Home

CoCo Administration and User Manual

1. cd lt COCO_HOME gt ant edit build_ant properties and position required properties especially adapt db pwd to use for cocouser and db url to reflect the MySQL server name Manually create the directory that you specified in the property coco data home in build_ant properties file and make sure that the user running tomcat has full access rights on it rxw ant configure Install R libs in a shell by typing you might have to log in as a user having permissions to remove install R libs gt gunzip untar configure make make install for gd 2 0 33 tar gz if not already done gt R CMD INSTALL lib R GDD_0 1 4 tar gz if not already done gt R CMD REMOVE gff3Plotter only if not the first installation gt R CMD INSTALL lib R gff3Plotter lt version gt tar gz VVVV In a shell type you should still be in the lt coco_HoME gt ant directory gt ant install example gt ant deploy gt chmod 755 lt TOMCAT HOME gt webapps coco WEB INF 1lib gt ant update genomes might take few minutes Point your browser to http localhost tomcat_port coco and login as admin admin Should you fail at any point re start from step 7 or before and ignore step 8 if it was successful 3 Detailed Installation Steps This section explains how to install CoCo step by step 3 1 Download CoCo Download the coco tar gz archive e g from furlonglab embl de and put it in a freshly made e g coco directory 3 2
2. 2 Add a feature map or clone map property by setting a new tiling chip clone map lt chipname gt which value is the absolute path to the features genomic position file This file is a 4 columns tab delimited file holding the feature_id the chr the start and stop position of the chip feature Note that ChIP chip result files must use these ids Take a look at demo data for an example 3 Add a genome lt chipname gt and genome version lt chipname gt property for the microarray which values must refer to taxid and genome version present in the configuration file 7 3 Defining LinkOut in coco properties LinkOut allows you to enrich CoCo interface by defining links to other web resources LinkOut are build using gene symbol values Just take a look at demo data to see how to define a LinkOut Note Defining LinkOut is optional 7 4 Defining InSitu default file s in coco properties To our experience preparing in situ files might reveal to be quite time consuming Hence CoCo let you define default in situ result files one per organism that will be available to all users To define such a default file simply add a insitu default file lt taxid gt property The value should be the absolute path to a tab delimited file holding in situ results Format is a simple as Column 1 GeneSymbol Column 2 dev_stage Column 3 anatomy You certainly wonder what terms to use in columns 2 and 3 Well it is up to you The most important
3. is available and features having a value greater or equal to this cut off will be colored in red e if you did provide mock results additional cut off and conditions can be set Three Expression profiling parameters must be set if expression profiling datasets are the mock cut off must be set and is used to remove enriched features from the result set when their associated mock is greater or equal to this mock cut off you can define an additional condition that must be reached to define a feature as enriched this condition uses either the ratio or the difference as you set it and a cut off included in the configuration e Minimun and Maximum values between min and max the color used will follow a color ramp from blue min value to yellow max value Below the min value and above the max value color won t change anymore and remain yellow or blue e lt A cut off this name might be badly chosen as this cut off represents the center of your min max couple i e the point where the color will be grey indicating that the gene is neither up or down regulated in the experiment If the values found in your expression datasets are log ratios this cut off must be set to 0 Once positioned click Save to store them in the database so that they remain after you log out This allows CoCo to load them the next time you log in Using the Show all options Show main experiment only options you can show
4. Name of this configuration 2 Submit Reset The Configuration form in details This form has four main sections 1 Experimental Parameters 2 ChIP on chip Results 3 In Situ Results 4 Expression profiling Results Fill in the four section give your configuration a meaningful name and click Submit CoCo will then parse and validate all your files It will convert ids found in Expro and In Situ files to internal ids and finally pass all these data to R format communication is GFF which will then index them This process is time consuming so please be patient it should last for a couple of minutes Remark the form usually offers to either select a file stored on the server or upload a new one Once you uploaded a file it becomes available to you and other users if you decide to share it To keep interface happy thing about DO NOT HAVE SPACES OR characters like amp lt gt N e Using meaningful file names as file names are displayed in selection box e Using short file names as file names are displayed in selection box Try to not exceed 20 characters e Deleting unused old files e It might happen that an error occur while loading a configuration mistake in files before loading completes In such situations CoCo tries to delete uploaded files but it sometimes happen that CoCo can t do it After a failure always check if your files have been saved on the server or not using the Uploaded Files m
5. Once Tomcat deployed CoCo execute gt chmod 755 lt TOMCAT HOME gt webapps coco WEB INF 1lib Important Before logging in you must create a user properties file and place it in the directory you indicated in build_ant properties file property coco data home This has been done for you in you choose to install demo data 3 10 CoCo Management and Maintenance Common installation and management tasks are Define Genome Annotation s e Define Tiling array s Define LinkOut Define InSitu default file s e User list The first tasks are realized by adding information in the coco properties while adding users in done in user properties Important considerations about coco properties The lt coco_HOME gt conf coco properties file has been generated when you executed ant configure Running ant configure again will certainly replace this lt COCO_HOME gt conf coco properties To get a coco properties from scratch use ant reconfigure Hence you might want to reflect the changes you make in lt COCO_HOME gt conf coco properties in lt COCO_HOME gt template coco properties or edit lt coCO_HOME gt template coco properties directly and launch ant reconfigure after modifications Summary 1 When editing coco properties you always have the choice between e Editing coco properties in lt COCO_HOME gt template followed by ant reconfigure gt recommended at installation time e Editing
6. a user properties file needed for login is available and can be used later as a guide to add new users To install demo data simply run the ant task install example 1 e gt cd lt COCO_HOME gt ant gt ant install example Note the coco properties file in lt coco_HOME gt conf has now been replaced with one containing properties to use demo data We suggest you to use this file as a starting point when adding your own data in latter steps Once done you can login using the username pwd combination admin admin and try to create your first configuration using demo data available in lt COCO_HOME gt demo chip chip and lt coco_HOME gt demo expro These directories contain ChIP chip results obtained for Drosophila melanogaster Mef2 transcription factor and time series gene expression data using an over expression of Mef2 respectively These data are a subset of published results Sandmann et al Dev Cell June 10 6 797 807 3 9 Build the coco war and deploying CoCo gt cd lt COCO_HOME gt ant gt ant buildwar OR if you deploy CoCo locally gt ant deploy A file named coco war should now be present in lt coco_HOME gt dist If you haven t executed ant deploy you can copy this file in lt TOMCAT_HOME gt webapps dir wait a bit to let Tomcat the time to deploy coco and point your browser to http localhost coco assuming tomcat listens the port 80 and you should see the CoCo login page
7. for R to load all data integrated in the configuration the time for R to generate the picture the time for R to shutdown We are thinking about using a solution like Rserve to cut down the picture generation time This should be a first step to proper Affymetrix dataset handling But still loading datasets of 6M point each will require a pretty big server for picture generation several Go of RAM 9 2 Other considerations Besides technical aspects we are not sure that displaying all features of an e g an Affymetrix chip is a very likely use case CoCo has been initially designed to cope with small medium size tiling arrays These arrays are usually made with clones ranging from 500b to several Kb The pictures generated by CoCo accommodate perfectly with these sizes of feature you can zoom out and display genomic regions of several Mb and still see tiling array features With oligo based tiling arrays i e from 25 Affymetrix to 60 NimbleGen bases zooming out becomes quickly useless Indeed we observe that users like to display from 30 to 100 Kb around enriched features to have a good overview of the surrounding genomic environment Unfortunately oligo features of 60 bases become invisible when displaying more than 30 Kb The examples below show views obtained with a NimbleGen chip 380K features In addition the common approach when analyzing results from high density arrays is to run a region discovery algorithm like MAT http c
8. http www hibernate org for database access and management You should then be able to easily deploy CoCo on other RDBMS Note about Tomcat JVM Memory CoCo requires quite some memory in specific situations This might lead to OutOfMemoryError To prevent this edit the startup sh script in lt TOMCAT_HOME gt bin and add the line just before last line export CATALINA_OPTS Xmx750m 2 Installation Overview For the impatient here is the complete list of instructions to get CoCo quickly installed and configured Important Tomcat MySQL and R including required packages must already be already installed Installing R libraries you certainly need to be root untar conf make make install for gd 2 0 33 tar gz R CMD INSTALL GDD_0 1 4 tar gz Use Bioconductor installation procedure to install geneplotter biobase annotate package You can also go for the easy solution and install the whole Bioconductor Please see instructions at http www bioconductor org docs install howto html Installation steps 1 om 10 download coco tar gz and put it in a freshly made e g coco directory tar xzvf coco tar gz gt creates a coco directory that we ll refer to in the rest of this document as lt COCO_HOME gt Create coco db in MySQL log in as root and do CREATE DATABASE coco use coco lt COCO_HOME gt src sql coco sql GRANT ALL ON coco TO cocouser lt servername gt IDENTIFIED BY lt a_password gt
9. 1 LD Or Upload New Exp Name A Mock select 3 pe oe Mock File Choose File no file selected Exp select m Exp File Choose File no file selected MEE Or Upload New Exp Name Choose File Mock select 3 dica RA Mock File Choose File no file selected Additional Results 2 Exp select m Enp File Choose File no file selected Additional Results 3 Or Upload New Exp Name Pe A Mock select z i taid af am Mock File Choose File no file selected Exp select g 7 Exp File Choose File no file selected Additional Results 4 Or Upload New Exp Name 4 OE ook ele 7 pidad New ies Mock File Choose File no file selected Sticky genomic 4 select B Or Upload New File Choose File no file selected fragment list EE A A In Situ Results Upload additional in a select s Or Upload New File Choose File no file selected situ data Ignore default in situ data O used by default Expression Profiling Results Exp Name Expro Results select B Or Upload New le Choose File no file selected Exp Name ak Expro Results 2 select g Or Upload New EE File Choose File no file selected Exp Name SEE Expro Results 3 select 5 Or Upload New File Choose File no file selected eT Exp Name EE Expro Results 4 select Or Upload New le Choose File no file selected Exp Name p Expro Results 5 select 2 Or Upload New File Choose File no file selected
10. All gt New Region gt Import gt Help gt Admin Tools gt Log out 6 2 Importing regions into CoCo You can import regulatory regions in bunch To do this you must first assemble a tab delimited file see format below and then choose Import in the Regulatory Region menu The picture below depicts the import page In addition to the file containing region definitions you have to provide 1 The organism 2 The genome version the genome version the region coordinates found in the file refer to 3 A Data Origin Name all regions will be attached to this origin This is quite important as origins can be used to filter regions later on Here you should indicate where you got these regions from e g Flyreg 4 The group world rights if you don t share these regions other people won t see them If you come with predictions you certainly don t want to share them Alternatively if regions origin is literature we encourage you to share them 5 Ignore Ambiguous Gene option can be use to ignore lines for which gene symbol relative to either transcription factor or target genes can t be uniquely resolved i e the symbol either has no match in CoCo db or maps to multiple genes We advise to first NOT use this option This will cause CoCo to either successfully upload all regions or print a report about ambiguous genes In this latter case no regions will be uploaded and you can either decide to correct your file
11. EX above cutoff W below cutoff In Situ data stages only E anatomy only E stages and anatomy E other Bit Gene Search nau Search B Search Chip Feature by ID Search m Or Go To Chr 3R 15 From 19391307 To 19691306 Go m Navigation Control 5 5 5 r Show 299 Kbp Global Picture Display Options Show all options Save Options as 58 Main ChIP options Mef2_ChIP_6 8 and Mef2_ChIP_6 8MOCK ChIP cut off gt 0 7 mock cut off lt 0 3 and no transfo os N BAMb B45Mb B 5Mb B 55Mb BEMb B 6 5Mb LA 001 003 0 o Bye H A HHH z o ma 003 009 o HE oe 6 7 mma HHH o 905 o 901 905 Ed AI o D HHH caoz 610367 coomo 5 IEH H or Amp pcs A H H H z a H H a s Pi H z L cemas com cos EH PHE HH cia on ov amo 90 EE se HH OM OR SODA pa oos oe osar EN BE A o vomar c gt 002 046 ay EBr HH DI HHH 3 2 2 Positioning picture parameters Cut offs used for color coding are controlled in the Global Picture Display Options panel Here you can position cut offs for ChIP chip and expro experiments Each ChIP chip experiment has its own settings while expression profiling cut offs apply to all expression profiling experiments ChIP chip parameters e if you did not provide mock results a unique cut off ChIP cut off
12. Regions Conditions Anatomy Stage s gt Search gt New Region Comments on the region gt Import gt Help o PAR Confidence given to this region tentative DI gt Admin Tools gt Log out eS EE sd ne Se EE m e Symbol or Acc Regulation Type unknown Confidence given to this region tentative Comments on region target gene association Save Reset 6 4 Creating a region and adding experimental evidences while browsing experimental data While browsing your experiments you can click on chip feature This will open up an option window that allows you to perform well different actions One of them is named Create a new regulatory region based on this fragment as shown on the picture below You have clicked on chip feature D599 6 e3 2R 5449064 5452214 in the experiment named mef2_1P_6 8 Please select the action you want to perform e View Fragments description details e Create a new Regulatory Region based on this fragment e Add a Experimental Binding Evidence to an Existing Regulatory Region based on the ChIP chip result of feature D599 6 e3 Select Regulatory Region select Next Close Window Please contact charles girardot for feature requests comments or reporting bugs If you follow this option the regulatory region definition page will show up This page see picture below is similar to the regulatory region creation accept that 1 the
13. chip feature ID specify a genomic location follow links provided in the result table or click on chromosome overview pictures The genome browser is the place where all data are displayed together and certainly one of the main features of CoCo As the user browses or zoom in out CoCo assembles genomic region views representing the requested chromosomal region Down the Navigation Control panel clicking the View All Regulatory Regions Overlapping With will open up a new window displaying all regions found in the picture 3 2 1 Picture Description Genomic region view pictures are organized in three main zones see picture below In the central part ChIP on chip zone each tiling array feature is represented as a rectangle colored in red or grey depending on whether its enrichment value is above or under the user defined threshold thresholds can be set for each dataset individually A third color black is used for features defined as sticky A rectangle is draw for each dataset microarray features resulting in the stacks as shown on the picture To cope with time series datasets order within stacks follows order given at configuration creation The plus and minus genomic strands together with genes including exon intron boundaries are represented above and below the ChIP on chip zone respectively Genes are colored according to available in situ patterns and four colors are used to reflect whether genes are expressed at
14. column file listing microarray features one per line found to be sticky Features marked as sticky will be masked on pictures We initially introduced this sticky notions based on two different observations 1 some array features always show up as enriched and 2 some features are not clearly mapped to the genome especially after a new genome assembly has been made available In both situations we want to flag them clearly and ignore them in result list After using CoCo we actually found that this sticky file shows to be useful in other situations e g you want to mask lots of features to easily trace the behavior of very interesting regions over different datasets Important The order in which you specify datasets is kept in CoCo and in displays The picture below shows an example where we mixed experiments already uploaded on the server with new experiments new file upload We also upload a sticky files Note that e you must not give an experiment name for experiments that are already on the server you must give an experiment name for new experiments e each track might take a mock dataset or not me Exp File Mef2 6 8 Chip Exp Name 7 Choose File no file selected Or Upload New Mock File Choose File no file selected Exp File Additional Zxp select Or Upload New Exp Name Choose File Mef2_8_10_ChIP_all txt Results 1 TE prea New Tye IP 8 10 Mock File Choose File no fi
15. e esse ese ese see see se ee ee Ge ge ee ee GR Ge ee ee Re Ge ee ee ke 16 4 MANAGING YOUR FILES iets ese ee eN ee Ge iii 17 5 MANAGING YOUR CONFIGURATIONS ee esse ee ssee esse esse Ge Ge EG EG EG Ge GE EG EG Ge Ge Ge 18 6 DEFINING REGULATORY REGIONS AND TARGET GENES use seeseeseeseese see see see sesse see see sesse see see see se es 19 6 1 REGULATORY REGION MODEL IN COCO ccccsseesscesseessesseesseeescesseeeseenseesseenseenseesseesseenseesseeeseensees 19 6 2 IMPORTING REGIONS INTO COCO datea ei ER r R EE ATEA EA ERA RANER 21 6 3 CREATING A REGION FROM SCRATCH iese see se ese ee ese ee ese ee ese ene Ge ee ee Gee GR ee ee ee Re ea ee ee Re ek ee ee ee 24 6 4 CREATING A REGION AND ADDING EXPERIMENTAL EVIDENCES WHILE BROWSING EXPERIMENTAL DATA vise sees see epos ige ses ede ses sg ige ee sege gee es eed gee ge pie ese gek eed eg ek oe ee ge seges eg eg ode ee gek Ese es ee ae ge beg ee eg 25 6 5 ASSIGNING TARGET GENES TO REGIONS WHILE BROWSING EXPERIMENTAL DATA esse se sees 26 6 6 SEARCHING FOR REGULATORY REGIONS ccssccsscssseesseesceeseesseesseeeseesseeeseenseesseeeseeeseesseesseesseeesees 27 SANA AA N NE EE EE N AN s 28 S FILE De ad Ed OE Ee ee ee ee ee ek Ie 28 Sal CHIP GHIP DATA FIL SO a OE OE Lge hs 28 8 2 INSITUDATA HILE naii a ii lalala 29 8 3 EXPRESSION PROFILING DATA FILE esse see se ese ee see ee ese ee ese ene ee ee ee ene en ese ee ee eke ea ee ee Re ek ee ee ee 29 SA FERM LIST ELE SG SS GE GR DE GE GE De N e
16. form is pre filled with region boundaries corresponding to the chip feature boundaries 2 the form is pre filled with the transcription factor accession concerned by the current configuration 3 the form exhibits an experimental evidence The cryptic text can t be modified By default the evidence will be added to the binding event that is created at the same time the region is created unless you select the checkbox in front 4 The region origin can t be modified and is set to Experimental Define a new Regulatory Region n Region Organism Drosophila melanogaster l Genome Version Region boundaries Chr 3R HA From 14568604 To 14572041 Transcription Factor CG1429 Mef2 Binding Event Observed Spatio temporal Binding gt ses Anatomy Stage s Conditions ge C Ignore the following ChIP chip Evidence not editable ExpN mef2_ChiP_4 6 ID D726_2_c5 Val Exp 0 1306 Comments on the region Confidence given to this region Target Gene Symbol or Et Te Tay Confidence given to Acc Number Regulation Type ERA this region tentative tentative Comments on region target gene association gt Save Reset Close Window Another option offered in the option window is Add an experimental evidence and let you attach the selected feature s experimental results as an evidence supporting the binding of a TF onto an existing region Indeed this option wil
17. select the one they need On the example picture below we create a configuration for a ChIP chip time series of the Drosophila melanogaster transcription factor mef2 The microarray used for the ChIP chip experiments is genomic_tiling_Dm_version2 The spatio temporal focus is set to stage4 to stagel7 and we provide a anatomical term list fly_muscle_anatomical_terms txt by selecting a Term List File already present on the server Note the stage4 17 short cut used to specify a range When Dev Stage field is filled CoCo tries to guess if you mean a single developmental stage or a range by analyzing your input If your input look like lt word gt lt number1 gt lt number2 gt CoCo understands that you meant all combinations of lt word gt lt number gt where lt number gt increments from lt number1 gt to lt number2 gt For stage 4 17 it translates into the list stage4 stage5 Stagel6 stage17 This feature is only available for the Dev Stage field and no spaces can be used i e embryonic stagel 17 won t be understand as a stage by CoCo In such a situation simply create a file listing all your terms i e embryonic stagel embryonic stage2 embryonic stage17 one term per line Experimental Parameters Organism Drosophila melanogaster Array genomic tiling Dm version2 Transcription Factor used in the ChIP mef2 Dev Stage Or Chooze gt Or Upload File E
18. selected Or Upload New Exp Name File Expro Results 5 select EET pro Results Choose Fi no file selected 2 5 Naming examples As you might have noticed we guite insist on names you should give to your files experiments This is because these are used in displays and long names will badly display especially in the genome browser Please try to keep experiment names less than 15 letters best is 10 3 Browsing data Once you created a configuration you can start working There are two main visualization pages 1 The Overview page 2 The Genome Browser 3 1 Overview Page The overview page see picture below depicts each chromosome and a result table summarizes all ChIP on chip results Note that the table keeps the order in which the ChIP chip datasets have been submitted Thanks to chromosome pictures users can gain understanding about how enriched regions spread across the genome and thus identify clusters of enriched regions At the top left of the page in Genome Overview options you can set cut offs used to define enriched regions in the chromosome pictures and compute pictures again At the bottom of the page a result summary Result List is presented in a result table Here again we can modify cut offs are re order the table you can order by score or genomic location You can also ask to display to n closest genes in an extra column This feature might take some time and we recommend using it on
19. CoCo Administration and User Manual COCO ADMINISTRATION AND USER MANUAL occccococconococnonocconoccononcnonocconocconocoronocconocconocoss 1 INTRODUCTION A O Ee 4 USER AAA A E RT TO 5 INTRODUCTION Me es ee ee ee Ge OE ae Ee GE EA GE 5 1 LOGIN IN MENU OVERVIEW sesse esse esse esse esse ese ese es ge see Ge Ge Ge Ge Ge Ge Ge Ge Ge EG Ge Ge Ge EG EG EG 5 1 MENU DESCRIPTION vs idiote dese dees ese ds ge ee ee gee dei de oe iio 6 2 CREATING A CONFIGURATION sscscsssssssssesssesssssesssssssssssesssesssesssessssssscssssssscsssesssesssessssssssssssassessces 6 2 1 EXPERIMENTAL PARAMETERS SECTION esse sesse see se ese ee ese ene ee ee ee ene ee ese ee ene ene ee ee ee ene ee ee ee ese ee 8 2 2 CHIP ON CHIP RESULTS SECTION sera esel Een Ge oe eek Eer dia 9 2 5INSTEURESULTS SECTION A ei enn ida iese ee gebere ioon Gesig eke see se gee ee Ee 11 2 4 EXPRESSION PROFILING RESULTS SECTION sesse sesse ese ese ene ee ese ee ene ene ee ee ee ene ee ee ee ee eke ea ee ee ee 11 Z NAMING EXAMPLES ciao anal 12 3 BROWSING DATA SEE ED OE ED ER EE GE E E 12 BLOMERVIEWPAGE SE ee Se oe eed ee Ge ee de ee de Ge ed N ee ee ee ee ee ed Se de ee ee eg Ds 12 3 2 GENOME BROWSER rnini aee RS EE OG Ke DAE bee ERGE Re DER ENE ee Ee Re Ee oen ee Re Ne DERE Be Ee ER Ne SR Beek De 13 3 2 VPictureDEseri A E NENA 14 3 2 2 Positioning picture parameters sesse se ee ee ee ee AR ee AA Re ee AR Re ee ee Re ee ee Re ee 15 3 2 3 Interactivity saving conclusions as YOU browse
20. Uncompress the archive In a shell uncompress coco tar gz e g gt tar xzvf coco tar gz This creates a coco directory that we ll refer to in the rest of this document as lt COCO_HOME gt lt COCO_HOME gt contains following directories e src where all sources are stored JAVA Perl R SQL ant contains ant build file and build_ant properties that allows you to perform a wide range of tasks from building to managing CoCo e doc documentation material e web contains JSPs and web application definition XML files e conf contains CoCo property files e demo some demo data files e logs empty dir used to create log files e lib all needed JAVA and R libraries template configuration file templates used by the ant re configure task to generate installation specific property files 3 3 Create the CoCo SQL Database Create a coco database in MySQL or another RDBMS and a cocouser with all rights on coco database Note CoCo uses Hibernate http www hibernate org for database access and management You should then be able to easily deploy CoCo on other RDBMS Note 2 To date CoCo has been used on MySQL 4 only but nothing should speak against using any InnoDb MySQL version Practically create a database coco in mysql and a cocouser with all rights on this coco db Go to lt cOCO_HOME gt src sql and execute the coco sal to
21. additional properties lt property name hibernate connection driver_class gt com mysql jdbc Driver lt propert y gt lt dialect for MySQL gt lt property name dialect gt net sf hibernate dialect MySQLDialect lt property gt lt property name hibernate show_sql gt false lt property gt lt property name hibernate use outer _join gt true lt property gt 3 7 Configuring CoCo CoCo is configured and maintained using a simple property file named coco properties and located in lt coco_HOME gt conf This file has been generated when you ran the ant configure task in Set up project The file generated is a minimum file that needs to be filled with correct values reflecting your organisms chips If this is the first time you install CoCo we recommend you to jump directly to Installing demo data and test your installation before configuring CoCo with your own data Alternatively please refer to CoCo Management and Maintenance to learn how to fill in properties The coco properties file contains help describing properties and explaining how you should add new ones 3 8 Installing demo data CoCo comes with a set of example files that let you run CoCo with further configuration and then check that CoCo is fully functional before configuring your final CoCo server Example files contains all files necessary to work with the Drosophila melanogaster genome genome annotation release 4 0 In addition
22. as defined by TileMap or MAT 2 remove all features having a enrichment ratio less than e g 0 3 3 You can rescan the clone map file and keep only those features that appear in any of your result files 4 If you want to have a nice display make sure that all your chip chip datasets after filtering contain the same set of features by adding back features found in one dataset but not in others 5 Contact GirardotWembl de I should have perl scripts you can adapt that do steps 2 and 4 CoCo installation and maintenance document CoCo is a JAVA web application using R http lib stat cmu edu R CRAN for statistics and image generation Data are stored in a MySQL www mysql com InnoDb database 1 Pre requisites CoCo can be downloaded from http furlonglab embl de methods tools coco CoCo needs the following third party products to be installed prior to CoCo installation All these third party products are well established e JAVA gt 1 4 2 available e Ant available gt 1 5 Tomcat installed tested with 5 0 version e R version 2 1 0 the version is important as an essential library we use doesn t work in newer versions We hope to fix this soon e Additional R packages a gd 2 0 33 tar gz b GDD_0 1 4 tar gz e Bioconductor package geneplotter_1 5 4 tar gz installed in R e MySQL installed should work with all version supporting InnoDb table engine tested with 4 0 versions Note CoCo uses Hibernate
23. ase contact charles girardot for feature requests comments or reporting bugs 4 Managing your files Once files have been uploaded to the server as you create configurations they are stored and made available in drop down selection boxes You can see your files using the Uploaded Files menu There you can delete or share your files with colleagues Three sharing levels are available no sharing group sharing and world sharing world meaning here people that have login In the list of files you see both yours and the one you have access to Deleting a file can be done only if this file is not used by existing configurations This limitation is made because it is sometimes e g at software upgrade time necessary to recomputed configurations F CoCo Server Analyze your ChIP On Chip data Online 20 Y Online users 1 admin Uploaded Files File name Owner Exp name Type Group rights World rights Extra gt Start Analysis C Mef2 10 12 ChIP_all txt File ID 7 admin Mef2_ChIP_10 12 CHIP_ON_CHIP_EXP_TYPE nogroup Delete gt Configurations O Mef2_10_12_mock_all txt File ID 6 admin Mef2_ChIP_10 12MOCK CHIP ON CHIP EXP TYPE nogroup Delete gt Uploaded Files C Mef2_6_8_ChIP_all txt File ID 1 admin Mef2_ChIP_6 8 CHIP ON CHIP EXP TYPE nogroup Delete EE Mef2_6_8_mock_all txt File ID 2 admin Mef2_ChIP_6 85MOCK CHIP ON CHIP EXP TYPE nogroup Delete a Region O Mef2_8_10_ChIP_all txt File ID 3 a
24. ate field Don t position other fields and they won t be used to filter Filtering options e Genomic location only regions that are fully included in the specified boundaries will be shown e Region Origin when used only regions from selected origins are displayed Proposed origins are those for which you have read rights on at least one region Note that the Experimental origin is always proposed e Region Confidence when used only regions with selected confidence are displayed e Only display my regions when checked only regions that you created owned by you are displayed e Transcription Factor Filter here you can select transcription factor and only display regions bound by these Note that all TFs proposed in the list are TFs that bind at least one region defined in CoCo Unfortunately you might not have the read right on these regions Target Gene Filter here you can define a gene list and display only regions that have one of these genes as target gene Note that all proposed genes in the list are target genes defined in CoCo Unfortunately you might not have the read right on these regions CoCo Server Analyze your ChIP On Chip data Online 9 U UY Online users admin P r P P P Region Filtering Criteria only filled fields are used to filter regions Organism Drosophila melanogaster Genome Version 4 Q Only display my regions gt Start Analysis Location Filter Chr 4 75 From T
25. bdominal intrasegmental apodeme l abdominal intrasegme 2 abdominal lateral transverse muscle abdominal 1 lateral transverse muscle 2 abdominal 1 lat abdominal 1 lateral transverse muscle 4 abdominal ventral acute muscle abdominal 1 ventral acute muscle 2 abdominal 1 ventral acute muscle 3 abdominal ventral acutemuscle 2 abdominal 1 ventral longitudinal muscle abdominal 1 7 dorsal transverse muscle abdominal 1 7 lateral oblique muscle abdominal 1 7 lateral transverse muscle abdominal 1 7 lateral transverse muscle 2 abdominal 1 7 lateral transverse muscle 3 abdominal 1 7 lateral transverse muscle 4 abdominal 1 7 ventral acute muscle 2 abdominal 1 7 ventral longitudinal muscle 4 abdominal 1 7 ventral oblique muscle 3 abdominal 1 7 ventral oblique muscle 4 abdominal 1 7 ventral oblique muscle 5 abdominal 2 7 lateral transverse muscle 1 abdominal 2 7 lateral transverse muscle 2 abdominal 2 7 lateral transverse muscle 3 abdominal 2 7 dorsal transverse muscle 6 Defining Regulatory Regions And Target Genes The primary goal of CoCo is to help users in finding regulatory regions and defining their target genes Regulatory region boundaries definition and target gene assignment represent very valuable knowledge As we ll explain CoCo provides different means to realize these tasks and provides a detailed model to store them To fully benefit from these data even years after decision has been made it is important to
26. coco properties directly on the server the file is located in lt TOMCAT_HOME gt webapp coco WEB INF classes org embl coco and manually reflecting changes locally gt recommended once your server is already in production 1 e you don t want to restart Tomcat 2 After modifying coco properties do one of the following At installation time after edition of coco properties in COCO HOME template do cd COCO HOME ant ant re configure ant uploadgenomes all ant deploy VVVV At maintenance time after edition of coco properties on the server do login as admin go to the Admin Tools menu only available when logged in with the admin role Click on the Reload application properties link Done VV VV 7 1 Defining new genomes in coco properties Defining new genomes in CoCo is certainly the most tedious task in CoCo but still very easy For this you need to 1 Add the taxid if not yet listed of your organism in the genome supported taxids property this property accepts a comma separated list of NCBI taxids Add the genome annotation version of the genome annotations in the corresponding supported genome versions lt taxid gt this taxid specific property accepts a comma separated list of genome annotation versions List if not done already the chromosome names of this organism in the taxid specific chrname list lt taxid gt this taxid specific property accepts a comma separated l
27. comelntro jsp i ora QA PC Boulini7 Furlong Lab COCO FunGen Localy EMBLw FlyAnnotationProjecty BASEw Informatiquey Persov bioinformatiquey gt F CoCo Server Analyze your ChIP On Chip data XQ a Online Y Online users 1 admin I want to use the following existing configuration Mef2Test ok gt Start Analysis gt Configurations gt Uploaded Files gt Regulatory Regions gt List All gt New Region gt Import I want to create a new configuration gt Help gt Admin Tools gt Log out Please contact charles girardot for feature requests comments or reporting bugs Al 1 1 Menu Description The left menu is always available Depending on whether you have started an analysis or not different sub menus are made available Main menu items are Start Analysis the first page you see when you log in This page allows you to either select an existing configuration to work with or create a new one Configurations this menu lets you view and manage your configuration s Once a configuration has been selected sub menus will appear Uploaded Files this menu lets you view and manage your files s This is where you should go to delete or share your files Regulatory Regions this menu lets you view manage and import regulatory regions Help provide some help like a link to download this document Admin Tools available only if user has the role admin Allow to reload properti
28. create tables in coco db In MySQL these tasks can be performed using following commands gt mysql u root p lt ROOT_PWD gt you should now be logged in mysql as root or any user with sufficient privileges then execute gt CREATE DATABASE coco gt use coco gt N lt COCO_HOME gt src sql coco sql gt GRANT ALL ON coco TO cocouser lt servername gt IDENTIFIED BY coco gt exit Note if your MySQL doesn t have any server name simply use localhost Make sure your settings work by trying to login as cocouser using the mysql console tool and use coco database i e mysql u cocouser pcoco h lt servername gt use coco show tables select count from genes VVVV If login succeed you can go on 3 4 Set up project To be able to configure and build CoCo you need to give Ant a few properties e GO lt COCO_HOME gt ant e Edit build_ant properties and provide required properties help embedded e Manually create the directory that you specified in the property coco data home in build_ant properties file and make sure that the user running tomcat has full access rights on it rxw e Then run gt ant configure 3 5 Install required R library CoCo needs different R library to be installed In addition CoCo uses functions from Bioconductor packages To ease things simply install the whole Bioconductor following the procedure explained at http www bioconductor org docs install howt
29. dmin Mef2_ChIP_8 10 CHIP_ON_CHIP_EXP_TYPE nogroup Delete eat rd C Mef2_oe_5 6 txt File ID 4 admin mef2_xpro_5 6 EXPRO_EXP_TYPE nogroup Delete mi C Mef2_oe_6 7 txt File ID 9 admin Mef2_oe_6 7 EXPRO_EXP_TYPE nogroup Delete gt Help C Mef2_oe_7 8 txt File ID 8 admin Mef2 oe 7 8 EXPRO EXP TYPE nogroup Delete gt Admin Tools C fly_muscle_anatomy_terms txt File ID 5 admin ANATOMICAL_TERMS nogroup Delete gt Log out Update marked Group Rights same World Rights Same Save changes Reset 5 Managing your configurations You can see your configurations using the Configurations menu There you can view details or delete your configurations Note that sharing configurations is not yet possible We encourage users to delete old configurations as configurations occupy quite some space on the server E CoCo Server Analyze your ChIP On Chip data Online 90 Online users 1 admin Configuration List Configuration Main Additional gt Start Analysis ne as ChIP chip ChIP chip ExpPro Anatomy Dev stage Chip name Org Maintenance gt Configurations testmef2 Mef2_ChIP_6 8 2 1 muscle stage stage3 stages genomic_tiling_Dm_version2 Drosophila Re create gt Current config melanogaster Result Table gt Generate Pictures In batch z E men test Mef2_ChIP_6 8 1 1 Pea stage stage10 stage genomic_tiling Dm version2 DOSPhila_ R gt Uploaded Files neur
30. e As 29 SO STICKY PILE MADE Ee ee ee A de ee Dee ie ee SONE ee a a 29 8 6 REGULATORY REGION did ENE EE EE RE EE EE EE Ee N 29 8 GENOME ANNOTATIONS ie eed sege Gee Se ed eg Ee Ee Gee Ge Ee SE GE seed Ee Ee Ge EN ee RENE GE EE SEE Ee EG See aiii 29 9 SUPPORTED ARRAYS NOTE ABOUT HIGH DENSITY ARRAYS oue seesesse see soe sesse see see see sesse see see sesse 30 9 1 TECHNICAL CONSIDERATIONS cccccssscccessssccesssseccessssecceesseeceseseeceesssecceesaeecceesssecceeeeecessateecessaees 30 922 OTHER CONSIDERATIONS in iet A A ee sd ge 30 9 3 WHAT YOU CAN DO WITH THE CURRENT COCO VERSION esse see see see se ee ese ene ee ese ee ese ene ea ee ee ee 32 COCO INSTALLATION AND MAINTENANCE DOCUMENT ese sesse sessesessese sesse se ssese sesse ses 34 1 LS IAN AAA 34 2 INSTALLATION OVERVIEW ont inicia 34 3 DETAILED INSTALLATION STEPS csssccsesisvssiensascsscdsconseenssestneccevsssensvesthedsevedsunsessseecessssconcessseecssee 36 3 DOWNLOAD COCO a ratos 36 3 2 UNCOMPRESS THE ARCHIVE cccsccssscssscssscsssesssesssessscssecsssesssesssesseessscssecsssesssesseceseesseseeseeeeeenees 36 3 3 CREATE THE COCO S OE DATABASE AAA AA 36 SET UP PROJECT A A A A A ORONA 37 3 8 INSTALLING DEMO DAT Ankkaan ehi se siele ss bak ERS E ER E E EEAO E EEA 39 3 9 BUILD THE COCO WAR AND DEPLOYING COCO ccccsccssssssscesscesscesscesscesseesscesscessecsscesscesscesseenees 39 3 10 COCO MANAGEMENT AND MAINTENANCE esse esse esse esse ee see
31. egion chromosome e region start the regulatory region start e region stop the regulatory region stop As a minimum file CoCo will accept a file containing these 3 columns only though it is really not informative to load such regions Optional headers e region strand if applicable the region strand e region confidence the confidence you have about this region to be real One of tentative predictive confirmed reviewed If not provided it is defaulted to tentative e region comment a free description about the region use this field to store e g literature information e ff a transcription factor binding to the region e ff_binding_anatomy the anatomical part s where the binding occurs comma separated values accepted to specify multiple anatomical parts e tf_binding_stage the stage s at which the binding occurs comma separated values accepted to specify multiple stages e tf_binding_site_start the transcription factor binding site start in case you know where exactly the transcription factor binds within the region i e its binding site you can precise it e tf_binding_site_stop the transcription factor binding site stop see above e tf_binding_site_strand the transcription factor binding site strand see above e target gene the gene which expression is modulated by the transcription factor in the given spatio temporal conditions e target_gene_modulation the type of expression modula
32. enu and delete those that CoCo haven t succeeded to clean as they are certainly corrupted If you leave them you won t be able to upload a file having the same name and you will pollute the server with corrupted files that will appear in select menus Note that mandatory fields have a red star close to their name 2 1 Experimental Parameters Section Select the organism and the microarray design used to perform the ChIP on chip experiments Select the transcription factor used in the Chromatin IP In case you plan to mix datasets gained with ChIP using different transcription factors you should provide the one used for the main ChIP chip experiment We ll come back to this later Indicate the Developmental Stage and Anatomical term lists the configuration should focus on These term lists are supposed to reflect your experimental setup and will be used to color genes using available in situ data For example if you used samples collected from the heart muscle at developmental stage 2 you are certainly interested to clearly see those genes expressed in the heart at the same stage or maybe at stages 1 to 4 CoCo offers you different ways to provide this term lists but you should use only one of them 1 A free text box this option is relevant when you have a unique term e g heart muscle For the developmental stage only you have the option to give a range This must be in the form lt prefix gt lt from gt lt to gt e g sta
33. es after a change e g after addition of a new chip or db connection settings update Log out close work session In addition to these basic operations other context dependent possibilities will be offered 2 Creating a Configuration A configuration is a space where users can integrate ChIP on chip datasets with microarray expression profiling data in situ patterns and genome annotations In addition to the organism and the ChIP on chip microarray design a configuration is given a developmental stage term list and an anatomy term list Once a configuration has been created users can visualize all data on interactive pictures that can be accessed in a genomic browser fashion Input data format is tab delimited and data should be already normalized Please see the file format chapter to know more about file formats Experimental Parameters Organism Drosophila melanogaster Array genomic tiling Dm_version2 Transcription Factor used in the ChIP Dev Stage Or Choose Existing Term List none 4 Or Upload New Term List File Choose File no file selected Anatomy Or Choose Existing Term List select 3 Or Upload New Term List File Choose File no file selected Exp select Exp File Choose File no file selected Main a Or Upload New Exp Name aose re E Mock select 3 ope ee ad Mock File Choose File no file selected Exp select gi r Enp File Choose File no file selected Additional Results
34. ese ee see ese ee see ese ee ee ee ee ee ee ee ee ee 40 Introduction The ChIP on Chip online CoCo application helps you analyzing your ChIP on chip results CoCo integrates results from ChIP on chip experiments together with in situ and gene expression profiling results All these datasets are put in a genomic context and displayed as meaningful colored pictures User can browse results in the fashion of a genome browser and define regulatory regions together with target genes while browsing User Manual Introduction CoCo is a web application that allows the user to search visualize and store different data associated with gene expression The program integrates ChIP on chip expression profiling and in situ hybridization data to create a user specified configuration The data can then be visualized and searched on a user friendly interface which displays all data as well as the surrounding genes The user can zoom in and out to different genomic regions and save images of the displayed data 1 Login in menu overview First ask your CoCo administrator for a login If you are the CoCo admin then please consult the installation and maintenance section You can then point your browser to http lt servername gt coco e g http localhost coco and login Once login you ll be looking at page like the below picture TYTYTY Welcome to the COCO Server lt Ma Ao d http localhost 8080 coco secure well
35. gel 5 would translate in term list stage stage2 stage3 stage4 stage5 Note that lt prefix gt can t have spaces and lt from gt lt to gt part must be numbers If you break these rules form validation fails 2 Select an existing term list here you can select term lists already available on the server A term list becomes available as soon as you uploaded such a file using the third option 3 Upload a file containing terms The file should have one column one term per line Now comes the question what terms should I use Well this depends what you used in your in situ files you must use the same and this is case sensitive Dealing with in situ data is quiet complex and can t yet be generalized even though ontologies for both developmental stages and anatomy exist for several organisms The basic problem is that in situ results have usually not been annotated using those The other problems is that these ontologies have thousands of terms with complex relations Because of this you always have to pre process in situ results and turn annotations into simpler classifications We never succeeded to go around this pre processing step so we decided to have a simpler approach in CoCo that allow to cope with all situations as you pre process in situ results you should build relevant term lists Once done give them a meaningful name and upload them in CoCo as admin share them with relevant groups and users will simply have to
36. hip dfci harvard edu wli MAT or TileMap http biogibbs stanford edu jihk TileMap index htm after normalization and look at enriched regions directly We think that the best approach would then be to allow users to upload regions directly This latest approach is certainly quite easy to implement and we ll implement this soon in CoCo 17195Mb 17 2Mb 17 205Mb 17 21Mb 17 215Mb 17 22Mb 17 225Mb CG7095 17204701 17207692 44 or eee ee ie CG6475 Figure 1 Display of a 30Kb Region Using a NimbleGen Tiling Array po 17 2Mb 17 21Mb 17 22Mb 17 23Mb 17 24Mb 17 25Mb 17 26Mb iia 44 A ep 1720321 235699 Tre59e45 Figure 2 Display of a 70 Kb Region using a NimbleGen Tiling Array 9 3 What you can do with the current CoCo version If you want to upload results from high density arrays in CoCo here is what you can do and what we did for NimbleGen array 1 Define the chip as usual In case you don t use the MM probes Affymetrix chips don t put them in the clone map This will save 3M lines and lots of compute time 2 Pre process your chip chip files as you build them for CoCo remove all those probes that are clearly in the background by applying some filtering of your choice Indeed these feature results are useless and they will only increase compute time Filtering could be 1 keep only features that belong to enriched regions i e
37. ion but will use all lines to create binding events and target genes 2 tf_binding_stage and tf_binding_anatomy accept comma separated values If you chose to use this a binding event will be create for each possible combination If this is not reflecting reality you must duplicate lines as needed For example if the TF mef2 binds a region 2R 100 200 at stagel in the heart stage2 in the heart and stage3 in both the visceral muscle and the heart you must define 2 lines e g only relevant headers are shown region_chr region_start region_stop tf tf_binding anatomy tf binding stage 2R 100 200 Mef2 heart stagel stage2 2R 100 200 Mef2 visceral muscle heart stage3 The region defined in this example will end up with 4 binding events 3 If you describe a target gene in the line it will be added as a target gene to all binding events described in the line From the previous example let s assume that the two lines have the value twist in the column target_gene gt the four binding events will have twist as a target gene but as 2 distinct relationships Indeed the first line could describe an activation type target_gene_modulation while the second a repression gene modulation type 4 To associate multiple target genes to the same binding event You should follow the same strategy as described in 2 6 3 Creating a region from scratch You can define a new regulatory regio
38. is to let your users know what terms have been used so that they can build relevant term lists to give to CoCo Note Defining default in situ result files 1s optional 7 5 Defining User s in user properties Well simply add a line holding required description for the new user s You ll find examples and format description in the demo user properties file Users can belong to different groups and have multiple roles In such situation list all roles groups using a comma separated value list in the appropriate column Note about groups guest and temporary users might have no group in this situation simply register them in the special nogroup group But be aware that all nogroup users might see each other data Roles known in CoCo e user the role for every user e admin an admin sees everything only admin have access to the Admin Tools menu e groupLeader a groupleader will see all data of his group even when not shared with the group
39. ist of chromosome names Note that the order in which you list the chromosomes are kept in CoCo display For each chromosome define a genome layout gff lt taxid gt lt version gt lt chrName gt which value in the absolute path to the annotation file in GFFv3 For a complete description of this format please see http flybase bio indiana edu annot gff3 html or http www sanger ac uk Software formats GFF GFF_Spec shtml Note gene symbols names and synonyms are extracted from the ninth column Used fields case sensitive are ID ends up as gene symbol Name ends up as gene name Dbxref dbxref_2nd and synonym all end up as synonyms For an example look at the demo data shipped with CoCo Note 2 to speed up computation time we encourage you to filter GFF you download and keep only relevant annotation types take a look at gff genome feature types to know which types are relevant 5 Add genome layout rdata lt taxid gt lt version gt lt chrName gt properties the same way you positioned GFF paths at step 4 Note These Rdata files don t yet exist You ll generate them at next step in which these genome layout rdata lt taxid gt lt version gt lt chrName gt will be used 6 Run ant update genomes 7 2 Defining new Tiling Arrays in coco properties 1 Add the chip name used for display in supported chipnames property this property accepts a comma separated list of chip names
40. king on different items will offer you different options If you click on a gene you ll be offered see picture below to either consult the gene summary page or define this gene as a target gene of a regulatory region ONO Gene Selected Select Action To Perform You have clicked on Gene CG10379 3R 19607508 19627396 Please select the action you want to perform e View Gene description details choose e Define CG 0379 as a target gene of an existing Regulatory Region Select Regulatory Region none Next Close Window Please contact charles girardot for feature requests comments or reporting bugs If you click on a tiling array feature you ll be offered see picture below to consult the feature summary page define a new regulatory region based on this fragment or add this experimental results as a supporting evidence of an existing regulatory region this option is offered only if the feature you selected overlaps with exiting regions You have clicked on chip feature D 99 6 e3 2R 5449064 5452214 in the experiment named mef2_1P_6 8 Please select the action you want to perform e View Fragments description details e Create a new Regulatory Region based on this fragment e Add a Experimental Binding Evidence to an Existing Regulatory Region based on the ChIP chip result of feature D599_6_e3 Select Regulatory Region select Next gt Close Window Ple
41. know how scientists have come to their conclusions CoCo addresses this issue by letting users give confidence to their conclusions and attaching experimental evidences For example CoCo automatically records evidences about the ChIP on chip results used to initiate a regulatory region definition These features allow users to accumulate conclusions about regulatory regions over time and should ensure reusability CoCo offers different ways to create regulatory regions 1 Import from file 2 Create a region from scratch e g that you found in literature 3 While browsing your results as explained before 6 1 Regulatory region model in CoCo In CoCo a regulatory region is not only defined as a genomic location Here is the model used in CoCo A regulatory region RR is a genomic region where transcription factor s TF bind to the genome Thus a region doesn t only describe a single binding site but a regulatory module The fact that a TF binds to a regulatory region is referred to as a binding event that occurs in specific spatio temporal conditions A regulatory region can have many binding events described for multiple TFs When describing a binding event for a given TF you can optionally specify spatio temporal conditions the exact binding site boundaries and a target gene In addition binding events can be supported by experimental evidence s A target gene is a gene which expression is affected by a binding event Because a regu
42. l be offered only if the selected feature overlaps with existing regions Simply select the right region from the drop down menu and complete the wizard 6 5 Assigning target genes to regions while browsing experimental data As you browse your results clicking on a gene will offer you the possibility to assign this selected gene to existing regions as shown on the picture below Note that the regions proposed in the drop down menu are those found in the current genomic region i e in the picture you clicked on You have clicked on Gene CG7714 3R 14551116 14552156 Please select the action you want to perform e View Gene description details choose e Define CG7714 as a target gene of an existing Regulatory Region Select Regulatory Region 3R 12636468 12636491 abd A Next Close Window 6 6 Searching for regulatory regions The Search sub menu of the Regulatory Regions menu lets you list regulatory regions stored in CoCo As you click on the link no regions will be displayed and a message will invite you to position filtering criteria as shown on the picture below Note regions are displayed for a unique organism Note 2 only regions on which you have read rights will be displayed CoCo offers an extensive filtering interface It works the following way e To view all regions don t fill any filtering fields and simply click on Filter e To filter on some criteria fill the appropri
43. latory region may be associated with many binding events it may have many target genes Actually CoCo allows you to associate or assign more than one target gene to a binding event and by extension to a region This is especially useful in situations where 1t is unclear what gene s are affected by the binding event Confidence values are attached to both the regulatory region and target gene assignment Different values are available reflecting different level of confidence e Tentative is the lowest confidence level and indicates well a possibility e Predictive comes after tentative and indicates that the conclusion comes from a prediction tool that s the real difference between tentative and predictive e Confirmed indicates that the region or the target gene assignment has been experimentally confirmed experimental evidence should be available e Reviewed indicates that the region or the target gene assignment has been published in the literature The picture below shows an example of a unreal regulatory region created while browsing experimental data Online users 1 me Regulatory Region Description Page Edit Region Delete Region gt Start Analysis y Chr 2R gt Configurations gt Current config Start 5449064 gt Overview Page gt Generate Pictures In batch Stop 5452214 Strand gt Uploaded Fil ELA Confidence tentative gt Regulatory Regions Organism Drosophila melanogaster gt List Al gt Ne
44. le selected Exp File Additional select Exp Name Choose File Mef2_10_12_ChIP_all txt Results 2 a Or Upload New e Wp 10 12 Mock File EE Choose File Mef2_10_12_mock_all txt Exp File Additional Exp _ select gt Exp Name Choose File no file selected ER Or Upload New Results 3 select Mock File _ _ Choose File no file selected Exp File Additional Exp select _ Exp Name Choose File no file selected EE Or Upload New 5 Results 4 select Mock File EE Choose File no file selected Mef2_6_8_mocl Sticky genomic select Or Upload New fragment list File Choose File sticky_fragments_demo txt 2 3 In situ Results Section The in situ result section lets you attach in situ results to your configuration Because preparing in situ results might be quite time consuming and delicate task CoCo offers the possibility to define a default in situ dataset per organism This has to be configures by the administrator and will be available to all users If present this dataset is included by default in every configuration but you can indicate that you don t want to use it In addition you can specify or select a file holding additional in situ results e g collected in your lab Note that the file name will be used as the dataset name in displays so you might want to keep it short and meaningful The picture below shows an example where we upload in situ results in addition t
45. ly for a subset of the results this is controlled but the Display top x result option If you choose to display sticky features they will appear on a grey background The chromosome overview pictures are built with results of the main ChIP chip experiment only The same is true for the result table ordering when ordering by score is selected These are examples where it is a bit tricky to find a way to apply user selection to all results together Online users 1 admin ChIP cut off 0 7 Gene Search Search M Search Chip Feature by ID ES Mock cut off _ Or Go To Chr 3R_ From 1 To 10000 Go m gt Configurations Display top x result gt Current config Global Picture Display Options Show all options Save Options gt Overview Page z Draw Overview EE Kataris Ta baidi Main ChIP options Mef2_ChIP_6 8 and Mef2_ChIP_6 8MOCK gt Uploaded Files ChIP cut off gt 0 7 mock cut off lt 0 3 and notransfo gt Regulatory Regions gt List All gt New Region 4 gt Import 5 gt Help a ii II gt Admin Tools r R _ I gt Log out x LI 2 EE AA p x o Result List Your result filtering criteria produces 62 Enriched Fragments A ChIP cut off 0 7 Display top x result 10 Display Sticky Features O e Options EA r ma Submit Mock cut off 0 3 Sort by score Show the 0 closest genes res Fea
46. ms Note that the value of the synonym attribute can specify a comma separated list of synonyms CoCo will register each synonym in the gene synonym table 9 Supported Arrays note about high density arrays CoCo virtually supports all arrays as long as the server running CoCo has enough memory and users are patient enough In practice we have observed acceptable compute times with arrays up to 100K This means that the 6M Affymetrix tiling arrays will certainly not work well with the current release of CoCo on small servers e g 1Go memory As we are starting to use Affymetrix arrays we are willing to address this issue soon so you should keep checking the coco download page for new release in the next months 9 1 Technical considerations In the current version of CoCo each tiling array must have a clone map file a file containing the genomic coordinates of each feature found on the tiling array that is used to find feature positions at configuration creation This mapping is performed by loading the whole map in server memory This results in high memory requirement for high density arrays and pretty long processing time to create a configuration That s the first technical limitation The second limitation would then be the time required to assemble each picture on user request Indeed CoCo starts a new R process for each picture generation Thus the time to build a picture is the sum of the time for R to start the time
47. n using the New Region sub menu of Regulatory Regions menu Simply fill in the form see example below If your CoCo installation supports multiple organisms changing the organism value will update both the genome version and chromosome lists The form is split in two section 1 Regulatory Region Boundaries and Binding Condition this section holds values to define both the regulatory region and the Transcription Factor binding event This section must be filled 2 Target Gene Assignment here you can define a target gene of the region Mandatory fields are marked with a red star Note that the spatio temporal conditions are not mandatory as it is sometimes hard to find this information published regions but we encourage you to fill these fields whenever you can As usual in the Transcription Factor and Target Gene fields simply specify the official symbol or gene accession ask your administrator about which database accession you can use CoCo will anyway complain if it can t map uniquely your input to the database CoCo Server Analyze your ChIP On Chip data Online 30 YV Online users 1 admin Define a new Regulatory Region SEET Region Organism Drosophila melanogaster Genome Version 4 ron s Dd From roe gt Configurations Region boundaries Chr 4 From To gt U ded Files Transcription Factor Binding Event Observed binds EE Spatio temporal Binding se SR latory
48. o Oaka Region Origin Filter Experimental Region Confidence Filter tentative gt Current config multi select allowed flymine multi select allowed predictive gt Overview Page flyreg reviewed gt Generate Pictures In batch confirmed gt Uploaded Files Show only regions CG31211 Drosophila melanogaster Show only regions abd A CG10325 Drosophila melanogaster 7 bound by TF s 1 28 CG9397 Drosophila melanogaster controlling gene s Abd B CG11648 Drosophila melanogaster gt Regulatory Regions multi select allowed abd A CG10325 Drosophila melanogaster multi select allowed Adf1 CG15845 Drosophila melanogaster SEES ac CG3796 Drosophila melanogaster a Aef1 CG5683 Drosophila melanogaster gt New Region alphaTub848 CG1913 Drosophila melanogaster Antp CG1028 Drosophila melanogaster h EE Zee Filter Reset gt Help gt Admin Tools gt Log out Please position filtering criteria 7 Help in CoCo CoCo has three help resources 1 this document 2 help displayed on you move your mouse over e g form fields 3 little help icons are usually present at every page Clicking them will open up a context sensitive help window 8 File Formats 8 1 ChIP chip Data File CoCo accepts ChIP chip results as a tab delimited file holding 2 columns e the first must hold a feature ID e the second a experimental value What you provide here is up to you but the value should certainly be some kind of enrichmen
49. o html Alternatively make sure to have the following packages from Bioconductor e Geneplotter e Annotate e Biobase To install R libraries that come with CoCo please read following instructions Note you might have to log in as a user having permissions to remove install R libs Tf not the first installation gt R CMD REMOVE gff3Plotter Then Vv gunzip untar configure make make install for gd 2 0 33 tar gz gt R CMD INSTALL lib R GDD_0 1 4 tar gz gt R CMD INSTALL lib R gff3Plotter lt version gt tar gz 3 6 Update Hibernate Configuration Important This step is optional if you use MySQL as the hibernate cfg xml generated by the ant configure task in Set up project generated this file for you already You can check it out If you change database properties in the future this is anyway how you can tell hibernate about the changes Procedure Edit the Hibernate Configuration file hibernate cfg xm1 located in lt COCO_HOME gt src java Update the three following lines with MySQL coco database information lt property name hibernate connection url gt jdbc mysql localhost 3306 coco lt property gt lt property name hibernate connection username gt cocouser lt property gt lt property name hibernate connection password gt aPWD lt property gt Replace values with appropriate settings Note that if you use another RDBMS than MySQL you ll certainly have to update these
50. o use the default in situ file in this case BDGP in situs In Situ Results Upload additional in none B Or Upload New File Choose File insitu_fly_embryo txt situ data Ignore default in situ data _ used by default 2 4 Expression Profiling Results Section The last section lets you attach gene expression profiling results to your configuration This section looks pretty much like the other sections You can specify from 0 to 5 expression profiling datasets by either selecting available datasets or uploading new files If you choose to upload new file s you must give an experiment name to each of them and you should as usual keep these names short and meaningful Important The order in which you specify datasets is kept in CoCo and in displays The picture below shows an example where we mixed experiments already uploaded on the server with new experiments new file upload Note that e you must not give an experiment name for experiments that are already on the server e you must give an experiment name for new experiments Expression Profiling Results Or Upload New Exp Name g File Expro Results 1 Mef2 oe 5 60 MEA no file selected Or Upload New Exp Name File EE sente 2 id EN Expro Results 2 _Mef2_0e_6 7 gt y Choose Fi no file selected Tad Or Upload New Exp Name File M Mef2 oe 7 8 Choose Fi Mef2 oe 7 B txt Expro Results 3 select Or Upload New Exp Name File Rs Choose Fi no file
51. oblasts abdo at R m Mef2 timeSerie Mef2_ChIP_6 8 4 3 bes stage 10 stage stag genomic_tiling_Dm_version2 Drosophila Re create gt Regulatory Regions ig iz neuroblasts abdo gt ER ii EERS melanogaster Result Table gt List All gt New Region gt Import gt Help gt Admin Tools gt Log out Clicking on a configuration name brings you to the detailed Configuration view shown on the picture below or E CoCo Server Analyze your ChIP On Chip data Online 0 Online users admin Details for the configuration Mef2_timeSerie gt Start Analysis gt Configurations gt Current config gt Generate Pictures In batch Change Configuration Name Delete gt Uploaded Files Chip name genomic_tiling_Dm_version2 EE Organism Drosophila melanogaster gt ns Ps rr RED Main gt New Region ChIP chip Mef2_ChIP_6 8 aed Exp gt Help Mef2_ChIP_6 8 Mef2_6_8_ChIP_all txt Mef2_ChIP_6 8MOCK Mef2_6_8_mock_all txt Sn LO ChIP chip files Mef2_ChIP_8 10 Mef2_8_10_ChIP_all txt SE Mef2 ChIP 10 12 Mef2_10_12_ChIP_all txt Mef2_ChIP_10 12MOCK Mef2_10_12_mock_all txt mef2_xpro_5 6 Mef2_oc_5 6 txt Expro files Mef2 oe 6 7 Mef2_oe_6 7 txt Mef2 oe 7 8 Mef2_oe_7 8 txt In situ files Sticky files Vi 7 gt a stage 10 stage 1 stagel 2 stagel 3 stage 14 stage15 stage2 stage stage4 stages stage6 stage7 stages stage Pc3 neuroblasts abdominal dorsal transverse muscle abdominal intersegmental apodeme a
52. or hide partially this panel In the example below we have three ChIP chip experiments two of them have associated mock results In the main experiment we define enriched features as features where e the ChIP understand enrichment of the test dataset enrichment here we loaded log transformed enrichment over genomic DNA is greater them 0 7 e AND the enrichment in the mock dataset is less than 0 5 AND the difference between test and mock value is more than 0 5 The second experiment has no mock data and thus a unique ChIP cut off can be positioned Finally the third experiment has no transformation condition i e enriched features have to be over 0 7 and less than 0 3 in the mock The configuration has expression profiling datasets which values are log ratios We then set expro cut off to 0 and min max to 1 5 1 5 respectively Global Picture Display Options Show main experiment options only Save Options Main ChIP options mef2_1P_6 8 and mef2_1P_6 8MOCK ChIP cut off gt 0 7 mock cut off lt 0 5 and exp mock gt 05 N Options for ChIP mef2_IP_8 10 ChIP cut off gt 0 7 Options for ChIP mef2_IP_10 12 and mef2_IP_10 12MOCK ChIP cut off gt 0 7 mock cut off lt 0 3 and no transfo re 2 os Expression cut off 0 0 Min Value 1 5 Max Value 1 5 3 2 3 Interactivity saving conclusions as you browse The picture is interactive and you clic
53. or re run using the Ignore Ambiguous Gene option on regions with ambiguous genes will be ignored wal E CoCo Server Analyze your ChIP On Chip data Online 90 Y Online users 1 admin Upload Regulatory Regions a gt Start Analysis gt Configurations All regulatory regions in your file must be from the same organism and coordinates must refer the same genome version gt Uploaded Fil o gt pes Organism Drosophila melanogaster gt Regulatory Regions Fo y Beas Genome Version 43 ie shed see Data Origin Name i ale e g FlyReg gt Help Group rights Ria gt Admin Tools World rights SEE Region File Choose File no file selected Ignore Ambiguous genes _ Upload File Format The file format is quite extensible and allows you to give extensive details or not Here is the complete list of columns that can be found in the file column order doesn t matter Note that there are both mandatory and optional columns and that providing some columns imply that you provide others i e optional columns might become mandatory Important headers must appear as the first valid line in the file Comment lines 1 e line starting with can be found prior to the header line Headers must be written as described and must respect the case comments more comments region chr region start region stop more tab delimited columns with headers Mandatory headers e region chr the regulatory r
54. t value or statistical score reflecting the likelihood that this feature is enriched 8 2 In Situ Data File CoCo accepts In Situ results as a tab delimited file holding 3 columns e Column 1 GeneSymbol e Column 2 dev_stage e Column 3 anatomy 8 3 Expression Profiling Data File CoCo accepts Expro results as a tab delimited file holding 2 columns e the first must hold a feature Id or gene symbol accession e the second a experimental value What you provide here is up to you but the value should certainly be some kind of expression value e g log ratio or statistical score reflecting the likelihood that this feature gene is differentially expressed 8 4 Term List File Term lists are provided as simple file with a unique term per line only one column 8 5 Sticky File Sticky feature lists are provided as simple file with a unique feature ID per line only one column 8 6 Regulatory Region File Please see chapter 6 8 7 Genome Annotations CoCo uses GFFV3 file format http flybase bio indiana edu annot gff3 html to define genome annotations Particularly important aspect is what you put in the last column e CoCo uses the ID attribute of gene annotations to create gene symbol In CoCo gene symbol must be unique Every gene annotation must have an ID attribute e If present CoCo uses the Name attribute of gene annotations to create gene names e If present CoCo uses the synonym attribute to create synony
55. the stages and or anatomy specified in the configuration Finally the upper and lower zones represent expression values for genes found on the plus and minus strand respectively Each expression dataset has its own track and color coded rectangles aligned with their corresponding genes are draw whenever result is available in case more than one results is available for a gene the mean together with standard deviation is used and displayed Rectangles are colored using a color ramp from blue under expressed to yellow over expressed where the minimum and maximum values are user defined Information about genes expression values and enrichment folds is displayed while moving the mouse over the picture In addition clicking on genes or ChIP on chip features opens dialog pages allowing users to undertake actions like accessing gene or feature report page creating regulatory regions or assigning genes to regions The picture below presents a genomic region view example where enriched features are found in a gene dense region Here the use of CoCo is certainly needed to find which gene s are under control of Mef2 gt Configurations gt Current config gt Overview Page gt Generate Pictures In batch gt Uploaded Files gt Regulatory Regions gt List All gt New Region gt Import gt Help gt Admin Tools gt Log out Hide Legend Legend ChIP on Chip enriched not enriched EE sticky fragment Expro data
56. tion one of activation repression unknown e target gene confidence the confidence to give to the gene assignment One of tentative predictive confirmed reviewed e target_gene_comment a free description about the gene assignment e target gene evidence a free comment about evidences supporting this assignment If the assignment comes from a prediction i e bioinformatics tool the following fields can be provided e target gene prediction origin the name of the prediction tool together with parameters used e target gene prediction score a score for the gene assignment if available e target_gene_prediction_score_type if a score is provided its type i e a small 255 char description of the score meaning e g p_value As explained dependencies between fields occur e tf_ fields should be found ONLY if the f field is described e target_gene_ fields should be found ONLY if the target_gene field is described AND if tf is described e tf_binding_site_start and tf_binding_site_stop should be both provided or both empty If f_binding_site_strand is provided both tf_binding_site_start and tf_binding_site_stop should be provided e If target_gene_prediction_ field s are provided target_gene_prediction_origin is mandatory Regulatory region creation rules when importing regions 1 If multiple lines hold the same region_chr region_start region_stop CoCo creates a unique regulatory reg
57. ture ID Sticky Mef2_ChIP_6 8 Mef2_ChIP_8 10 Location Annotations D1029_2_g5 0 3 70 mock 3 70 323 2L 17195974 17199231 DI029 2 gs DI1006 1 all 0 3 11 mock 3 11 261 3R 26078055 26080696 D1006_1_all D294 2 ell 0 3 09 mock 3 09 2 75 2L 8479188 8482399 D294 2 ell DI142 3 g2 0 3 03 mock 3 03 230 X 18035631 18039278 D1142_3_g2 D708_2_b12 0 3 03 mock 3 03 1 69 2R 9995662 9997807 D708_2_b12 From the overview page several options are provided to switch in genome browsing mode 1 you can click on chromosome pictures this will open the genome browser centered on the region you clicked The region displayed will be quite broad 2 you can follow a link from the result table The region displayed will be sharper the size of it is actually a property the administrator can set By default it is 30 Kb 3 you can use one of the search option search by gene use symbol or synonyms search by microarray feature ID the microarray is the tiling array not expression microarray or specify a genomic location Finally you can set directly in this page the parameters you want to use in the genome browser mode in Global Picture Display Options This Global Picture Display Options panel as well as the search toolbox will be available in genome browser mode as well 3 2 Genome Browser From the overview page users can start browsing data in a genome browser fashion To enter the genome browser users can search by gene symbol or
58. w Region Genome 4 gt Import Version gt Help User Comment Creation Date 2006 06 30 16 22 03 0 gt Admin Tools Owner me gt Log out Group rights R World rights Transcription Factor Binding Events Observed Transcription Factor Anatomy Developmental Stage Experimental Evidences Target Gene s Assigned to this event Gene Mef2 CG1429 Mef2 CG1429 muscle stage5 1 view Modulation activation Confidence tentative From the above picture you can see that the region has one binding event defined for mef2 The column Experimental Evidences indicates that the binding event is supported by one experimental evidence If you click the link a page displaying evidence details supporting this binding event and by extension the regulatory region is shown In this example the evidence recorded the experiment chip feature ID and feature enrichment values P y CoCo Server Analyze your ChIP On Chip data Online 0 wy Online users 1 me Regulatory Region Description Page Back gt Start Analysis 7 PAPE E N Experimental Evidence s Supporting the binding of CG 429 Mef2 on genomic region 2R 5449064 5452214 gt Configurations gt Current config 7 7 m gt Overview Page Experiment Name Chip Name Chip Feature Name Experimental Values Creation Date o heer mef2_IP_6 8 genomic_tiling Dm_version2 D599 6 e3 Exp 0 8454 Mock 0 2006 06 30 16 22 03 0 gt Uploaded Files gt Regulatory Regions gt List
59. xisting Term 7 List stage4 17 New Term List Choose File no file selected Anatomy Or Choose Or Upload File Existing Term fly_muscle_ana List New Term List Choose File no file selected Lis 2 2 ChIP on chip Results Section CoCo accepts up to five ChIP chip experiments One experiment is mandatory and is call the main experiment When uploading more than one experiment try to define the most relevant as the main experiment CoCo uses the main experiment data in different situations where it is not possible to mix all datasets together or not yet implemented As you read this document we ll point you such situations For each ChIP chip experiment you should either e select files from drop down menus or e upload new file s and give these file s an experiment name Please use short experiment name e g 10 letters To cope with common design ChIP chip experiments can be made of two different files 1 e two different result sets a test and a mock result set where the mock represents results obtained in the same conditions as test but using a mock antibody in the ChIP step In such designs two hybridizations using two channels platforms are performed the first hybridization measures test sample signal over genomic DNA while the second measures the mock signal over genomic DNA Providing mock results is optional Finally at the end of the form section you can provide a sticky file This file is a single

CoCo Administration and User Manual

Contents

Download Pdf Manuals

Related Search

Related Contents