Home

Cell Illustrator User Guide

image

Contents

1. fractional number as its quantity Double This type of entity is used to represent concentrations e g the number of ions or enzymes Other type of entity is a generic entity It holds string or logical value String or Boolean 3 3 Connectors Connectors are used to connect entities to processes input connectors or processes to entities output connectors There are three types of connectors Figure 2 process connectors inhibitory connectors and association connectors An input connector can take all three types In contrast an output connector is always of the process type Correspondingly all types of connectors can connect input entities to a process In contrast only a process connector can connect a process to an output entity name c2 c3 c4 firimg style 1 resh Ttdle nocheck threshold rule script 0 true E Process Inhibitory Association Connector Connector Connector Figure 2 The threshold is an input connector s parameter It can be a value or a script e g m 10 This parameter is used to define the minimum value of the input entities needed for the activation or deactivation of the linking processes A process and an association connector become activated 1f the threshold of the connector is smaller than the value of input entity On the other hand inhibitory connectors are activated when input entity is less than this threshold Inhibitory connectors are used t
2. 60 30 t 0 1000 2000 3000 0 1000 DFF40 oligomer DNA fragment 90 1200 90 o 900 600 30 300 t t t 0 0 2000 3000 0 1000 2000 3000 0 1000 2000 3000 90 1200 900 60 600 30 30 300 t t t 0 0 2000 3000 0 1000 2000 3000 1000 2000 3000 90 1200 900 60 600 an 30 300 t t t 0 0 2000 3000 0 1000 2000 3000 1000 2000 3000 Figure 5 Simulated time courses of some intermediates during apoptosis for the Fas ligand concentration n 210 450 and 600 References 1 Harada H 1999 Regulation of apoptosis by BH3 domain only proteins Jikkenigaku 17 1603 1606 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 82 of 88 2 Hugunin M Quintal L J Mankovich J A and Ghayur T 1996 Protease activity of in vitro transcribed and translated Caenorhabditis elegans cell death gene ced 3 product Journal of Biological Chemistry 271 3517 3522 3 Kuwana T Smith J J Muzio M Dixit V Newmeyer D D and Kornbluth S 1998 Apoptosis induction by caspase 8 1s amplified through the mitochondrial release of cytochrome c Journal of Biological Chemistry 273 16589 16594 4 Nijhawan D Honarpour N and Wang X 2000 Apoptosis in neural development and disease Annual Reviews of Neuroscience 23 73 87 O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 83 of 88 Appendix B
3. Using Differential Equations in CI In this example the following differential equations are modeled with Cell Illustrator ml t ml 0 mioat m2 t midi ml 0 10 m2 0 0 Chart 1 e g T 8 5 4 a 2 1 Figure 1 Differential equation results m1 m1 ma Iz cj P1 c2 zz 10 0 0 0 m1 10 0 0 Figure 2 Figure 1 shows the differential equation results For the model in Figure 2 the simulation results for runs with Sampling Interval equal to 0 1 pt 5 pt and 10 pt are shown in Figure 3 Figure 4 and Figure 5 respectively Comparing simulation results to those in Figure 1 you can easily notice that the smaller the sampling interval becomes the more similar accurate the simulation result is Note that Cell Illustrator offers modeling features that are alternative to modeling with differential equations and allow the simulation of complicated systems in an easier way 1 discrete behavior and continuous behavior can be mixed in one model in Cell Illustrator 11 the notion of delay can apply to discrete processes 111 activity and threshold concepts can be used in CI 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 84 of 88 Chart 1 ce z B8 P 8 5 4 3 2 1 Figure 3 Sampling interval is 0 1 pt amp Chart 1 x e Figure 4 Sampling interval is 5 pt Chart 1 a ce Figu
4. e g Microsoft Word or Excel 6 3 2 Simulation History Frame For your model you can run simulation many times changing the simulation parameters or making corrections to the model itself The Simulation History frame allows you to view and manage all the simulation log files CSV and CIL that have been saved for the given model file 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 65 of 88 If the simulation results are logged see Simulation Log each simulation run is saved into a separate CSV file along with the model representation at the time the simulation was done A log file with the extension csv 1s created in the folder project name output where project name is the name of the Cell Illustrator model file For example logs for the Cell Illustrator apoptosis csml sample project are kept in the apoptosis output subfolder All saved simulation logs are accessible in the Simulation History frame Creation Date Log Type Jan 26 2010 Coy Jan 26 2010 CIL Jan 26 2010 Coy wi View Settings istory Chart Settings F Simulation Settings Figure 79 This frame allows you for viewing and comparing of several simulation runs You can also manage the list of logs by opening the context menu with the following commands Open Log Open Directory Export CIL CSV File Open Backup CSML With the Rerun command you can open the selected log files for
5. 2 Import of CelIML or SBML models CelIML www cellml org and SBML www sbml org are widely used XML file formats for representing mathematical relationships in biological models In Cell Illustrator you can import models from CelIML SBML files into the CI workspace During the import operations the models are converted to CI models and can be edited analyzed simulated saved as any other models SBML CellML models differ from the CI ones E g SBML uses differential equations for representing relationships between entities while CI use the Petri Net model Therefore the conversion process has the following limitations o notall SBML models can be imported to CI o the simulation results might differ for a SBML model imported to CI It is recommended to run a simulation after a successful model import to verify whether the results of the converted model are identical to the simulation results of the original CellML model Before running the simulation the user should check and set the proper Simulation Time and Sampling Interval in the Simulation settings frame 8 3 Import of CSO and BioPAX models Cell System Ontology 3 0 CSO3 0 www csml org online services csml 30 ontology is an OWL format for storing biopathways information It 1s based on CSML3 0 which makes it a very complex and powerful format BioPAX level 2 www biopax org is most popular ontology for storing biological pathways data BioPAX Level 2 covers metabolic pa
6. Analyzing Simulation Results ssseessse 67 6 4 Running SECC OImillallOE sscuiodoceipen doe e te padieni e ege hu emend 68 t TOOLS FOR MODEL ANAL Y 99 2 conciones sco eben ca ieu cce i nrc 72 1 NaVIgaloE ueiieoio itio tenor venie Lo ied od on dace oit COEN 72 TPZ MANU N CAC I erecta cise T E UP 73 7 59 nt and PUN AGA onn a ve No PR FEVER andes enon ed 14 T5 EXDOPC ODIOM Serur ee ne ee 14 8 MAKING COMPLEX MODELS eere nere rne n enne 75 8 1 Grouping and UNGroOuping ccceccceeccceececeeeceeeeseeeseeeeeeseeeeeeeeeeeeeeeeseees 15 8 2 Import of CelIML or SBML models eeseeeeeeene 76 8 3 Import of CSO and BioPAX MOEIS ccc cceccesceceeeteneeeeeeseeeseeeeeeeeaeeeness 76 9 CELL ILLUSTRATOR PLAY ER ensi dace aa ER Ev beR WEE noia su su a EN MS xa ER VER QU 77 APPENDIX A TUTORIAL MODELING FAS LIGAND INDUCED APOPTOSIS 19 APPENDIX B USING DIFFERENTIAL EQUATIONS IN Cl 86 APPENDIX KINE TIC STYLE Sorron 88 CH REFERENCE S ripis a Eaa 90 2002 20 Human Genome Center Institute of Medical Page 3 of 88 Science The University of Tokyo All rights reserved 1 Introduction Cell IlIustratorTM CT is a software tool that enables biologists to model elucidate and simulate complex biological processes and systems 11 It allows researchers to model metabolic pathways signal transduction ca
7. Arc concentration change Defines which Connects a place to a transition A test entities are associated with each other arc does not consume any quantity of the but not as inputs or outputs Association source place by firing Connector allows the modeling of enzymatic reactions as well as other catalytic processes 3 2 Entities Entities are abstract elements that can represent any type of biological concepts like mRNAs DNA proteins ligands and compounds They can also represent cellular structures like mitochondria cell nuclei cells or biological phenomena like transcription and translation An entity Figure 1 contains a value which 1s interpreted as its quantity or concentration You can input a more meaningful name and value for the entity Also each entity has an associated variable el e2 e3 representing its quantity 30 123 2 ATGC respectively The mathematical equations used for simulation are expressed in terms of these variables O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 9 of 88 There are three types of entities discrete entity continuous entity and generic entity I name variable cAMP el enzymeA e2 mRNA e3 I i l quantity I 30 123 2 AUGC I i Discrete Continuous Generic i Entity Entity Entity Figure 1 Discrete entity holds an integer quantity Integer or Long In contrast a continuous entity has a continuous e g real
8. Concept Section 6 1 5 2 Modify a Model The user can move any element within the model 1 e change its location To carry out this task click on the Selection Mode button from the Draw toolbar Then click on the element in the canvas and drag it to move it You can also drag the mouse to select multiple elements and then right click the mouse to bring up a popup menu which allows to you to arrange copy cut duplicate or delete the selected elements A mouse right click on a selected single element will bring up a customized popup menu with additional options to modify the element The Draw and Element toolbars contain buttons for frequently performed editing operations This includes a e Insert Frame Create a background frame e Ld nsert Text Annotate the model e A Insert Image Load an image and insert the image to the canvas e amp Zoom In Zoom in on the canvas e amp Zoom Out Zoom out from the canvas e 4 Reset Zoom Restore the default models magnification O 2002 20 Human Genome Center Institute of Medical Page 37 of 88 Science The University of Tokyo All rights reserved sia Fit In Canvas Restore the default models magnification Uu Group Oroup together chosen elements to treat them as one for editing Ungroup Undo a group operation K A Set Color tools Fill Color Line Color or A Text Color e Set Stroke Change t
9. Nagasaki M Doi A Matsuno H Miyano S 2003 Recreating biopathway databases towards simulation In Computational Methods in Systems Biology Volume 2602 of Lecture Notes in Computer Science Springer Verlag pp 191 192 10 Nagasaki M Doi A Matsuno H Miyano S 2004 A versatile Petri net based architecture for modeling and simulation of complex biological processes Genome Informatics 15 1 180 197 11 Nagasaki M Doi A Matsuno H Miyano S 2004 Genomic Object Net I A platform for modeling and simulating biopathways Applied Bioinformatics 2 181 184 12 Nagasaki M Doi A Matsuno H Miyano S 2004 Integrating biopathway databases for large scale modeling and simulation In The Second Asia Pacific Bioinformatics Conference Volume 29 of Conferences in Research and Practice in Information Technology Australian Computer Society pp 43 52 13 Nagasaki M Doi A Matsuno H Miyano S 2005 Bioinformatics Technologies Chen Yi Ping Phoebe Ed Springer Press 179 243 ISBN 3540208739 Note Genomic Object Net is the research version of Cell Illustrator When Cell Illustrator is cited please cite the literature 11 O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 88 of 88
10. Simulation gives the user the ability to insert reference data contained in CSV file into the simulation run visually compare simulation results versus reference data Reference Sii Reference Simulatio Simulation Data Name Entity Mame el es m3 Set values When Presenting Data Figure 81 To setup a reference simulation Load the data file The file should be a tab separated values file with time series data stored in columns The first row must have the column headers names The first column must be the time column while the next columns should be identified by unique names Map the column names from data file to entity names of the simulated model Choose the method and moment of setting values from the list of available choices Only At Simulation Start When Presenting Data At Each Step Press the Run button If the Reference Simulation setup was correct the simulation will be started in the CI workspace Use the standard buttons Step Play Play Pause Fast Play etc to continue the simulation in the usual way the results of the simulation are displayed on charts and or saved to log file 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 68 of 88 6 4 5 Advanced SECG topics The SECG program is composed of two components one holds common functionality the second contains information specific for the simulated model The S
11. University of Tokyo All rights reserved Page 45 of 88 5 4 5 External References Frame Using this frame you may define for each model element a list of references to external databases or vocabularies The defined links can then be opened and viewed in a web browser 5 5 Biological Elements and Pathway Fragments While you can use the menu options and toolbar buttons described in Section 5 to add elements biological pathways can be created faster and more intuitively by dragging and dropping appropriate elements from the frames that support building of pathways These frames give the user the access to many predefined elements or pathway fragments and the ability to insert these fragments to the active canvas The Library toolbar frames gives a quick access to these useful 5 5 1 Biological Elements Frame The Biological Elements frame contains the Entity Process and Cell Component Sheet The Entity Sheet contains biological elements that will contain a certain quantity e g mRNA and protein Figure 59 O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 46 of 88 Biological Elements Element Type Element Size f Continuous C Discrete Generic sca fioo w Filter w alear Sort Order E Id 6 Alphabetical Xx Mawviga Biologi a Parts Li B Project E Search j BioP AC x Graph Figure 59 The Process Sheet contains
12. a linear way but one can change it into a quadratic function by writing a script The example in 4 5 could be modified as follows Step 6 Change the Kinetic Style of the process to Connector Rate and set the Rate parameter as 0 001 ml ml Step 7 Make sure that other parameters are the same as in the previous Tetramerization example 1 e cl Stoichiometry is 4 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 28 of 88 p53imonaomer m1 p53 tetramer m2 cz 10 0 0 001 m1 m1 i Figure 36 In the model in Figure 36 the speed of the tetramerization process is proportional to the square of p53 monomer The results of the simulation are presented below Figure 37 Chart 1 Mame ipb3tmona 50 T 100 125 p53 monaomerj p5S3 tetramer Figure 37 4 8 Separation In a cell binding proteins are sometimes separated with proteins A model can be created here with the following steps Stepl Create three entities and set their names e g p53 mdm2 p53 mdm2 and set their value e g 50 0 50 Step2 Create one process and set its name e g separate Step3 Connect the p53 mdm2 entity to the separate process Connect the separate process to the p53 entity and the mdm2 entity Step4 Set the separate speed by editing Kinetic Style and Kinetic Script options to values Custom and 1 0 Again 9 combinations of element types in t
13. a script to define Initial Value property of an entity Activity Speed Add Update and Delay of a process Threshold Updater and Firing Rule of a connector Double click on one of the above properties in the Element Settings frame open the Script Editor in which you can define the script and its language Cell Illustrator enables the usage of several different scripting languages such as simplemath java java bulk js javascript pnuts The Script Language combo box allows for choosing a specific language for each single script The default language for the whole model can be set in Simulation Settings frame This default language will be used whenever the script language is not set in the Script Editor 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 53 of 88 Script Language java 0 Unheded Check Ok Cancel lif irand gt 0 5 return true else return false x C Functions 9 Entities c IE z A ez Cube Figure 66 Figure 66 shows the Script Editor In the middle a script pad is placed where you can edit a script either by clicking on the function buttons and selecting entities from the table or by typing in text You can check whether the syntax of your script is correct by pressing the Check button If the check fails error messages will appear in a tooltip when you place the mouse over the Check Failure message 5 6 1 Exampl
14. and e2 and the discrete process pl between them The entity el 1s connected to pl with the input process connector cl The process pl 1s connected to e2 with the output connector c2 The kinetic style of the process is Custom The entities el and e2 have the initial value 10 and 0 respectively The input connector c1 has the Threshold parameter equal 2 The threshold of an input connector controls the activity of process pl The process pl has the delay parameter equal 1 and the add parameter equal 2 This model can be used for auto catalytic reactions e g protein auto phosphorylation O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 13 of 88 In the HFPNe theory time is measured in virtual time units called Petri net time pt At time 0 the value of el is 10 and the threshold of cl is 2 and the process pl is enabled When all input processes are enabled the connected discrete process is really executed after the delay time of the process In this example the delay parameter of the process pl is 1 Therefore the process is executed at time 1 Consequently at time the entity el decreases and e2 increases by the speed of process pl i e 2 Thus at time 1 the value of entity e1 and e2 becomes 8 and 2 respectively At time 4 the value of the entity el becomes 2 and the process pl cannot be enabled because the value of el is less or equal than the threshold of c1 At th
15. biological elements that denote biological processes e g phosphorylation and translocation see Figure 60 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 47 of 88 Biological Elements Filter Sort Order f Id Alphabetical XX Naviga a Biologi Figure 60 The Cell Component sheet contains useful pictures of cellular components e g cell and mitochondrion Figure 61 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 48 of 88 Biological Elements Entity Process Cell Component zm LP lement Type Element Size Fact Global Entity m 200 b Filter w lear Sort Order E Id 6 Alphabetical XX Naviga Biologi gl Parts Li B Project E Search BioPAC x Graph Figure 61 5 5 2 Project Manager Frame CI Project Manager 1s a user interface to pathway models stored on C7 Online Server To access the remote C7 Online Server the user must be logged in In this frame you can Browse and search through pathway libraries defined on the C7 Online Server and open selected models in CI workspace Organize your models and projects stored on Cell Illustrator Online Server Share your projects with other users and use projects shared by others See the CI Reference Manual for details 2002 20 Human Genome Center Institute of Medical S
16. col carry B nimi 0 mi4 200 release cyto lz Element Lists le Is e T Mouse position 6 852 Shaw Keyword search dialog Done 53 78 126 0 17 23 08 716M2 B39 i Figure 84 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 72 of 88 8 Making Complex Models This chapter presents features of CI that are especially useful for large scale modeling 8 1 Grouping and Ungrouping When editing a model it is often convenient to treat a group of elements in the canvas as one element This can be accomplished with the Group command This command is accessible in three ways via the popup menu in the canvas Element Group in the main menu and the Group icon in the Edit toolbar Once elements are grouped you can select and move them around easily just as one element The grouped elements can be ungrouped with the Ungroup operation Again there are three ways to ungroup elements via the popup menu in the canvas Element Ungroup in the main menu and the Ungroup icon in the Edit toolbar e m3 ay Cut Ctrl Copy CEri c Chr IJ Delete Delete ctrl E Copy Duplicate Ctrl M T Delete Delete T ungroup Ctrl Shift 6 Duplicate Ctrl M Arrange E Group Cri B set Image Edit Image Figure 85 2002 20 Human Genome Center Institute of Medical Page 73 of 88 Science The University of Tokyo All rights reserved 8
17. dimerization or tetramerization are not deterministic but stochastic To improve modeling of the biological processes one can use the Stochastic Mass kinetic style of the process Starting from the previous model the stochastic tetramerization process can be modeled as follows Step 6 Change the Kinetic Style of the process to Stochastic Mass Set the Standard Deviation to 0 2 This parameter controls the distribution of the randomly generated process speed 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 27 of 88 Step 7 Make sure that other parameters are the same as in the previous Tetramerization example 1 e c Stoichiometry is 4 p53 imonomer mi ci pi c2 p53 tetramer M eset 10 0 0 max 0 001 m1 0 2 gauss0 0 0 0 Figure 34 In the model shown in Figure 34 the speed of the tetramerization process represents stochastic behavior Parameters related to Stochastic Mass kinetic style are described in section 3 4 Processes The example results of the simulation are presented below Figure 35 Note that subsequent simulation runs will give different results due to the stochastic nature of the model Chart 1 Eram a 5 10 15 20 25 a0 35 40 p53 monomer p53 tetramer Figure 35 4 7 Custom Tetramerization Tetramerization can also be defined by a custom formula In the example in Section 4 5 the process speed depends on the p53 monomer in
18. large models SECG will run large models much faster than the standard engine Customization integration of simulation model The command Export Simulation Source Code enables you to generate source code that can be customized by a programmer and or reused in another software 6 2 3 Simulation Log The results of each simulation can be logged and saved into a CSV Comma Separated Values file or CI log files CILs The CIL files can be viewed and analyzed in the Cell Illustrator Player program The CSV file can be analyzed with an external spreadsheet application A CI log is an XML file that consists of two parts the input model and a time series of values for the logged properties The time series data may include entity values and or process connector speed process state firing waiting not firing To track value changes for all elements you need to Set the Save Log File option to the on value or select the Save Log button on the Simulation toolbar Define the elements to be logged in Logged Elements combo box Specify the Log Update Interval which defines how often entity values are saved in the simulation log 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 62 of 88 Logged Elements combo box enables you to define the elements to be logged and offers the following choices All Entities Log all entity values but do not log
19. of simulations for the same model and the simulation results may differ from one simulation run to another for example For a model with random factors or for a model with changed parameters 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 59 of 88 6 2 How to Set Up and Run Simulation 6 2 1 Simulation Settings The next step after creating a model is to set up simulation process parameters The Simulation Settings frame has the following options that control the simulation flow Sampling Interval Simulation Time Chart Canvas Update Interval Simulation Speed Continuous and Discrete Weak Firing Firing Accuracy and other Execution and Log File parameters Figure 76 Simulation Time Sampling Interval Chart Update Interval Canvas Update Interval Continous Weak Firing Discrete Weak Firing Firing 4ccuracy Use SEC SECG Calculation Method Update After Step Randomize Processes Default Script Language simplemath Log Save Log File Log Update Interval Logged Elements Speed Set Speed of Simulation Figure 76 During simulation the model changes its state in a certain time sequence In Cell Illustrator simulation time 1s measured in virtual time units called Petri net time pt The Simulation Time option sets the duration of the simulation process in Petri net time If the Simulation Time is set 1000 as in Figure 76 the simulation lasts from O pt
20. other elements connectors and processes All without Firing Log entity values and process connector speed for all elements but do not log process state All with Firing Log everything entity values process connector speed and process state for all elements Chart Log these elements which are displayed on charts see Chart Settings frame By default no one element is added to he chart In such a case nothing will be logged during simulation and a CIL file will not be created during the simulation Log Log these elements which have the log option set to true This option is displayed in Element Lists frame By default the Log option of all entities processes and connectors is off In such a case nothing will be logged during simulation and a CIL file will not be created during the simulation Chart and Log Log all these elements which are displayed on chart or have the Log option set A CIL file is created if at least one entity or process is selected for logging During the simulation the data 1s logged at time points defined by the Log Update Interval setting in the Simulation Settings The log interval value needs to be selected carefully so that an appropriate number of steps is logged during the simulation In general the number of steps that will be logged is given by the following formula SIMULATION TIME N STEPS 7 LOG INTERVAL It is recommended to select the log interval in such a wa
21. replay in the Cell Illustrator Player application 6 4 Running SECG Simulation SECG Simulation Engine Code Generator is a new mode of running simulations in Cell Illustrator SECG is based on the same HFPNe model as the standard simulation engine described above however it 1s faster more customizable and adds some viewing and running modes The basic concept in SECG consists in generating a computer program which execution will perform the simulation Also the SECG module allows exporting the source code of this program this allows running the simulation from other program and enables controlling the simulation behavior and data in a more precise way The simulation can be run using SECG by selecting Use SECG option in Simulation Settings frame The SECG simulation is performed in the Cell Illustrator workspace in the standard interactive way using the Step Play Play Pause Fast Play etc buttons the results of the simulation are displayed on charts and or saved to log file Note The javac compiler that is included in the JDK is required to run SECG simulations The JDK Java Development Kit must be installed on your computer JRE is not enough to run SECG 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 66 of 88 6 4 Passing Parameter simulation The SECG module offers a method for performing a series of simulations in a bulk on the same model with varying th
22. substring U current position 5 else 7 mli return ml 4 n C 0 8 lt c Q Functions J CC 7 8 9 f 2 EE a transcripted protein 4 5 6 lt Bk 4s 1 2 3 i 0 E gt y Figure 69 Selected succeeding states of the model in Figure 68 are presented in Figure 70 transcription transcripted pratein sequence mi mapgmyethlpls transcription transcripted protein sequence m mapaqrmyetfhilpnispeellksggvngyeqevi 2002 20 Human Genome Center Institute of Medical Page 55 of 88 Science The University of Tokyo All rights reserved Figure 70 5 7 Preferences Frame Width Height Fill Color LES Line Color El o 0 0 255 Stroke L 31 0 Biological Property FF Process Fact Proxy Settings URL Templates Restore Close Figure 71 With the Preferences frame you can set the global properties of Cell Illustrator This section briefly introduces the major options available in the frame The frame displays a tree of option types and and the list of available options and their settings for a selected part of the tree For example Figure 71 shows the global settings of the figure to be used for display of continuous entities One can define the size of the figure its color and other parameters The values from the Entity Process Connector Fact Fact Edge sections are used as the defaults when a new element is created The values of Simulation Settings secti
23. the simulation are presented below Figure 41 Chart 1 LV M 30 40 50 amp Q fo B p53 mdm2 p53 mdm2 Figure 41 For models based on discrete elements it is difficult to create a model that fulfills this condition 4 9 Inhibition A specific drug sometimes inhibits activities of transcription The transcription model in Section 4 3 is modified with an inhibitor activity as follows Stepl Create an entity and set its name e g doxorubicine Step2 Create a process and connect it to the doxorubicine entity with the process connector Step3 Connect the doxorubicine entity to the transcription process with the inhibitory connector Step4 Do not change the threshold of the inhibitory connector the default value is 0 The type of the doxorubicine entity can be either discrete or continuous With the model in Figure 42 one can simulate the situation where the concentration of doxorubicine is growing If the inhibition connector threshold is set to 0 for continuous processes the translation will be stopped immediately after the first sampling Interval For a discrete process Translation will be stopped after the first step as during the first step process pl is enabled 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 31 of 88 p2 doxorubicine m2 p2 doxorubicine ma 1 0 0 0 1 O10 MRNA pos m3 pi mRNA pos m
24. with the increase of the Fas ligand concentration 200 lt x 1200 Z E 1000 z z 800 Autocatalytic dix o presen E 600 ia in m apsen E 400 D O O QO O 200 400 600 800 Concentration of Fas ligand 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 80 of 88 Figure 3 Simulated relationship between the amount of DNA fragmentation and the Fas ligand concentration At higher concentration of Fas ligand the direct pathway from caspase 8 to caspase 3 contributes to the fragmentation To examine the effect of autocatalytic process of caspases DNA fragmentation is simulated for both cases of the presence and absence in this process There are two pathways from activated caspase 8 to caspase 3 one through several steps including the cytochrome c release from mitochondria when the concentration of activated caspase 8 1s low and the direct one to caspase 3 when the concentration of activated caspase 8 is high 3 We assume arbitrarily that the direct pathway starts when the concentration of activated caspase 8 1s larger than 30 Reportedly the removal of the Bid by gene knockout method increases the resistance of liver cell apoptosis by Fas ligand while it does not affect the apoptosis of thymus and embryonic cells If the second pathway 1s included in the scheme DNA fragmentation increases slightly especially when the Fas ligand concentrat
25. Arid Mass D 1 m3 0 1 1 0 1 0 1 0 4 Element Settings wr View Settings Chart Settings F Simulation Settings Figure 56 The major process parameters displayed in the Process Sheet are Type Calc Style Firing Style Activity and Kinetic Style The Calc Style property determines its Type which defines whether the process is discrete continuous or generic and One can choose between two Firing Style settings and and or For a process with the default and firing style 1f all input process connectors and association connectors of the process are activated and no input inhibitory connector of the process is activated the process is enabled For a process with the or firing style the process 1s enabled if at least one of Association or Inhibitory input connector fulfills its condition and all input process connectors fulfill their respective conditions The Kinetic Style defines the method for calculating the values of connected entities The property can have the following options Custom Connector Custom Mass Stochastic Mass Connector Rate Michaelis Menten Hill Function Each style requires a specific set of parameters that you need to enter at the bottom part of the Element Settings frame For example you must specify the Coefficient Coefficient2 Standard Deviation and Connector Stoichiometry for the Stochastic Mass style The most general style is Custom If a process has the Custo
26. Der ot Input OnLItr es stoichiometry where m product is the product of all input entity values For the input entity variables mi m Mn In prouet m gm Ta Page 86 of 88 stochastic mass parameters coefficient coefficient standard deviation and stoichiometry for each connector connector rate parameters rate and stoichiometry for each connector 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved continuous speed calc style discrete add calc style continuous speed calc style discrete add calc style delta is randomly generated using the Gaussian distribution with the calculated delta mean and specified standard deviation delta mean is calculated using the formula for the mass kinetic style Celta mean deltla mass COSOSTIricrentgo OcoerrtroTrTernto stoichiometry delta is randomly generated using the Gaussian distribution with the calculated delta mean and specified standard deviation delta mean is calculated using the formula for the mass kinetic style delta Mean decus mass COGTITICLOeNUS COSTIUC Lents stoichiometry delta rate Storenvometry sampling interval delta rate stoichiometry Page 87 of 88 CI References 1 Doi A Fujita S Matsuno H Nagasaki M Miyano S 2004 Constructing biological pathway models with hybrid functional Petri nets In Silico Biolo
27. ECG allows the change and extension of the common module see the EngineBase options in Simulation Setup SECG menu This enables the usage additional functions scripts or performing operations which would normally would be impossible in the Cell Illustrator for example saving and using previous values of the entities The source code of the engine generated by SECG can be exported to file in Java or other languages The source code file can then be edited which allows for better control over the simulation Also the simulation source code can be integrated or executed in another program which is useful for performing parameter optimization or generating some advanced statistics from the engine run The exact reference for advanced usage which contains description how the SECG framework works and detailed information of the content of each source code file is obtainable from Masao Nagasaki masao ims u tokyo ac jp or check http www csml org Note Profiting from the SECG engine possibilities to the full extent requires some software development skills O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 69 of 88 7 Tools for Model Analysis 7 1 Navigator For a large model the Navigator frame Figure 82 provides a convenient way to locate and display in the canvas window a desired part of the model The Navigator displays a small view of the entire model with a green rec
28. I workspace and to store and organize them on the remote CI Online Server Additionally CI Online gives the user convenient access to libraries and databases of bio pathway models In CI Online Cell Illustrator 1s the front end where the user prepares a pathway and simulates it The Cell Illustrator client communicates with the CI Online Server over the internet HTTP or HTTPS protocol The CI Online Server is a place where the user logs in to store and manage his pathways and to share them with other users 2 3 Licensing Information To run Cell Illustrator you need to obtain a CI Online account For details please contact your local Cell Illustrator reseller CI offers several license types Professional Standard Classroom Access and Player The main differences between the license types are in the simulation capabilities of CI and are briefly summarized at https www cellillustrator com lineup Note This manual describes all functions available in the CI CI Professional license type A given function described in this manual might be disabled or not available in the limited versions CI Access and CI Player 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 6 of 88 2 4 CI Memory Customization When starting CI Online you can set the maximum memory for Cell Illustrator 1n the Login dialog box O 2002 20 Human Genome Center Institute of Medical Science The Uni
29. JREs are available for download at http java sun com downloads To run Cell Illustrator Online Open the Cell Illustrator Online website at https cio bioillustrator com or https cionline hgc jp in your web browser is f f E 5 a 5 3 N GEN Ww Geli Niustireator 9 m UL CU ER D0 Main menu Gell IlIustrator Player REGISthation Cell illustrator Java Web Start is required to run CIO Download the latest Java Runtime Enviroment JRE from http java com and install it on your PC Get your login account through the registration and try CIO If you are already registered you can check your license status Register to obtain the CI Online account Run Cell Illustrator 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 5 of 88 https cianline hgc jpicifileserver Password one Remember Password Memewemey ji id Inthe C7 Startup frame enter the login data that you obtained during the registration and press Sign In Note It is recommended to select the Remember Password option After a successful login the login and password will be stored on your local computer and you will not need to enter the login data again After this CI Online is downloaded from the server This may take several minutes CI Online is a client server solution that enables the user to create manipulate and simulate bio pathway models locally within C
30. OOH terminal part translocates to mitochondria where it triggers cytochrome c release The released cytochrome c binds to apoplectic protease activating factor 1 Apaf 1 together with dATP and procaspase 9 and activates caspase 9 The caspase 9 cleaves procaspase 3 and activates caspase 3 The other pathway occurs when caspase 8 cleaves procaspase3 directly and activates it The caspase 3 cleaves DNA fragmentation factor DFF 45 in a heterodimeric factor of DFF40 and DFF45 Cleaved DFF45 dissociates from DFF40 inducing oligomerization of DFF40 that has DNase activity The active DFF40 oligomer causes the internucleosomal DNA fragmentation which is an apoptotic hallmark indicative of chromatin condensation Note this is not a CI model The pathways consist of several steps lt where two different pathways from caspase 8 are assumed gt and many molecules including Fas receptors the caspase family which includes aspartic acid dependent cysteine proteases and products of their zymogens the Bcl 2 family which includes pro and anti apoptotic proteins cytochrome c and DNA fragmentation factor The apoptosis starts from the Fas ligand binding to Fas receptors and ends in the fragmentation of genomic DNA which is used as a hallmark of apoptosis Thus the amount of DNA fragmentation can be assumed to be proportional to the cell death We have designed a model through utilization of the known facts about the Fas induced apoptosis pathways shown in Fi
31. Tokyo All rights reserved P age 34 of 88 Chart 1 3 4 CAK p53 p53 P Figure 49 In the presented reaction CAK influence on the model is enabling disabling the reaction The CAK activity will be adjusted with the threshold of the association connector However in most of the systems the reaction rate depends on enzyme concentration and substrate concentration In the example in Figure 50 the speed of the catalysis process is defined with the equation m1 m2 10 The model is made of continuous type elements and the threshold of association connector is reset to 0 p2 enzyme m3 1 0 0 0 7 i i O10 i substrate m1 pi product mz m1 m3 0 00 0 0 Figure 50 The results of the simulation are presented below Figure 51 Chart 1 4 5 amp T CAK p53 p53 P Figure 51 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 35 of 88 Cell Illustrator User Interface and Model Creation At Cell Illustrator startup a single main window pops up You can create open save and print a biological pathway model file using options in the File menu Figure 52 This window has the Menu Bar at the top that allows you to manipulate the model diagram Cell Illustrator 5 0 Professional demo intra csml org l B x File Edit Element Simulation View Analyze Window Help ga ejixooecmmellis e amp sa
32. ass z mOoOO00 788612 pe3im s ub cso30 c Protein TRANSPATH Kass MO000069108 TPS3SIMP1 v s cso30 c Protein TRAMSPATH kass WOOO EF Paih Tath cie Protein TRANSPATH Hass mongnaadd35440 BAF47rihy pe3rh cso30 c Protein TRAMSPATH kass Mogg 7738 PSD a5rihyIRSpBe36 cso30 c Protein TRANSPATH Kass PTOOO0018953 pe31ubin t csa3 c Probein TRANSPATH Kass MOOU00S85 36 Delta133p53 cso30ic Protein TRANSPATH Kass MOo00039149 PML 3 p53 p515Hp cso30 c Protein TRANSPATH Kass Fo000045160 pe34pt cso30 c Protein TRANSPATH Kass MOO00083781 pS3tm sePhbirins ceso3 ciProbein TRANSPATH Klass gt Found 466 Molecules Far pos a ee v Merge Auto Layout Create Mew Canvas Import Biologi a Parts Li B Project i Search E BiaPAC x Graph Figure 63 zi CE ac E 5 5 4 Parts Library Frame The Parts Library frame can be opened with the button from the Library toolbar In this frame you can register a model file as a reusable item Once registered the model is accessible in the tree and the elements in the file can be inserted into the active canvas in the same way the command File Import Model works With Parts Library you can organize a repository of frequently reused models Cell Illustrator comes with several pre registered parts to demonstrate this functionality 2002 20 Human Genome Center Institute of Medical Science The Uni
33. at time the process pl cannot be enabled and el and e2 reach a stable state The results of the simulation are presented below Figure 5 Figure 5 3 7 Discrete Elements Example 2 Connector Speed In the previous model the values of el and e2 increase and decrease by the same amount 2 at each time interval because their value depends on the add value of the process pl However modeling differing increase and decrease in the value of connected entities is sometimes necessary e g when a monomer becomes a dimer or a trimer and vice versa To deal with this type of process set the process kinetic style to connector custom and then define the value changes of the connected entities separately m2 iD T iT r3 c c2 Figure 6 O 2002 20 Human Genome Center Institute of Medical Page 14 of 88 Science The University of Tokyo All rights reserved The example in Figure 6 demonstrates the use of the connector custom kinetic style The difference between the previous example and the model in Figure 6 is as follows 1 The add parameters of connector c1 and c2 are 2 and 1 respectively At time 0 the value of el is 10 and the threshold of cl is 1 and the process pl is enabled After the delay of pl i e 1 the process is executed The add value of cl and c2 are 2 and 1 respectively Thus the value of el and e2 becomes 8 and at time 1 respectively At time 4 the value of el and e2 are 2 and 4 resp
34. ations for quantities change are dmi dA dt dt At time 4 the value of el is the same as the threshold of cl 1 e 2 At that time the process pl cannot be enabled and el and e2 reach a stable state The results of the simulation are presented below Figure 9 10 0 7 5 Figure 9 The same conversion process can be applied to the discrete model in Section 3 7 Figure 10 Figure 10 The equations for quantity changes are dml dm2 _ 22 dt dt The results of the simulation are presented below Figure 11 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 16 of 88 Figure 11 3 9 Continuous Elements Example 2 In the previous two examples discrete examples were converted into similar models with continuous elements For kinetic based modeling one should use a continuous model rather than a discrete one In the following model Figure 12 the consumption ratio of el is proportional to the value of el 1 e the speed of process pl follows the mass action kinetics el ml pi e2 me O a c2 O 10 0 E m1 0 05 0 0 Figure 12 The difference between the present model and the model in Section 3 8 is that the speed of pl is changed from 1 to 0 05 ml In this model process pl continuously transports entity el to e2 at the rate of 1 20 of el s concentration ml Since pl 1s connected to el with quantity represented by m1 and its output
35. axml 9 New Model 1 csml 7 xx Ui DOS x Biological Elements Bi i D 1 m1 Figure 52 To create a new model the first step 1s to open an empty canvas window for the model which can be accomplished by either 1 clicking on the top icon in the left toolbar 2 clicking on the File in the Menu Bar and selecting the New option 5 1 Add Elements To add a new element an entity or a process select one of L e LC Ell ind EH icons from the top toolbar and then click on the canvas workspace where you want to place it You may continue and place more components of the selected type on the canvas Alternatively you can press the right mouse button anywhere on the canvas and choose the desired element from the popup menu using Insert Entity Insert Process function The new element will be placed at the center of the active canvas Lastly you can select a predefined element in the Biological Elements frame see Section 5 5 1 Each entity has three labels associated with it for you to edit by double clicking see Figure 1 On the upper left is the Name of the entity for associating biological concepts with the entity The upper right label is the Variable representing this entity It is used in mathematical equations to define the rate of reaction in the model The lower left label represents the current value of the entity You can edit it O 2002 20 Human Genome Center Institute of Medical Scie
36. be edited have a white background while the columns that cannot have a gray background To change an element property displayed in a gray out column you need to select a corresponding element in the canvas and open the Element Settings frame The Element Settings frame displays all the properties of a single element selected in the canvas The frame may display depending on the type of selected element a list of various properties divided in categories Simulation View Shape Image Biological Properties Custom Biological Properties etc Any modifications made in the Element Settings will be immediately reflected 1n the Element Lists and the canvas and vice versa O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 39 of 88 5 4 1 Entities For entities several properties are available for editing in both frames see Figure 53 and Figure 54 A default setting is automatically displayed if you have not entered any value You can change the name of the entity to reflect its biological meaning and you can also enter the initial value of the entity The current value of the entity is displayed once simulation data is available To show hide the selected element in a simulation output file check the Visible box Clicking on the column header you can sort or filter the table You can also control if a given column property should be visible or hidden in the table by a right click o
37. cience The University of Tokyo All rights reserved P age 49 of 88 Project File Sort circadian Previous Next 73 Y Jj CT File Server Mv Projects Shared Projects Libraries Transpath DB Academic COXPresDB E csml org et Gene regulatory networks m Switching mechanism of lambda phage z Circadian rhythms in Drosophila l 4 Circadian rhythms in Mus musculus H Metabolic pathways H Signal transduction pathways csml org zipped CIOS 0 only Transpath DB Professional Figure 62 To open Project Manager Frame choose icon from the Library toolbar 5 5 3 CSMLDB Search Frame The CSMLDB Search Frame is an interface that allows for submitting queries to the remote CSMLDB database To access the CSMLDB database the user must be logged in to the C7 Online server Using this frame you can search in the CSMLDB database for molecules entities and or reactions processes and import them into active canvas To open CSMLDB Search Frame select icon from the Library toolbar 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 50 of 88 rch CSMLDB Entity Process Fact Search For pss Find Databases Max Rows suc Academic x Search In W Name w Id Use Synonyms Case Sensitive Organism Type f ID Mame Type Properties M moo O023312 pe3ihiMdmzrh cso30 c Protein TRAMSPATH k
38. d from the canvas context menu A newly created chart is added to the tree To add an entity process connector to an existing chart select the element on the canvas highlight the chart name in the tree right click on it and select the Add Selection from Canvas item You can also delete charts Highlight names of charts to be deleted in the Chart Tree and choose Remove Selected from the context menu Chart Settings Charts B Chart 1 i v Bl proteintinactive v Bl protein active 33 auto phosphorylation a Chart 2 wl Bl protein active m P we View Settings P Simulation Settings Figure 77 6 2 5 Run Simulation Once the setup described in the previous section is completed you can start and control the simulation using the VCR like controls in the Simulation toolbar The toolbar allows you to start gt and pause 1 l the simulation Also you can step through the simulation with the Step Play 1 gt button The buttons and invoke simulation with animated tokens moving between entities The simulation can be stopped at any time with the Stop and Initialize M button These simulation commands are also available in the Simulation menu in the Menu Bar During the simulation active components in the model are highlighted in red You can also see the values of the entities updated during the run Charts defined in the Chart Settings frame are displayed and updated with the frequency specified by the Plot Update I
39. del of degradation can be created with the following steps Stepl Create an entity and change its name e g p53 and set its value e g 10 Step2 Create a process and set its name e g degradation Step3 Connect the p53 entity to the degradation process with the process connector Step4 Set the degradation speed by editing the Kinetic Script parameter of the degradation process For the discrete process the default speed is 1 0 For the continuous process set the Kinetic Style to Custom and then edit the Kinetic Script speed In Stepl and Step2 the following combinations are possible Figure 16 a discrete entity and a discrete process a continuous entity and a continuous process a continuous entity and discrete process g m1 Bp m3 10 10 E E threghald threghald pl p3 stand and 1 1 1 Figure 16 The results of the simulation are presented below Figure 17 O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 19 of 88 Figure 17 If the degradation speed depends on the value of p53 protein say m in Step 4 the speed should be made dependent on m Figure 18 eS m ef mb 100 100 threshold threshold degradation degradation EE im 1 mito m6 0 Figure 18 The results of the simulation are presented below Figure 19 Figure 19 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserv
40. discrete discrete continuous and continuous continuous Figure 24 Cell Illustrator facilitates the modeling of the path DNA gt MRNA Protein not only as a concentration function of DNA mRNA and Protein Global transcription and translation mechanism can be established with a string input DNA sequence intermediate product proper mRNA and product Protein This mechanism can produce simultaneously many proteins from different DNA sequences transcription cq h NA p53 m1 transcription c3 mEMA p53 m3 1 40 0 10 m transcription co mRNA p53 m2 1 10 Bie Figure 24 The simulation results of the discrete model the are presented below Figure 25 Chart 1 p53 cytoplasm Figure 25 Normally mRNA synthesis is associated with its degradation The models presented below Figure 26 combine the transcription process with the degradation process in Section 4 1 The same speed m 10 is applied 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 23 of 88 transcription mRNA posimi m1 c transcription DRNA ph3im3 m3 1 1 0 0 1 0 0 0 g 0 0 gj p4 p 1 m1 1 ma g transcription c3 MRNA_pss m2 m2 1 1 0 maig Figure 26 The results of the simulation are presented below Figure 27 Chart 1 mFEH amp ps53 m1 mPFNA_poSsim2 mFEFHh amp p53m 3 Figure 27 4 4 Complex In a c
41. e 1 Stochastic Behavior of Translocation Process Many processes 1n a cell show stochastic behavior In order to model a stochastic translocation process the model in Section 4 2 was modified by applying a script in Figure 66 to the Activity option of process pl The results of the simulation of the discrete model with the sampling interval equal to 1 are presented in Figure 67 The figures compares 2 discrete processes one randomly activated and the second always active process Note that the value changes of the p53 nuclei random entity are not steady Chart 1 p53 nuclei pS3_nucleifrandom p53 cytoplasm p53 cytoplasm random Figure 67 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 54 of 88 5 6 2 Example 2 Transcription Process with Generic Entities and Processes Generic entities are of two value types For example you can use an entity that holds a String variable to describe a translation transcription process with a generic process transcription transcripted protein sequence m1 Figure 68 In this example a script shown in Figure 69 1s assigned to the Updater property of the connector Script Language java CHEESE Check Ok Cancel l String sequence mapquyethlplspeellkscqewndgywuqewvl int size sequence lengthi 3int current position ml lengthi 1 iif current position lt size 5 ml sequence
42. e selected parameter value s This mode allows comparing the simulation results when initial parameters differ It allows checking for example how the amount of control entity influences the system or choosing the right value of some kinetic constant The changing parameters are selected and the range of their values is defined in the SECG Passing Parameter dialog box After that for each combination of parameter values a simulation is run This mode is faster than running each simulation as a different SECG simulation and it is run outside the Cell Illustrator Chart Simulation 17 1 Parameterized Simulation Mame ez l ez n 5 egz 1 e2 1 5 e e7 0 er 0 5 er 1 e7 1 5 d no ES ISIS ISIS Entity ez Oo oOo 82 0 e2 0 5 e2 1 e2 1 e2 2 er efr 0 5 e7 1 e 1 5 ef 2 Figure 80 The results of Passing Parameter simulation are presented as 2D or 3D charts After running Passing Parameter simulation a control panel window appears which allows from comparing simulation results by selecting the chart type and parameters values The charts always present how the entities depend on one selected entity while the rest of the parameter values is fixed 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 67 of 88 6 4 0 Reference Simulation This simulation mode can be run by invoking Simulation Run Reference Simulation Reference
43. ectively At that time the process pl is not enabled and el and e2 reach a stable state The results of the simulation are presented below Figure 7 x Figure 7 3 8 Continuous Elements Example 1 In the previous two examples discrete entities and discrete processes have been used However in many cases entity values should be treated as continuous variables rather than discrete variables The model in Figure 8 has been created by conversion of the discrete model in Section 3 6 Figure 8 More specifically the differences between the models in Figure 4 and in Figure 8 are as follows Step 1 The type of entities el and e2 is changed from discrete to continuous Step 2 The type of process pl is changed from discrete to continuous Step 2 The delay of process pl is changed from 1 to 0 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 15 of 88 All other parameters are the same as in the discrete model in Section 3 6 As described in Section 3 6 the threshold parameter of the input connector works like the threshold of the activity of process pl In this example at time 0 the value of el is 10 and the threshold of c1 1s 2 and the process is enabled When all input connectors are enabled the connected continuous process 1s executed with no delay Since pl is connected to el with the quantity m1 and its output is connected to e2 with the quantity m2 the equ
44. ed P age 20 of 88 4 2 Translocation In a cell mRNA and proteins move from one compartment to others e g from nuclei to cytoplasm and vice versa A model for this can be created with the following steps Stepl Create two entities and change their names e g p53 nuclei and p53 cytoplasm and set their values e g 100 and 0 Step2 Create a process and set its name e g translocation Connect the p53 nuclei to the process and connect the process to p53 cytoplasm Step3 Set the translocation speed by editing the Kinetic Script parameter of the translocation process For the discrete process the default speed is 1 0 For the continuous process set the Kinetic Style to Custom and then edit the Kinetic Script speed In Stepl and Step2 possible combinations of p53 nuclei translocation p53 cytoplasm triple are discrete discrete discrete discrete discrete continuous continuous discrete continuous continuous discrete discrete and continuous continuous continuous Figure 20 p53 nuclei m1 c1 translocatian 9 p33 cytoplasm m2 100 i 1 11 i p53 nuclei _ m3 c5 translocation cg P93 cytoplasm m4 100 i 1 10 0 0 Baa Tule m c7 translocation eg D93 cytoplasm _ m6 100 0 zu 1 10 0 0 p53 nuclei m eg translocation cig Pas_eytoplasm ms 100 0 ui 10 0 p53 nuclei m8 E translocation eq Pes_cytoplasm m10 k T 100 0 1 0 0 0 Figure 20 The simulation results of the discrete model are pr
45. ed by editing the speed option of the catalysis process Set Kinetic Style to Custom and Kinetic Script speed to 1 0 Step4 Create one entity and change its name e g CAK Step5 Create a process and connect it to the CAK entity with the process connector Set kinetic style to Custom and kinetic script speed to 1 0 Step6 Connect the CAK entity to the catalysis process with the association connector Step7 Change the threshold of the association connector to 1 the default value is 0 The type of CAK entity 1s continuous With the model in Figure 46 one can simulate the situation where CAK enzyme concentration is growing If the association connector threshold is set to 1 the catalysis process will be enabled when the CAK entity value reaches 1 level 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 33 of 88 p2 CARK ms pos m1 catalysis p53 P m2 100 0 0 1 0 00 Figure 46 The results of the simulation are presented below Figure 47 Chart 1 CAF p53 p53 P Figure 47 If the threshold of the association connector is modified to 5 Figure 48 the results will be different The results of the simulation are presented below Figure 49 p2 CAK mb 1 0 0 0 i i i 50 i pas m4 catalysis i poa P _ m5 0 0 10 0 0 10 0 Figure 48 2002 20 Human Genome Center Institute of Medical Science The University of
46. ell a set of proteins binds and creates a complex A model for this can be created with the following steps Stepl Create three entities and set their names e g p53 mdm2 p53 mdm2 and set their values e g 100 50 0 Step2 Create one process and changes its name e g complex Step3 Connect p53 entity and mdm2 entity to the complex process Connect the combine process to the p53 mdm2 entity Step4 Set the combine speed by editing Kinetic Style and Kinetic Script options to values Custom and 1 0 In Stepl and Step2 9 combinations of element types in the pattern p53 mdm2 p53 mdm2 complex process are allowable 1 e discrete or continuous discrete or continuous discrete or continuous discrete and continuous continuous continuous continuous Figure 28 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 24 of 88 paa m1 pi c3 p53 mdm2 m3 pha mig p epg P3 mdmz m18 100 100 NN 1 0 0 0 cli 0 0 mama madm2a mis s0 s0 p 53 m cd p2 cB p53 mdm2 m6 ps3 mi c19 p co p53 mdm2 mei 100 100 iE 1 0 0 0 c20 0 0 madm2 mi mdm2 ma 50 0 50 0 p53 m t7 p3 c9 p53 mdm2 m p53 m22 c79 pe cod p53 mdm2 m4 100 0 100 0 aoe 1 0 0 0 c23 0 0 mdm2 ms mdm2 maa 50 50 p53 m10 c10 ni4 eq P923 mdmz m12 p53 m25 c25 pa c27 p53 mdm2 m27 100 0 100 0 E S 1 0 0 0 c26 0 0 mdm2 mi4 mdm2 _ m26 50 0 50 0 p53 m13 c13
47. eric processes The activity of continuous processes is different The firing of a continuous process occurs continuously Consequently a continuous process does not hold the Delay parameter O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 11 of 88 To employ generic processes one must use the Pnuts scripting language Use of Pnuts is briefly described in Section 5 6 Another important parameter associated with a process 1s its Kinetic Style If the process kinetic style is Custom then the value change after the process fires 1s the same for all connected input output entities For all other kinetic styles the value change at each entity depends on connector specific parameters such as Connector Stoichiometry and Connector Speed Thus it may be different for different input output entities 3 5 Element Combinations All combinations of discrete and continuous CI elements are presented in Table 2 and Table 3 Table 2 Output Connectors L v t L v t Transport quantity Transport quantity Sequentially v t through the OH OWER v t through the transport quantity Connector at d t Connector at each v t through the intervals d t interval Connector 56660006 Not Allowed The variable m t represents quantity or concentration of the entity at time f The variable w t is the threshold of the corresponding connector The constant d t 1s the delay time of the corre
48. esented below Figure 21 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 21 of 88 xl a0 Figure 21 If the translocation speed depends on the quantity of p53 nuclei say m1 as in the degradation example Section 4 1 the process speed should be made dependent on m1 10 in Step3 The factor of 1 10 was used in the example in Figure 22 p53 nuclei m1 ef translocation 9 h93 cytoplasm m2 Q 100 0 0 0 m1 10 0 0 Figure 22 The results of the simulation are presented below Figure 23 x 30 BO FO p53 nuclei p53 cytoplasm Figure 23 4 3 Transcription Transcription modeling can easily be done using CI A model can be created with the following steps Step1 Create a process and set its name e g transcription Step2 Create an entity and change its name e g mRNA p53 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 22 of 88 Step3 Connect the transcription process to mRNA p53 entity with the process connector Step4 Set the transcription speed by editing the Kinetic Script option of the transcription process For the discrete process the default speed is 1 0 For the continuous process set the Kinetic Style to Custom and then edit the Kinetic Script speed In Stepl and Step2 possible combinations of the pair transcription mRNA p53 are discrete
49. f Cell Illustrator is based on the Hybrid Petri net with extensions HFPNe theory 10 and uses the concepts of entities connectors and processes With discrete elements you can execute basic logic based simulations You can also execute differential equation based simulations with continuous elements and combine discrete and continuous elements 1n more complex models The most advanced simulations would employ generic elements and scripting For a short introduction to the HFPNe theory see Section 3 It is easy to prepare a basic simulation just by drawing biological pathways on the canvas A simple model can be then turned gradually into a more advanced one by adding detailed information to HFPNe elements in the biological pathways e g kinetics data and detailed regulation rules To build a model quickly drag and drop biological elements from the Biological Elements frame Section 5 4 4 and link them with connectors observing the connection rules listed in Section 3 5 All element parameters are initialized with default values the initial value of an entity 1s 0 the threshold parameter of an input connector 1s 0 and the speed parameter of a process 1s 1 by default For any biological process to take place during simulation the process representing it needs to be enabled A process is enabled when all the following three conditions are satisfied 1 all of its incoming process and association connectors are activated 11 no input inhibi
50. gure and biochemical knowledge about reactions Figure 2 shows the model representation that we have described with Cell Illustrator 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 78 of 88 Fas ligand d cytochrome c comp ex 3g Lol T oe ipis oe 3 T B T spase 9 ER t comp O DFF complex Q1 Cleaved DFF45 4DFFA5 cleave release cytochrome c from mitochondr ion Figure 2 A CI model representing the Fas induced apoptosis obtained from Figure 1 For Bid m11 Procaspase 9 m21 Procaspase 3 m25 DFF m30 DNA m37 the initial concentration of each compound is assumed to be 100 On the other hand for FADD m4 Procaspase 8 m5 Apaf 1 m17 dATP ADP m18 when two compounds react together without the stimulation of apoptosis the initial concentrations and the rate are assumed to be 39 039 and m1 m2 5000 respectively to keep the stable state condition Each compound is assumed to be produced by the rate of 0 5 represented by a process without any incoming connector and to degrade by the rate of its concentration divided by 200 represented by a process without any 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 79 of 88 outgoing connector which keeps its concentration at 100 under the stable state condition This degradation rate also applies t
51. gy 4 3 271 29 2 Doi A Nagasaki M Fujita S Matsuno H Miyano S 2004 Genomic Object Net II Modeling biopathways by hybrid functional Petri net with extension Applied Bioinformatics 2 185 188 3 Matsuno H Doi A Hirata Y Miyano S 2001 XML documentation of biopathways and their simulations in Genomic Object Net Genome Informatics 12 54 62 4 Matsuno H Doi A Nagasaki M Miyano S 2000 Hybrid Petri net representation of gene regulatory network Pacific Symposium on Biocomputing 5 341 352 5 Matsuno H Fujita S Doi A Nagasaki M Miyano S 2003 Towards biopathway modeling and simulation In 24th International Conference on Applications and Theory of Petri Nets ICATPN 2003 Volume 2679 Lecture Notes in Computer Science pp 3 22 6 Matsuno H Murakami R Yamane R Yamasaki N Fujita S Yoshimori H Miyano S 2003 Boundary formation by notch signaling in Drosophila multicellular systems experimental observations and gene network modeling by Genomic Object Net Pacific Symposium on Biocomputing 8 152 163 7 Matsuno H Tanaka Y Aoshima H Doi A Matsui M Miyano S 2003 Biopathways representation and simulation on hybrid functional Petri net In Silico Biol 3 389 404 http www bioinfo de isb toc vol 03 html 8 Nagasaki M 2004 A Platform for Biopathway Modeling Simulation and Recreating Biopathway Databases Towards Simulation Ph D Thesis The University of Tokyo http genomicobject net pub nagasaki phd pdf 9
52. hanism s dynamic behavior e Model import previously created models or sub models can be imported from SBML CellML files or from public or proprietary libraries and databases like KEGG 9 12 The documentation for the Cell Illustrator software includes the manuals listed below Cell Illustrator User Manual Introduction to Cell Illustrator system Cell Illustrator Reference Manual Detailed description of Cell Illustrator functionalities Cell Illustrator Player Reference Manual Detailed description of Cell Illustrator Player functionalities This manual describes version 5 0 of Cell Illustrator Online CIOS 0 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 4 of 88 2 Getting Started 2 Cell Illustrator Online Cell Illustrator is available as a Java Web Start application CI Online CI Online can be run directly from the website without the need to install software on local computer CI Online is a client server solution that enables the user to create manipulate and simulate bio pathway models locally within CI workspace and to store and organize them on the remote CI Online Server Additionally CI Online Server gives the user convenient access to libraries and databases of bio pathway models 2 2 Run Cell Illustrator Online CIO In order to run Cell Illustrator correctly Java version j2sdk 1 5 0 or higher needs to be installed Java Runtime Environments
53. he set p53 mdm2 p53 mdm2 separate process are possible in Stepl and Step2 i e discrete or continuous discrete or continuous discrete or continuous discrete and continuous continuous continuous continuous as in Figure 38 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 29 of 88 p53 mi e p1 c9P53 mdm m3 p53 m16 p1 c2P53_mdm2 m18 50 1 0 0 0 50 50 10 0 0 50 0 c3 c3 mdm2 _ m2 mdm2 m 0 0 p53 m4 c1 p1 c2 93 mdm m6 p53 mig T p1 2053 mdm2 m21 50 50 10 e 50 0 c3 mdm2 mdm2 m28 0 0 0 0 p53 m7 el p1 c2P53 mdm mg p53 m22 el p1 c2P53 mdm m24 50 0 mdm2 mdm2 m23 0 0 p53 _ m10 m p1 c9P53 mdm m12 p53 m25 p1 2053 mdm m27 50 0 f 1 0 is 50 50 0 T 10 00 soo C C mdm2 m mdm2 m 26 0 0 0 0 p53 m13 T p1 oh53_mdm2 m15 0 0 Figure 38 The results of the simulation are presented below Figure 39 Figure 39 If the speed of the separate process strictly depends on the value of p53 mdm2 say ml this can be expressed by a respective equation defining the speed parameter of the process In Figure 40 m3 20 for the model where all elements are of the continuous type 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 30 of 88 53 53 3 p mi pi eg mdm2 m 50 0 m3i20 00 50 0 c3 mdm mz 0 0 Figure 40 The results of
54. he stroke pattern of selected elements e Toggle Grid Turn the background grid on or off Cm e gt Toggle Antialiasing Turn on or off antialiasing 5 3 Change Element Properties For additional editing of the properties of elements in the model the Property Frame toolbar at the far right of the window is provided Note that the Window Show Frame menu contains all options provided in this toolbar The actions for each button are as follows Element Lists show the Element Lists frame see Section 5 4 R a Element Settings show the Element Settings frame see Section 5 4 kl Biological Properties show the Biological Properties frame e Q3 External References show the External References frame see Section 5 4 e X Navigator show a small view of the entire model to navigate the model in the canvas window see Section 7 1 e V view Settings show the View Settings frame In this frame the user can customize the global view settings of the active canvas e g show and hide entity labels e Chart Settings show the Chart Settings frame to create time series plots see Section 6 2 3 O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 38 of 88 x Simulation Settings show the Simulation Settings frame to change simulation parameters see Section 6 2 e Simula
55. how Mame Show Firing Operation Show Firing Style El Shape Depth Line Color EH 0 0 0 255 Line Style Straight E UR JaA 71 1 Element Settings e View Settings Chart Settings a Simulation Settings Figure 58 5 4 4 Biological Properties Frame The Biological Properties Frame displays a list of all elements of the selected type Entity Process Connector Fact Edge Fact Vertex Group along with their biological properties and enables you to view assign additional biological information such as location cell component biological role biological event GO ID list of Pubmed ID s etc similarly to the Element Lists frame in this frame you can easily change a property of an element by double clicking on the corresponding cell in the table For selected properties such as Cell Component or Biological Event you can select the values from a list of predefined choices Please note that changing the Biological Event of a process will assign a new graphical symbol image to the process on the canvas Clicking on the column header you can sort or filter the table You can also control if a given column property should be visible or hidden in the table by a right click on the column header and choosing the Select Visible Column item from the popup menu Any selection made in the Element List will be immediately reflected in the canvas and vice versa 2002 20 Human Genome Center Institute of Medical Science The
56. ia Print and Print Area options in the File menu 7 4 Export Options Analysis of model structure and simulation results can be also carried out outside Cell Illustrator with third party tools The options Export Entity List Export Process List and Export Connector List in the File menu store respective data in a file in the CSV Comma Separated Values format Such a file can be later uploaded to a spreadsheet for example The option Export CSO Model available since version 4 0 allows to export model to Cell System Ontology 3 0 format Cell Illustrator 5 0 Professional demo intra csml org a x Edit A Mew Element Simulation View Analyze Window Help PFE ELTE LTTE Ctrl M e Open ctri O e Saphila z ceml P Fas induced a pathway 1 csml 4 gt El it al Open Recent Files i m I m Close File Chri ai Sy rinrebrold a Ea a3ndirespmo Ee Clase Al Files S3 x pd wi B l 0 5 A threshold EE Save Ctrl 5 mind l 30 a m P Save As Ctreshift Behe Jhreshak m a rr 1 B z m Insert Li Yao AVE i m8 amp m1 1 2500 c Export d Simulation Source Code gt Export All Image 57 Look amp Feel CSO Model CE Print visible Area Ctrl Shift F SYG beta m1 2 200 q DE E EA A and mI E Print Ctrl P Html Report beta Ctrl g CIL simulation lag and a 300E 6 Sigla 0 mi3 200
57. in CI Player show entity value changes 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 75 of 88 over the whole simulation period with a vertical bar indicating the time point visualized in the canvas Figure 87 Process state indication in CI Player from left to right firing waiting and not firing One advantage of replaying the simulation from a log as opposed to rerunning it in Cell Illustrator is the capability to deterministically repeat the simulation even for a model of stochastic nature Also you can quickly navigate to a specific part of simulation and step through it forwards and backwards for debugging purposes For more details please refer to the CI Player online help and CI Player Reference Manual You can open them from the CI Player Help menu O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 76 of 88 Appendix A Tutorial Modeling Fas Ligand Induced Apoptosis Apoptosis Induced by Fas Ligand Apoptosis programmed cell death is known to participate in various biological processes such as development maintenance of tissue homeostasis and elimination of cancer cells Malfunctions of apoptosis have been implicated in many forms of human diseases such as neurodegenerative diseases AIDS and ischemic stroke Reportedly apoptosis 1s caused by various inducers such as chemical compounds prote
58. ins or removal of NGF The biochemical pathways of apoptosis are complex and depend on both the cells and the inducers Fas induced apoptosis has been studied in detail and its mechanism has been proposed as shown in Figure 1 4 We will represent this mechanism as a model with Cell Illustrator Fas ligand m1 cell membrane Fas receptor Apoptptic Stimuli Apaf 1 m17 exces FADD m4 Y H Bax Bad Bim D Procaspase 8 m5 k Be 2 Bcl XI dATP ATP Bid i terminal 2 m11 m13 Cytochrome c m18 e e E Mitochondr ion m16 Caspase 8 Procaspase 3 m25 Alo m8 RR o o gw 9 Ww C m23 m20 y aspase 3 am Q t gh gt in 9 m21 DFF m30 A DFF45 m33 34 35 nucleus 2 Pe B igomer of DFF40 AN m36 m22 DNA fragmentation 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 77 of 88 Figure 1 Proposed steps of apoptosis induced by Fas ligand Fas ligands which usually exist as trimmers bind and activate their receptors by inducing receptor trimerization Activated receptors recruit adaptor molecules such as Fas associating protein with death domain FADD which recruit procaspase 8 to the receptor complex where it undergoes autocatalytic activation Activated caspase 8 activates caspase 3 through two pathways the complex one occurs when caspase 8 cleaves Bcl 2 interacting protein Bid and its C
59. ion connecting with this transition to use their process functions firing speeds Arc Connector Universal Place A universal place can handle a list of variables of various types double Boolean string etc Continuous Transition u Continuous Process A continuous transition 1s a continuous E iot A biological process or a reaction like process consuming quantities of its input n aoe transcription translation or enzymatic places and producing quantities into its l l ME pip Glico edene a Co MONS reaction which consumes some entities in ntity and pr her flow that is specified by a speed of firing npu quani ty mop ee v output according to a rate formula Process Connector Connects input entities to a process and process to output entities Its weight parameter specifies an activation threshold Process Connector allows quantities flow in the model Normal Arc A directed arc Connects a transition to a place or connects a place to a transition An arc has a weight Inhibitory Connector Prevents a process from being activated Inhibitory Connector facilitates the modeling of the inhibition competition process Connects a place to a transition An inhibitory arc disables a transition If the quantity in its source place is greater than the weight of the arc the transition cannot fire Inhibitory Arc Association Connector Connects entities without causing Test
60. ion 1s high Figure 3 However the detailed mechanisms of the selections of these two pathways from caspase 8 are still unclear and need to be studied in future since the presence of autocatalytic process 1s proposed in caspases 2 1t 1s included in our model Figure 4 which increases the DNA fragmentation as shown in Figure 3 pro caspase Figure 4 A model representation of autocatalytic process in Figure 2 However if the large rate of the autocatalytic process 1s assumed in the caspase reaction the DNA fragmentation becomes independent of the Fas ligand concentration which then disagrees with the experimental results Therefore we can guess that autocatalytic processes must be slow if they are present To examine the effect of autocatalytic processes of caspases on the apoptosis by Fas ligand DNA fragmentation 1s simulated when the stimulation by Fas ligand 1s stopped after a short period Table 2 shows a simulation result where the apoptosis proceeds more with the increase of the autocatalytic rate of caspases even for a short period stimulation O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 81 of 88 Table 2 DNA fragmentation at four autocatalytic rates of caspases rate0 0 rate l mA mB 80000 rate2 mA mB 40000 and rate3 mA mB 25000 which are assigned to the process T4 1n Figure 4 The stop time represents the period after that Fas ligand stimulation i
61. is connected to e2 with quantity m2 the equations for quantities change are un me 0 05 ml dt dt The value of entity el 1 e ml decreases while the value of entity e2 1 e m2 increases during simulation The equation 0 05 m1 of process pl governs this change of value per unit time see Section 6 1 for Simulation Details The results of the simulation are presented below Figure 13 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 17 of 88 Figure 13 If consumption and generation speed should be different the Connector Custom kinetic style should be used and the corresponding speed add parameters for connectors cl and c2 should be set In Figure 14 the speed of c1 and c2 is 0 05 ml and 0 1 ml respectively Figure 14 The results of the simulation are presented below Figure 15 chart 1 BN xj Figure 15 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 18 of 88 4 Biological Process Modeling with Cell Illustrator This section briefly explains how to simulate major biological processes with CI The examples have been designed to help the novice user to become familiar with the Hybrid Petri Net modeling method and notation 4 1 Degradation In a cell nearly all mRNAs and proteins fragment into parts at certain speeds This fragmentation activity 1s called degradation A mo
62. l 1 0 0 0 1 1 0 0 Figure 42 The results of the simulation are presented below Figure 43 Chart 1 daxarabucin mRNA ps3 mRNA p53 a doxorobucin doxorubucin mRFWA_ p53 mRNA pos NM Figure 43 If the threshold of the inhibitory connector is modified to 5 Figure 44 the results will be different The results of the simulation are presented below Figure 45 p2 doxorubicine m2 p2 doxorubicine ma 1 1 0 0 5l MRNA p53 ml pi MRNA p53 m3 1 0 0 0 1 1 0 0 Figure 44 2002 20 Human Genome Center Institute of Medical Page 32 of 88 Science The University of Tokyo All rights reserved Chart 1 EJ Mame doxorobucin doxorubucin mRNA pSS mRNA p53 l doxorobucin doxorubucin mFAA pos mENA ps3 mE Figure 45 As in Figure 43 and Figure 45 if doxorubicine is treated the transcription of p53 is inhibited The inhibitory activity will be adjusted with the threshold of the inhibitory connector 4 10 Catalysis Enzymes catalyze a variety of chemical reactions The model of an enzymatic reaction can be created with the following steps The first three steps are similar to the translocation process in Section 4 2 Stepl Create two entities and change their names e g p53 and p53 P and set their values e g 10 and 0 Step2 Create a process and set its name e g catalysis Connect p53 to the process and connect the process to p53 P Step3 Set the catalysis spe
63. l number Marking A variable representing the state of a place O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Discrete Entity A countable biological component or event that is quantified or represented by an integer e g the number of molecules locating at the membrane an event that describes whether a protein is binding to DNA or not Quantity The number of entity items present e g the number of molecules binding to DNA ON OFF state of a gene expression etc Discrete Process A biological reaction that converts quantities in discrete entities into quantities in other discrete entities A switching mechanism can be also handled as a discrete process which receives some signals and sends out signals for a pathway Continuous Entity A biological entity like a concentration of a protein enzyme or ion etc the quantity of which can be represented as a real number Concentration Quantity System status representing quantity or concentration of proteins enzymes and ions Page 8 of 88 Generic Entity A biological chemical physical entity that gt is not pre defined May contain more data types like string that is used to represent DNA sequence Universal transition Generic Process A transition for a universal place A A complicated biological chemical universal transition allows the arcs physical process e g a translat
64. m kinetic style you need to specify directly its Kinetic Script i e the formula that governs the value change For other kinetic styles the kinetic script is predefined and you can only set the values of parameters in the equation 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 43 of 88 The Delay option is enabled for a discrete or continuous process see Section 3 4 and it takes a positive real value or zero A discrete process will fire just after its delay time if the process is still active Otherwise it can lose its chance to fire The delay 1s specified in Petri net time pt By default the Activity property of a process 1s set to true However you can write a script in the Activity field to turn the process on and off in the course of a simulation For an example on how to write a script see Section 5 6 In the View and Shape category you can modify the display properties associated with the process 5 4 3 Connectors Element Lists Entity Process Connector Fact Edge Fact vertex Group From To Firing Firing Visible pl Threshold t pl Threshald 0 pz Threshald 0 A p3 Threshald 0 Threshod S O do InputProcess InputProcess InputProcess Inpubassociation X X1 lt 1 XI rm tal al Irit iriri Biological Pr Simulation Hi 3 Simulation E E Gene Mining Path Search Figure 57 In the Connecto
65. n the column header and choosing the Select Visible Column item from the popup menu Element Lists Entity Process Connector Fact Edge Fact Vertex Group o CSO 7 Type Varia Initial Eval Visible Depth ma poe m pP Tu f o Protein Double m3 100 Imi T Biological Pr F Simulation Hi Q Simulation E Ii Gene Mining Fath Search Figure 53 If an entity is selected in the canvas and the Element Settings frame is activated the properties of the selected entity are presented there Figure 54 The type of entity cannot be edited in Element Lists However it can be changed in the Element Settings frame A continuous entity can only have the Double value type while a discrete entity be it Integer or Long or a generic entity can have the value type String or Boolean The Maximum Value and Minimum Value fields define the allowable value range The display properties associated with the selected entity can be modified in the View and Shape categories 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 40 of 88 Element Settings El Entity Element Id Name C50 Class Children Parents All Edges H Simulation Type Double Variable mz Initial Value 10 Eval Once Max Value Min Value Log alobal View visible Show Mame Show Variable Element Settings d View Settings Chart Settings a Simulation Setti
66. nce The University of Tokyo All rights reserved Page 36 of 88 to enter the initial value for the selected entity Note that for discrete entities this value must be an integer A process can have up to four associated labels see Figure 3 At the upper left is the Name of the process The lower right label is the Kinetic Script of the process The upper right label is the Firing Style of the process and the lower left label 1s the Delay of the process To add a connector to link two elements select a connector type from l and icons in the Draw toolbar above the canvas pane and then click on the element where the connection originates Next click on the element you want to connect to The connector will be automatically drawn to connect the two elements Note that a connector links entities to processes not entities to entities or processes to processes If you draw a connector between two entities a process with default parameters is inserted between them Also discrete entities cannot connect to continuous processes see Element Combinations in Section 3 5 Generic entities can only connect to generic processes Each connector has up to three labels above the connector is its name below is its firing style and below the connector is its threshold if threshold firing style is enabled or the rule if rule firing style is enabled see Figure 2 The threshold is a minimum value for the activation of the connector see Simulation
67. ngs Figure 54 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 41 of 88 5 4 2 Processes In the Process List in the Element Lists Figure 55 you can change the name of the process to reflect its biological meaning Also you can set the Kinetic Script and Delay Script property The Kinetic Script and Delay Script fields define respectively the speed add update function and delay parameter of the processes The Kinetic Value and Delay Value columns display current simulation values Element Lists Entity Process Connector Fact Edge Fact Vertex Group Mame Type Kinetic Kinetic Script Visible Continous 0 5 m2 ml Continous 1 m3 Conci us Biological Pr ir Simulation Hi Simulation E an Gene Mining Path Search Figure 55 If a process is selected in a canvas the Element Settings frame displays the properties of the selected process Figure 56 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 42 of 88 Element Settings El Process Element Id Mame C50 Class H Simulation Tvpe Calc Style Firing Style Kinetic Style Kinetic Script coefficient coefficient c4 stoichiometry cS stoichiometry Kinetic Script Language Delay Delay Script Delay Script Language Activity Activity Script Language p3 p3 ProcessBiological Continous Speed
68. nterval option in the Simulation Settings frame 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 64 of 88 6 3 Viewing and Analyzing Simulation Results 6 3 1 Simulation Charts Entity values in charts are color coded The Entity List on the right side of the window enables you to change the visualization and color of each single entity plot Right click on a chart to open a pop up menu with options to customize the chart Figure 78 caspase Properties Save a5 Print Zoom In i SO ee eee Zoom Out 1000 1 250 1 500 Auto Range traspase8 caspase3 caspases pro caspa Rasterize pra caspase3 pro caspases Copy to Clipboard Figure 78 e Properties opens a dialog box in which you can set the chart visualization details e Save as Turns on and off the tool tip for the graph If the option is enabled you can see the value of the data point once you move the mouse cursor over it e Zoom In Zoom Out Turns on and off the zoom in feature You can drag the mouse cursor over an area in the chart to zoom into it e Auto Range adjusts the range of x and y axis e Rasterize freeze a specified chart for later analysis or comparison e Copy To Clipboard Copy the time series to system clipboard in a tab separated text format You can paste the chart data to another application
69. o model repression see Section 4 9 Association connectors are used to model the situation when entities and processes need to be linked together but nothing is produced or consumed They do not allow transport of quantities They are used when the input entities concentrations do not change see Section 4 10 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 10 of 88 3 4 Processes Processes Figure 3 define the rate of entity value changes and interactions among entities Processes are used to model biological reactions such as enzymatic reactions or protein complex formation processes Processes can take multiple inputs e g consume different entities to produce new entities and have multiple outputs e g produce different entities such as breaking down sacharose into glucose and fructose Also like entities processes can be continuous continuous processes discrete discrete processes or generic generic processes depending on the input output and timing Only generic processes are able to handle generic entities i i name migrate phosphorylate translate Bliiincgstle Jand MI Hand delay kinetic script m1 3 1 1 1 i Continuous Discrete Generic Process Process Process Figure 3 In general terms if all input connectors of a process are activated the process is said to be enabled see Section 6 1 When a process is enabled the reaction
70. o other compounds in the network The rate of other processes is determined roughly by following Table 1 Synthesis and catabolism processes are added in the model for all proteins Autocatalytic processes are also added 1n the model to all caspases since they exist as proenzymes The pathway from caspase 8 to caspase 3 is assumed when the caspase 8 concentration is over 30 Protease is often synthesized as a proenzyme zymogen and changed to active form by other enzymes or by itself So the autocatalytic process is added to every caspase reaction Table 1 Functions assigned to continuous processes in the simulation of apoptosis induced by Fas ligand where mA and mB represent the contents of the corresponding continuous entities Rate Unimolecular reaction Bimolecular reaction Self effacement mA 200 Oligomer mA 20 mA mB 10000 Monomer mA 10 mA mB 5000 Enzyme binding mA 5 mA mB 2500 Enzyme reaction mA 10 By using the apoptosis scheme modeled as a model we simulated the amount of DNA fragmentation by varying the Fas ligand concentration and Figure 3 shows the simulated relationship It shows that under very weak stimulation very low amount of Fas ligand DNA fragmentation does not occur since the stimulation stops at the intermediate point because of the assumption of degradation processes With the increase of the stimulation the reaction proceeds to the backward intermediates and DNA fragmentation cell death occurs finally which increases
71. on e g Simulation Time Sampling Interval Plot Update Interval Log Update Interval etc are used as the default parameters when a new model is created The Canvas options define the way how a newly created canvas will look The background color and the grid size can be defined here Alternatively you can update major canvas options in the View Settings frame 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 56 of 88 5 8 Graph Layout Frame In this frame you can beautify the graph layout of the model This function can be especially useful when importing models from the BioPACS database or converting models from other formats Graph Layout B CB sds Eades Random rid Adjustment Sugiyama CUL Tree RadialTree Moen Circle FastOrganic MormalOrganic PrettyOrganic Enlarge Enlarge width EnlargeHeight Shrink Shrinkyyidth ShrinkHeight BendEdge BendEdge Bezier BendEdget Spline Information ee Option a Biolo a Parts B Prajec c Sear Bio E Grap Figure 72 The Graph Layout frame has several options for placing elements on the canvas such as Annealing Layout Circle Layout and others 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 57 of 88 6 Simulation 6 1 Simulation Concept The simulation architecture o
72. on Speed option in the Simulation Settings frame On the other hand for simulation started with Fast Play and Max Speed Play Cell Illustrator attempts to execute simulation at the rate defined by the Simulation Speed For example if the Simulation Speed is set to 100 then the simulation will proceed at 100 Petri net time units per 1 second of real time provided that your hardware is sufficiently fast You may need to increase the Sampling Interval at the cost of result accuracy if the simulation is too slow The difference between Fast Play and Max Speed Play 1s that the canvas is not updated until the simulation end if the latter option is executed Thus you should use the Max Speed Play for time consuming models O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 61 of 88 6 2 2 Simulation Engine Cell Illustrator offers two alternative engines for performing the simulation O standard engine O simulation engine code generator SECG Both engines base on the same simulation model Hybrid Petri net with extensions and give the same results However SECG executes the simulation in a different way first Java source code is generated for the model to be simulated then it is executed as a usual Java program Use SECG option determines whether to use SECG or the standard engine for the simulation execution The SECG engine is recommended for the following two cases Simulation of
73. p5 c15 p53 mdm2 m15 100 0 mdm2 mi4 50 0 Figure 28 The simulation results of the continuous model are presented below Figure 29 Chart 1 p53 mdm2 p53 mdm2 Figure 29 If the complex speed strictly depends on the value of p53 m1 and mdm2 m2 the speed of the complex process needs to be m1 m2 400 the factor 1 400 1s 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved P age 25 of 88 an example for the model where all elements are of the continuous type as in Figure 30 pos ml pi 53 mdmz ms c3 P 100 0 m1 m2 400 Ti meme n 50 0 Figure 30 The results of the simulation are presented below Figure 31 Chart 1 3 40 50 0 fo p53 mdm2 p53 mcdm2 Figure 31 4 5 Tetramerization In a cell some proteins create a self complex under certain conditions For example p53 can create a homotetramer The process of creating tetramers is called tetramerization The three examples below show how various kinetic styles processes can be used for the modeling of tetramerization The speed of tetramerization is linearly dependent on the mass of monomers in the first example In the second example stochastic perturbation is introduced Finally in the most advanced example tetramization is defined by a script A simple model of tetramerization can be created using the following steps Step 1 Create two entities and
74. r List of the Element Lists frame the Name Type From To Firing Style Firing Script and Visible properties are displayed Figure 57 When a single connector is selected in a canvas the Element Settings frame displays its properties Figure 58 The major properties that you can edit are CSO Class Connector Firing Style Firing Threshold and Firing Rule The CSO Class defines the connector type which can have the values process inhibitory or association The Connector Firing Style option defines how the condition for activating the connector is calculated It should not be mistaken with the Process Firing Style parameter The Connector Firing Style can take the Threshold Rule or No Check value In the Threshold style the value of the Firing Threshold determines if the connector is enabled if the current value of the source entity is larger than the threshold then the connector is enabled In case the Connector Firing Style is set to Rule the behavior of the connector is defined by a boolean expression script in the Firing Rule field No Check option always enables the connector 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 44 of 88 Element Settings E Connector Element Id cl Mame cl C50 Class InputProcess H Simulation Type InputProcess Kinetic Script 1 p 5 m2 m1 Firing Style Threshold Firing Script T Firing Script Language Log View visible S
75. re 5 Sampling interval is 10 pt 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 85 of 88 Appendix C Kinetic Styles This chapter gives more details on the kinetic style concept The table below specifies the mathematical formula that 1s used to calculate the entity value change delta for all the various kinetic styles Also the parameters used for each kinetic style are listed in the table Kinetic style Custom parameter speed add connector custom parameters speed add update for each connector Mass parameters coefficient coefficient stoichiometry for each connector 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Process type continuous speed calc style discrete add calc style continuous speed calc style discrete add calc style continuous speed calc style discrete add calc style Formula for the change value delta speed sampling interval OGiGa Connecuor Speec sampling interval CODDector ada derta lt E product UueDerrttcolente COSLITCIento m mber OF Input sent ves es stoichiometry sampling interval where m product is the product of all input entity values For the input entity variables mi m Mn m product m ma 7 eT derta mMm product cOoerrh5screnbr COSTITOTOIBE Inumb
76. rights reserved Page 52 of 88 Cell Illustrator 5 0 Professional demo intra csmLorg File Edit Element Simulation View Analyze Window Help tNJOOGQDOmeBllif saacumpmw ies LX Be T mmcey 000000 aax S O 1 1 Carbohydrate Metabolism A y Glycolysis _ Gluconeogenesis 1 csm LAE Glvcolysis Gluconeogenesis J9 Open KEGG Image Citrate cycle TCA cycle Pentose phosphate pathway Pentose and glucuronate interconversic dummy39 ns Fructose and mannose metabolism m34 ij Galactose metabolism i o Ascorbate and aldarate metabolism 5 3357 2 7 0511 iti Starch and sucrose metabolism Aminosugars metabolism Nucleotide sugars metabolism Pyruvate metabolism Glyoxylate and dicarboxylate metabolisi Propanoate metabolism Butanoate metabolism C5 Branched dibasic acid metabolism Inositol phosphate metabolism Global metabolism map qus 1 2 Fnerav Metabolic dummy obj B B e xx P gt r a gt ln C Add Inhibitor Add Effector Add Cofactor att ir ENRIKE E DTI n BIoPACS Library j E E E ir o T Gene Mining Mouse position O 1 Show frame BioPACS Library Done Selection 136 132 490 0 12 38 53 73M z 508M Figure 65 Importing KEGG maps with BioPACS 5 6 Script Editor An advanced user of Cell Illustrator may need to write a custom function to specify changes of a property value With the Script Editor you can write
77. ring the simulation These options have no meaning for simulations started with the Max Speed Play command The Weak Firing option solves a problem of non firing processes This may happen when the value associated with an entity is smaller than the calculated change of this value delta within a given sampling interval If this happens the process does not fire at all This behavior is an inherent consequence of the classical Petri net formalism However it may lead to an unexpected unrealistic simulation behavior e g when the kinetics of catalyzed reactions in biological systems are modeled a catalyzed reaction may not occur at all even if a substrate and a catalyst are present The Weak Firing can be enabled disabled for continuous or discrete processes By default the this option 1s off which is in agreement with the Petri Net model but is sometimes difficult to understand for normal biological reactions Cell Illustrator has several commands to run a simulation If the Play Play with Animation Fast Play and Max Speed Play 9 buttons in the Simulation toolbar are clicked the simulation will run until the end Clicking on Step Play and Step Play with Animation buttons executes a simulation step corresponding to one sampling interval If a simulation process is started with the Play or a Play with Animation button the program attempts to execute the simulation at the speed 1 pt per 1 second regardless of the Simulati
78. s stopped The initial Fas ligand concentration is set to be n 600 Variables mA and mB represent the contents of the continuous entities going into T4 DNA Fragmentation Stop time rate0 ratel rate2 rate3 10 0 0 0 1169 15 25 442 746 1862 20 417 581 885 2048 Figure 5 shows simulated time courses of the model in Figure 2 with CI Some intermediates during apoptosis at three levels of Fas ligand concentrations are measured These time courses might be useful to plan new experiments such as addition of inhibitors to various steps However it 1s necessary to estimate the realistic rates of each reaction by comparison with the experimental data It is also necessary to add other pathways through the Bcl 2 family 1 or p53 to describe the real apoptosis held in various cells and by various inducers CI 1s a very useful tool for biochemists to describe the complex biological pathways semi quantitatively on a figure n 210 Fas ligand caspase8 cytochrome c 600 40 60 20 30 15 300 20 30 10 10 t 5 0 0 0 0 0 1 2 3 0 1000 2000 3000 0 1000 2000 3000 n 450 600 40 60 20 30 15 300 20 30 10 10 t t 5 o 0 0 0 0 1 2 3 0 1000 2000 3000 0 1000 2000 3000 n 600 600 40 60 20 30 15 300 20 30 10 t 10 t t 5 0 0 0 0 0 1 2 3 0 1000 2000 3000 0 1000 2000 3000 0 caspase9 caspase3 90 60 30 t 0 1000 2000 3000 0 1000 1000 2000 90 60 30 t 0 3000 0 1000 90
79. scades gene regulatory pathways as well as dynamic interactions of various biological entities such as genomic DNA mRNA and proteins Cell Illustrator models are used to visualize biopathways interpret experimental data and test hypotheses In addition it provides researchers with model diagrams of publication quality and simulation result charts Cell Illustrator has been successfully utilized to model biological pathways like Circadian Rhythms of Drosophila Glycolytic pathway and Fas ligand induced Apoptosis 1 2 3 4 5 6 and 7 Using the graphical user interface of Cell Illustrator researchers can create and simulate their own molecular pathway models Some of the key features of Cell Illustrator are a Pathway construction and visualization construct models graphically by drag and drop entities that represent bio components in the biological pathways b Pathways simulation specify mathematical formulas for biochemical reactions in the pathway for simulation The simulation can be run inside the workspace window in an interactive mode or sent to a remote server Cell Illustrator Server c Simulation replay the simulation results can be logged and then visualized and analyzed in Cell Illustrator Player d Simulation animation Cell Illustrator s simulation results can be edited into an animation script file and then be used as the input to Cell Animator a module that produces high quality presentation of the studied mec
80. set their names e g p53 monomer and p53 tetramer Set the initial value of the monomer to 10 Step 2 Create a process and set its name e g tetramerization Step 3 Connect the p53 monomer entity to the process with a process connector 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 26 of 88 Step 4 Connect the process to the p53 tetramer entity with a process connector Step 5 Set the Kinetic Style of the process to Mass and set c Stoichiometry to 4 Leave the default values for the other parameters of the process Coefficient 0 01 Coefficient2 0 1 and c2 Stoichiometry 1 p53 tetramer m2 p53imanomer mi pi ci c2 10 J m1 0 0010 Figure 32 With the model as in Figure 32 the speed of the tetramerization process depends on the value of p53 tetramer To make the process of tetramerization faster the Coefficient value should be increased To change tetramerization into the dimerization process the cl stoichiometry value should be changed to 2 More details on parameters related to the Mass kinetic style can be found in the section 3 4 Processes The results of the simulation are presented below Figure 33 Chart 1 Mame ps3 menomer p53t tetramer p53 monomer p53 tetramer Figure 33 4 6 Stochastic Tetramerization Normal biological processes e g
81. sponding process and the function v t is the speed of the corresponding process at the time t Note that CI checks and prohibits the connections that are not allowed during model creation O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 12 of 88 Table 3 Input Connectors d t v t Process is enabled at intervals of the time d t while m t gt w t holds On w t mim o Process is enabled at intervals of the time d t while m t gt w t holds but no quantity change m t w t M t Process IS not enabled while m t w t holds Not Allowed On w t E Process is enabled while m t w t holds but no quantity change m t w t C Process IS enabled m t gt w t holds not while 3 6 Discrete Elements Example 1 m t w t d t v t Process is enabled at intervals of the time d t while m t gt w t holds On w t mm c Process is enabled at intervals of the time d t while m t gt w t holds but no quantity change EN c t Process IS enabled m t gt w t holds not while iT r a F ce ce Figure 4 v t Process is enabled at the speed of v t while m t gt w t holds Process is enabled while m t gt w t holds but no quantity change m t w t C Process IS enabled m t gt w t holds not while me The model in Figure 4 consists of two discrete entities el
82. tangle marking the portion of the model currently shown in the active canvas You can drag this rectangle around in order to navigate the model in the canvas window Navigator E ADR i a ns bo im 7 et hdi a mt D ae I TENE i a4 al occ fe I I I I Xx Navigator a Parts Library Po Project Manager Figure 82 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 70 of 88 7 2 Finder With the Finder frame you can search for specified elements in the active canvas Figure 83 As with the Navigator the Finder becomes very useful when the model in the canvas is large or very complicated x Search Far FR v Regular Expression Case Sensitive M only visible Search In v Search Ids W Search Names M Search Comments Search Variable Mames Search Variable Names in Script v Gene Extension Gene Probe ID Found 4 entity Fas ligand trimer entity Fas ligand entity Fas ligand trimer Fas receptor complex entity FADD Close Figure 83 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 71 of 88 7 3 Print and Print Area In CI you can output the whole canvas a part of active canvas or the simulation charts to a printer or an image file in the Postscript PNG or JPEG format This functionality is available v
83. tet qud eso v tadius sntus deno dinc totaal 23 zs iaceqee e m TU 24 dd COmpleX amsssdusicui tbe EEA 26 4 pL ettamerzalofisssismestee toc sU C XI M IM ADI DE LEE 28 4 6 Slochasuc TetramerizaliOrissesss eio diei aep aet uda dee ac date doo Eo ER Gn biens 29 4 7 Custom TeltramellzauODiuc cisco entis otov a a ense fel DL ars 30 4 o Separators busted eupvesta supe dec Dui otim od I Eo UN ME pAR I RREIU 31 A NND NR RR Tet Eee een EN ae Te ner Stee 33 BB VO ALAN nc E 35 5 CELL ILLUSTRATOR USER INTERFACE AND MODEL CREATION 38 XM PACCiEIO MOMS TE ETUR 38 5 2 WIOGINY a INNO CON accu at cence etate te duiladaketer dosi ten ie dati S Dra ote e ced 39 5 9 Change Element Properties 2 0 mbi b aae dot aget irienn 40 5 4 Element Settings Frame and Element Lists Frame 41 9 9 Biological Elements and Pathway Fragments suusssse 48 9 5 SCIIDE E OIEO carere iria te emiten ead a pd qun TA 55 Ou Preterences ETaITIG oce poreitieasoimiedborbiteLein eol P etas u dieci o oie tenir ncc cs Seu 58 5 0 Graph Layout Frame couture ee eei dvi ua aoa a opc ute rax adea 59 0 SIMULATION roren E 60 6 1 Simulation CON Ept painaa a a aaa 60 2002 20 Human Genome Center Institute of Medical Page 2 of 88 Science The University of Tokyo All rights reserved 6 2 How to Set Up and Run Simulation cece cecceeecseeeeeeeeeeeeeneeaeeeeeaees 62 6 3 Viewing and
84. that it represents will fire will be executed immediately or after a certain time interval This will cause a change in value for all connected entities For example 1f a process represents the reaction of a protein precursor which turns into an activated protein by phosphorylation the decrease of the precursor protein and the increase of the active protein on account of the speed of phosphorylation 1s simulated once the process fires The exact conditions for enabling a process depend on its Firing Style parameter and are described in detail along with other process parameters in Section 54 2 The process type determines its calculation style The continuous process has the speed calculation style the process calculates the speed of production or consumption of the entities On the other hand the discrete process has the add style the process calculates the concrete values that should be added subtracted to the entities Finally the calculation style of the generic process 1s update 1 e the process calculates a concrete value that will replace the entity value The Kinetic Script parameter specifies how to calculate the value change of connected entities in each of the three cases See the Reference Manual for details Each discrete and generic process has an additional Delay parameter that can be a constant or a function The firing of a discrete process occurs abruptly with the delay associated with the process The same applies to gen
85. thways molecular interactions and protein post translational modifications In Cell Illustrator you can import models from CSO and BioPAX files into CI workspace During the import operations the models are converted to CI models and can be edited analyzed simulated saved as any other models Each model opened in the workspace can be exported to CSO ontology 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 74 of 88 9 Cell Illustrator Player Once a CI log file is created during a simulation it can be opened in CI Player application for replaying and detailed analysis if Cell Illustrator Player 5 0 File Edit Element Replay View Analyze Window Help PRQQRHRS E J ex Navigator m Lun T T cr zl Tur un mRNA fn TIM E Element Lists E Ica Mouse position 420 238 Fit the model in canvas Done 30 26 43 0 17 50 25 23am 2 508M if Figure 86 CI Player Workspace and Charts The main window of CI Player very much resembles the Workspace window in Cell Illustrator However instead of the Simulation menu and toolbar of Cell Illustrator CI Player has the Replay menu and toolbar with commands for moving forward and backward over the time series in the log The model state entity values and process status at a given simulation time point is visualized in the CI Player canvas window On the other hand charts
86. tion History show the Simulation History frame The frame contains the list of all simulation log files for the active model file see Section 6 3 2 e 4 Graph Layout show Graph Layout frame to set up elements on the canvas in the specified way e Path Search Results show the Path Search Results in a tabular view see Section 7 2 e Comments show the comments of the selected element in a tabular view 5 4 Element Settings Frame and Element Lists Frame The frames that show the details about the variables and equations in the model are the Element Lists see Figure 53 Figure 55 and Figure 57 and the Element Settings see Figure 54 Figure 56 and Figure 58 They can be opened either by selecting the items Element Settings and Element Lists respectively in Window Show Frame menu or by clicking on the buttons andil in the Property Frame toolbar The Element Lists frame consists of the following sheets the Entity Process and Connector Sheet as well as the Fact Fact Edge and Group Sheet Each sheet contains a table of all elements of the given type entity process connector fact vertex fact edge or groups respectively in the active canvas and their main properties A selection made in the Element List will be immediately reflected in the canvas and vice versa You can easily change a property of an element by double clicking on the corresponding cell in the table Note that the columns which can
87. to 1000 pt Note that the value of the Simulation Time parameter does not necessarily correspond to the real time it takes to execute the simulation on your machine The latter depends also on the complexity of the model additional simulation parameters such as the Sampling Interval and the hardware on which the Cell Illustrator runs The Sampling Interval specifies the interval at which simulation steps are executed Again the Sampling Interval is defined in Petri net time units At each step 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 60 of 88 model state changes are computed If the interval is 0 1pt as in Figure 76 the model changes its state at 0 1 pt 0 2 pt 0 3 pt 1000 pt If the Sampling Interval becomes smaller the simulation becomes more accurate but calculation performance decreases For example for Simulation Time 1 pt Cell Illustrator needs to perform 100000 simulation steps if the Sampling Interval is 0 00001 pt and only 10 steps if the interval is O 1 pt The minimum sampling interval allowable in Cell Illustrator is 0 000001 pt You should be aware however that simulation with the minimum sampling interval may take a very long time If the sampling interval becomes very small the simulation result will be similar to a differential equation model see Appendix B The Canvas Chart Update Interval defines how often canvas charts are updated du
88. tory connector 1s activated 111 the activity parameter of the process 1s true The default value of the activity parameter is true but one can write a script that modifies this parameter Once a process 1s enabled the biological process can occur Let us take a look at the simulation of a simple case A protein precursor is phosphorylated to be an active protein protein inactive auto phosphorylation protein active m c1 c2 10 0 2 0 m1 0 1 m2 1 0 12 Figure 73 The initial value of the protein inactive 1s 10 While the concentration of protein inactive is greater than the threshold of its outgoing connector cl which is 2 0 the connector is in the enabled state An enabled connector in this case executes the activation of the Phosphorylation The protein inactive will be converted to the active protein protein active at the speed of phosphorylation which is V t dm0O dt In another words the concentration of protein inactive will decrease by the amount m1 0 1 m2 time and the concentration of protein active will increase by the same amount Once the concentration of protein inactive reaches the threshold value 2 the connector will be deactivated Consequently the process of phosphorylation is terminated The end simulation state is displayed in Figure 74 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 58 of 88 protein inactive auta phosphaor
89. uw THE UNIVERSITY OF IOKYO Cell Illustrator M User Guide 2002 2010 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved TABLE OF CONTENTS TINTRODUG TION dices aekeka e etu d vow a Eng eV CE eyes ro eu ni hpc t E ee d up a E a uc trt 5 2 GETTING STARTED cis sxixekes ue Y euvess FaVEl e nei Ri x EPEErk Y EYE EUR Fe visae v Pune ekVa E VKE 6 ZV EG MI ure aai E TENET 6 2 2 Run Cell Illustrator Online CIO seeseeeeeeneen 6 2 3 Licensing INFOFMATION ccceeccceeeceseeeseeceeeceeeteeeeeaceteeeteueteneeaseeseeetseetaeeas T 2 4 GL Memory GUstOmIZAllOl Tuo omo Ra pte b EDU t ERR UP Ed ul Pos tita 8 3 CELL ILLUSTRATOR ELEMENTS siinid aaa aS 9 S 4 onceDEOVerviGW aadbesidtaitibaa us rebat A ma D mtd P id 9 SEP Mic REOR Rm 10 ONG GONMECCIONS ETT TEE TRE UTE 11 OX EIOPBSSOS c d ic se M dM LM 12 2 5 Element COMDIMNANONS usc osse e cuui medi imn tenero RM DU Eu dien arte DR ELEME LE 13 3 6 Discrete Elements Example Tenedor a i 15 3 7 Discrete Elements Example 2 Connector Speed 16 3 8 Continuous Elements Example 1 ccccccccceccseeeeeeeeeeeeeceeeeeeeeeeaeeneeees 17 3 9 Continuous Elements Example 2 cccccccecccececeeeeeeeeeeeeeeeeeeeeeeeeaeeneeees 19 4 BIOLOGICAL PROCESS MODELING WITH CELL ILLUSTRATOR 21 2 Degradarea 21 4 2 ransiocdllOliusueisstudetaidunciekos bedticsntac tcsetdatet ideato
90. versity of Tokyo All rights reserved Page 51 of 88 Cell Illustrator 5 0 Professional demo intra csmLorg File Edit Element Simulation View Analyze Window Help OCO oms Li xt 4G R QQ THe BR ie s 2 x 4 Part csml amp 3t D cal Lieli aviar o reir MRI ERARA t3 t a hit amp Lf Project Mona a E rs gt T Gene Mining Mouse position 20 18 Select all elements Done Selection 12 30 31 A76Mz 508M Cell Illustrator 5 0 Professional demo intra csmLorg File Edit Element Simulation View Analyze Window Help OOGnmmllisf amp nasacZvp ELIP RE Parts Library Parts Library 44 cellComponent_images_with_volun EE H cellComponent images without v 99 part D csml 3 DELL a TEA fh eee yn ERATA gt Insert PD E rastro F Premana IE E lr gt gh Gene Mining Mouse position 358 3 Select all elements Done Selection 12 8 16 0 12 31 44 ezMz s08M m e US ERN Figure 64 Parts Library usage 5 5 5 BioPACS Frame BioPACS is a remote repository of metabolic pathway maps which were converted from the KEGG Web site to the XML file format used by CI 9 12 To connect to BioPACS open the BioPACS frame by clicking on the icon in the Library toolbar Select a KEGG map name and click on the Open button to open it in Cell Illustrator as a new canvas 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All
91. versity of Tokyo All rights reserved Page 7 of 88 3 Cell Illustrator elements CI 1s designed to model biological pathways using three types of abstractions entities processes and connectors 3 1 Concept Overview The modeling and simulation engine of Cell Illustrator is based on an extension of the Petri net methodology A Petri net models a system of conditions and instantaneous events their dependencies and mutual synchronization The Hybrid Petri net with extensions HFPNe employed in the Cell Illustrator adds the notion of continuous and generic processes and quantities which are essential to the description of biological systems 10 The following table presents a summary of the Petri net model elements their explanations and equivalent biological meanings Table 1 HFPNe elements Hybrid Functional Petri Net Elements Petri Net Component Symbol Biological Equivalent Discrete Place A discrete place holds tokens represented as a non negative integer Tokens A token is a unit held in a place Discrete Transition A discrete transition 1s a discrete process that consumes tokens from their input places and produces tokens into their output places The input and output places are connected to a transition by arcs A transition fires and the process is invoked if all the conditions for firing are met Every transition has a delay time for firing Continuous Place A continuous place holds a non negative rea
92. y that the number of steps will be less than 100 000 steps Logging hundreds of thousands of steps will cause the CIL file to be very large Since CIL stores the model in the same format CSML as the original project file you can open a log file with Cell Illustrator and edit the model However saving the edited model into the original log file might erase the time series and any other simulation history information Therefore you should use the Save As command of Cell Illustrator if you would like to recover the model from a log file for further editing O 2002 20 Human Genome Center Institute of Medical Science The University of Tokyo All rights reserved Page 63 of 88 6 2 4 Simulation Charts You can visualize the simulation results on time graphs The charts can display time series of entity values process connector speed of the selected elements Charts can be defined in the Chart Settings frame Figure 77 Click on the button in the Dialog toolbar to bring up the Chart Settings frame A table at the left side of the frame displays all entities in the active canvas The Chart Tree at the right side displays all defined charts for the model in the active canvas Select one or more elements on the canvas right click on tree root element and select Create New Chart item to create a chart that tracks the element value over simulation time Alternatively right click the selected elements on the canvas and choose Create Chart comman
93. ylatian pratein active m c EZ ut E Oo 5 2 m1 D 1 m2 F Figure 74 Figure 75 presents a simulation chart for this model Chart 1 Marne pratein inackive auta phosphorylatian pratein inactive protein active auto phosphoarlization Figure 75 When simulating a complex pathway you will often need to construct multiple entities with multiple connectors leading to one or more processes In these cases only when all process and associate input connectors are activated and all inhibitory input connectors are not activated the process 1s enabled For each simulation step all the activation conditions for all processes are checked and all activated processes are executed in a random order The simulation run can be logged and saved into CSV comma separated values files or CI log files CILs refer to Simulation Log Saving log files gives you the following possibilities e View and analyze the log using a standard spreadsheet application e View and replay a simulation step by step in CI Player application refer to Simulation Replay e Order and systematically analyze the results for a set of similar simulations refer to Simulation History Creating a log file and replaying it in CI Player 1s recommended in the following Cases e When the simulation may take a large amount of time e g fora complex model and accurate simulations e When you run a set

Download Pdf Manuals

image

Related Search

Related Contents

Norge 1200 GT 8V - Utilisation / Uso - 09/2010  Ram Models - SonnyCovic.com  Operating instructions  CCAM ACP : MODE D`EMPLOI Projet  SMS Smart Media Solutions FUNC Flatscreen WM ST A/S  PDF - Honda ATV  Screen Architect Ltd  SpikeMark Manual v2_1  1300 S & SM MANUEL DE L`OPERATEUR WD66-1300 S  Samsung LN26C350D1D  

Copyright © All rights reserved.
Failed to retrieve file