NETI 1.2 User Manual
Contents
1. Fit Iters 100 Target 1 288 Model Prior Kernel Background Times 4 Exponent Inverse Polynom 1 0 0 1 0 0 09g o 02522458mn0989n LE e E e M J Self Regulation x o E Graph v of P kkk p FE KKK PP EE B jy KPase PF K FKKPase P KK Se 0 cr ia e KKPase_PP_KK KKPase 0 2245mnumo PP KK K KPase P K Time K P KKK KK E1 Y KPase E2 s y PP_K KKK lw n wom a eam le Se a eas e PPK PI S 8 Sam ELI 18 M Hr i a 16 N y f j 4114 a a aml w 2a __ _ _____ 0 x2 181 788 NP 22 an oes2a220na09 oega3a2seon ao CoOnDaADANAD x Time Time Time v lemory 495728256 NETI 1 2 Manual Copyright C 2004 2007 35 Institut Curie Model Selection Result Times NETI12 D inWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help eS mM b M Fit Iters 100 Target Model Prior Kernel Background Times es Exponent Inverse ED The model selection 0 1 0 9 procedure after testing the eight models defined in the Template table selects the best model and eer Renee copies the identification of this model into the 10999 anaana NA Kinetics P Kkk p EM KKK PP KK B fy2. Network inference can File Run Window Help be performed in the i 6 6 BN JW automatic mode using E arget the Run button from l the Toolbar or the Menu Kernel Background Times T E Item RunlRun F5 Exponent Inverse Polynom 1 IE EL 1 0 0 y t i 0 0 ls a j iit l 3 comm aml 1 YR NU OE Self Regulation Y P_KKK_P us KKK Pase PP K Kinetics The procedure performs until the number of created links is higher than the ters value or until the 7 value is lower that the Target value Kinetics 1 266 NP J 0 10 1 194 NP 01 232498xu amp m 1oaDRaODaNAD 0u zs2seoesoncm Time Time Time lemory 515937296 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 23 Institut Curie Terminate Processing NETI12 D inWork Software Setup WWW_RE demodata WMAPKDistorted bin File Run Window Help l b mm W Kernel Background Times Exponent Inverse Polynom Processing can be stopped by pressing the 5 Mm ETETTITITY EN TTTTITIIT Mm ennanana Stop button on the Toolbar or selecting the Menu Item RunlStop P_KKK_P_KM KKK H KPase_PP_K KKPase_P_KK KKPase_PP_KK KKPase PP_KK_K KPase_P_K om PKR KK 0192323499299 KPase PPK a PK M 1x s pe T f 0 W 4 N 4 om fi Eas Pt 35 901 NP
3. 0 ao am zi amt REELE ana w cen2asanao cona2asanao Time Time Time lemory 500330760 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 17 Institut Curie Forward Selection Stopping Criteria Iters is the stopping criterion maximal number of links that can be created Target is the stopping criterion minimal limit for the 7 overall fitness criterion Novikov E Barillot E Regulatory network reconstruction using an integral additive model with flexible kernel functions submitted to BMC Systems Biology 8 Oct 07 NETI12 D InWork Software Setup WWW_RE demodata WAPkDistorted bin File Run Window Help i db d NX i i Fit Iters Target Model Prior Kernel Bw ground Times eXponent Inverse Polynom 1 0 0 1 0 0 ___ 100 m Self Regulation v 0l EC EE KEROKI OKH KOPOP HI Del ORDER ERE ABIRE ERE Kinetics e e nasasaad g ie Soom v e tms x toma toms oo nanasa NAD Kinetics 03 i 025n22242658Gov99g 0 00000 nansa nas Time 0o ceansansa nas Time e a Lj Kinetics e ti ao Kinetics e e rm Bs 2 8 m eme B R ceasensou ao Time esaeg2g 02 nao Time e eanaaes 802 n a o Time Memory 500330760 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie
4. should be defined Noise Type Proportional Noise to Signal 0 05 Seed E 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 40 Institut Curie Artificial Networks I Simulation model is defined by a set of ordinary differential equations with either first or second order kinetic rate laws The rates of kinetic equations are randomly selected from the interval Rate Range Rate Range Generated time series is rejected if it exceeds Overflow procedure tries to find the structure and kinetic parameters without the overflow 8 Oct 07 Simulator NETI File Run Help ib nt File Simulate Kinetics Rate Law First Rate Range Overflow imes r Network Structure arr Random Scale Free Time Step 0 01 2 Nodes 20 Out Times Table Permute Rows 0 Noise Type Noise to Signal Seed Proportional 0 05 p NETI 1 2 Manual Copyright C 2004 2007 Institut Curie Artificial Networks II Number of Nodes Network Topology Random Scale Free Random topology any two nodes are connected with the probability p independently of the other connections Scale free topology the number of links at each node is approximated by a power law distribution p k k Novikov E Barillot E Regulatory network reconstruction using an integral additive model with flexible kernel functions submitted to BMC Syst
5. lemory 502249576 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 21 Step by Step Inference Network inference can be performed in the step by step mode using the Step button from the Toolbar or the Menu Item RunlStep F4 At each step the procedure selects the node with lowest fitness and finds another node which can explain the behavior of the given node in the best way ensuring the lowest 7 criterion value for the given node 8 Oct 07 NETI12 D inWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help e o mNX JW m ers 100 Target Model Prior Kernel Background Times Exponent Inverse Polynom 1 0 1 0 0 gt ee L 1 t BESUCHEN MEN LL oe eee eee es Self Regulation Kinetics e e E RIE 2 fa lo la 3 ay P KKKP qu KKK PP Pk z KPas PP K KKPase_P_KK cms KKPase_PP_KK KKPase I PP_KK_K KPase_P_K m7 0 00000 0o02323a3202n1292 Kinetics K_KK KPase PP E Kinetics 181 788 NP Be 4B Se E885 Nt mis 165 363 NP ooa32asanao ooaaa2902n a Time Time Time v Memory 519813072 NETI 1 2 Manual Copyright C 2004 2007 22 Institut Curie Automatic Inference NETI12 D InWork Software Setup WWW_RE demodata MAPKDistorted bin
6. Kinetics o a Kinetics E Kinetics 47 Simulated Data Downloaded NETIT12 Simulated 1 bin File Run Window Help or ae W Fit Iters 100 Target Model Prior Kernel Background Times Exponent Inverse The simulated time series 9 1 0 9 and the corresponding structure have been downloaded in the processing application KKPase PP KK PPKKK Nf phis d NN Hy VN Kinetics Kinetics e e E a Time Time Memory 522614552 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 48 Institut Curie Prior Links from Simulator NETI12 Simulated 1 bin File Run Window Help Hl b mmp p rFit Iters 100 Target Model Prior Links Template From To All links from Simulator 1 Kk KKK are by default assumed as All Used to be unknown I Ade E2 P KKK All they can be included Kinetics for the adaptive model selection using the Menu e ey P ieee Cae Item All Used from the woe RAs in j MAS ANY context menu PRIKK N VS P KKK KK NC DN KPase AS PP K Kinetics Kinetics e e 1uzsascnsm Memory 522614552 8 Oct 07 NETI 1 2 Manual Copyright
7. 4 06 0 27 75 NP 46 E ETTTITIII EETTTITITII toanenanaa Time Time Time Running NETI 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 24 Institut Curie Results The quality of fit can be visually appreciated The resulting network graph 8 Oct 07 Institut Curie D O o are Setup REXd od APKDisto d b m File Run Window Help i d d mM Fit 3 2 Plot Iters Target 1 21 mx Model Prior Sak i Model i Prior Neo Es 005 N Ci sapere Kernel Background Times sum 7 is Exponent Inverse Polynom Ln ST xl 8 1 0 0 oon 4 1 0 0 1095 2 co 00000 C 1 E Ter 0gmae32anae ogmaasnanam cona2asoanaa Time Time Time Self Regulation PAX PK n 3 i a RR 04 A ren com Ac F X 1 mp FA 1 L4 Pk p I K PPS B e fuma Bos a KPase_PP_K KKPase P KK E cons E 02 8 KPase_PP_KK KKPase am xs 4 on KKK KPase_P_K noms a 2 mr eTTTTTETy conagansmnam Mm uwnnananau P_KKK_KK El Time Time Time KPase E2 Ld x i T PPK KKK t3 PK P KKK 2 025 PP pk 1 r1 wid E s s E A E gi HE g ajg FM 46 x2 7 1 266 NP 114 2 a5 4 a am am 47 x2 1 194 NP 116 conaasana 7202229mmgm oos2 32a23anaa v Time Time Time lemory 515937296 NETI 1 2 Manual Copyright C 2004 2007 25 Save Results To save the results use the Save Analysis button
8. P J C Inference Auto Model Randomize v KKPase_PP_KK EN PP KK K tam tamps v S umm v E ums x come ums ional 0 Kinetics ERER EEE 5 Kinetics Kinetics Kinetics Kinetics ao Kinetics Kinetics 0 19a2 72 2 2 82 uu a gg mM Time lemory 495033072 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie Save Statistical Characteristics NETI12 Simulated 2 bin File Run Window Help bob mm W of Prior Template To i The statistical zz dependencies can be EL KKK exported using the Save Analysis button from the Toolbar or the Menu inference v Auto Model Randomize Item FilelSavelAnalysis Ctrl S Kinetics KKPase PP KK EN PP KK K Yr i AR s Kinetics Kinetics Kinetics 0922234 28 u ag Time Memory 495033072 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 33 Institut Curie
9. 18 Initialization Press the Init button from the Toolbar or select the Menu Item RunlInit F3 to initialize the inference The default background behavior will be assigned to each node red line Gray line is the residuals the differences between the experimental and reconstructed time series 8 Oct 07 NETI12 D inWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help i blb mmy W ras met n2 Es Prior VV v Ernel Background Times Kinetics e e Exponent Inverse Polynom 8 1 0 0 y 0 0 1 1 2 3 i comm t0 oO reanananuu reanesanuu Self Regulation Table v oF 0 0y EC EP KC OND OII KAKPel op KI Se CHAR dice RES com Sams x tow 01005 Kinetics 00 onanan Na Kinetics Kinetics 181 788 NP 22 EF SANN Z erred an anasan CoAaDADANAH oo 3 a32a0n a2 Time Time Time v Memory 502249576 NETI 1 2 Manual Copyright C 2004 2007 19 Institut Curie Initialization Interaction Table NETI12 D MnWorkiSoftwareSetupVWWW_RE1demodataMAPKDistorted bin File Run Window Help blo mm DD M Fit Iters 100 Target h n 2 Model Prior Kemel Background Times as i Exponent Inverse Polynom B The interaction table is
10. Data can be downloaded Fle Run Window Help using the Load Data gt W rFit button from the Toolbar i a 3 Y arget or the Menu Item IEF FilelLoadlData I Kernel Background Times Ctrl O Exponent Inverse Polynom See next page for examples of the HJ J input data formats Self Regulation 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 5 Institut Curie Input Data Format Without standard errors El E2 KKK KK K PK X 0 00003 0 00003 0 00003 0 00003 0 00003 0 0003 0 0003 0 0003 0 0003 0 0003 x3 x4 0 003 0 00296 0 00291 0 00287 0 00284 0 00001 0 00002 0 00003 0 00003 D x2 x3 l x4 O 1 0101 2 0202 3 0303 4 0404 1 19743 1 1904 1 17974 1 16613 1 15009 DL uu we lu 7 1 a gent xj l PK KPase_PP_KK tyes 4 0 00253 0 00948 00194 00313 00484 0 000001 00004 00012 00006 p 1 19999 1 19985 1 19931 1 19795 f 149533 y 0 00001 0 00008 200008 000114 00059 p es dn SO ed n odl o a OO 1 1 0 12 012 011997 011985 011955 011897 y 0 0001 0 00013 000015 y 0 f 00005 00016 00046 0 00104 a 0 ee y ui i 4 5 0 1 2 7 1 2 x2 With standard errors SE is indicated in the table cell 1 1 Time x2 SE 2 0202 3 02E 05 2 79E 04 0 00286195 a 2 14E 05
11. Time 0o 0522425258789 Time Kinetics L Kinetics e P p Bs 2 8 R e LJ ao INIIAI Time 0o oeoasasasnsaonao Time 19 00 2249mn8Smm Time v lemory 500330760 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie Inference Model Definition Inverse Model Using the context menu of the Model table the default representation for the kernel function can be selected Inverse kernel function 8 Oct 07 NETI12 D InWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help i d d mm p Fit Iters 100 Target Model Prior Kernel Background Times Exponent Inverse 0 1 0 1 Polynom Kinetics e e Qn Inverse 02z2zzosonm79 Polynom Self Regulatior Differential DA Y o p E EE Kt ce KKPe lp KAKCe OR K EAINNT EER FHERR Kinetics Kinetics 03 nanasa NAD 025n22242658Gov99g 0 00000 nansa nas Time 0o ceansansa nas Time A Sn 7 Kinetics e e rm Bs 2 8 m eme B R 0o oeasasasos ou oo Time esaeg2g 02 nao Time anLZ 052328253877989m9 Time lemory 500330760 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 10 Inference Model Definition Polynomial Model NETI12 D InWork Software Setu
12. format to be able to restore it using the Menu Item FilelLoadlLoad Experiment Ctrl R in the future to reanalyze J 0 10 the data 1 194 NP cs232asenae 1oaDRaODaNAD 0u zs2seoesoncm Time Time Time Kinetics 1 266 NP lemory 515937296 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 2T Institut Curie Set Batch Options NETI12 D InWork Software Setup WWW_RE demodata WMAPKDistorted bin File Run Window Help i amp o mmu Fit ers Target Prior Sow AP fi TIUN Kernel Background Times EK gn UEN E go Ny Exponent Inverse Polynom t z ZUA AA L EL Using the Menu Item i FilelSet Batch Options 1m NEL EU O 0922450mng99 anaana NA oo2 2a202nae all settings can be saved Ur Time Time PRK to be applied to other data i T A sets P KKK P KPase PP K KKPase PP KK Kinetics 000000 090222580859 Time PPIK pra Y Kinetics Kinetics 1 266 NP b aw 0o 1 194 NP H nasara cenaasgagnas o2323a2380ne Time Time Time lemory 515937296 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 28 Institut Curie Adaptive Model Selection 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 29 Prior Links Table Prior links are defined in the Links table
13. C 2004 2007 49 Institut Curie Selected Prior Links from Simulator Prior links can be selected manually using the check box field Link of the Links table Prior links can be selected randomly using the button Randomize the number of selected links is defined by the corresponding spinner box 8 Oct 07 NETI12 Simulated 1 bin File Run Window Help i o e m NEO W Fit Iters 100 el Target 1 2 A Model Prax Links Template z From To E1_KKK KKK E1 KKK E IP KKK E1_KKK lE2 P kkk 7 JJ JJ ul C Inference v Auto Model P_KKK_P_FR KKK PP EK B ey KPase PP K A RKPase_P_KK bo 4 gt ae lt gt Kinetics KKPase_PP_KK PP_KK_K Kinetics am Kinetics 0922328 Nu a2 o m Time lemory 522614552 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 50 Multi Run Simulations NETI12 Simulated 1 bin Using the button Runs File Run Window Help the simulation procedure 9 m i with the follow up processing is repeated Modell Prior 100 times to collect the Tinks Template statistics To Link A different network structure different link parameters different time sampling and different noise realizations may be generated at each
14. L a 2 Piy as there are mo u EE UM arroaren interacting nodes so far Self Regulation ac v 0l EC BR KEID OO OKH KAPOD HI Ded Ri NDS ERI PABA ERE Kinetics i naasna Time The numter of iterations links and the fitness criterion are shown in the f i log window wasesanas 19324582329 Teasnnasanau Time Time Time Memory 502249576 Kinetics Kinetics 181 788 NP 22 EW SANN 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 20 Institut Curie Initialization Network Graph The interaction table can also be presented as a network graph 8 Oct 07 NETI12 D InWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help 6 6 NX We Fit ers Target Model Prior Kernel Background Times Exponent Inverse 1 Polynom 1 d 0 m Self Regulation P kkk p d KKK KPas amp PP K Pase PP KK PP KK K P_KKK_KK KPase PP_K PK pp ek 181 788 NP PP_E amp _ B e kKPase P KK KKPase KPase P K E1 E2 KKK P_KKK p K 22 E Kinetics ere Ae EESE e Kinetics 00 nasaan na Time Kinetics MEL eel 9022258099 00 naaa Time 13 wid s wasi E om 005 y A oog2 3 a20nae Time
15. S tables Kernel and cei ie im NITH Times of the Model PP_KK_K KPase_P_K tab P_KKK_KK El 1 Times table iret re PK P_KKK PP K PI Kinetics Kinetics N j a oe O x2 181 788 NP 22 nara DDIIN Time Time Time lemory 495728256 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 36 Institut Curie Network Reconstruction with the Selected Model NETI12 D InWork Software Setup WWW_RE demodata WMAPKDistorted bin File Run Window Help CECCOLI Target on Sax Bon Zsil c r an P kkk P BM KKK PP EK B fg S aes KPase_PP K KKPase_P_KK com am KKPase_PP_KK KKPase 0 onazoscumom PP KK K KPase P K E Inverse i a Further processing can be c continued in the either tamm tam anaana NA step by step or Time automatic inference a T te mode Self Regulation Kinetics Kinetics P KKK KK El KPase E2 KKK Kinetics Kinetics 1 Qf Lan 0 x2 181 788 NP H nasaan oo2aaasennamu DDIIN Time Time Time lemory 495728256 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 37 Institut Curie Network Simulator 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 38 Run Network Simulator NETI12 D InWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help DX dd
16. from the Prior tab The nodes to be linked are selected from the lists of measured nodes The number of prior links can be increased using the button or can be decreased using the button Only the selected links will be used for the adaptive model selection 8 Oct 07 NETI12 D inWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help 8o o mm D WW Fit Iters 100 Target Model Prior aAA AV AA Links Template From Kinetics oa KPase_PP_K 1 1 2 3 P KKK P KK KPase_PP_K Pase_P_KK HU KKPase PP KK KKPase omi KPase P K M7 10000 onanan Na Kinetics El E2 Esser 0099 waDaADA Nw 0nnxzosomasg Kinetics e e Kinetics 1 194 NP 116 7 O x2 161 786 NP 22 cwanuananma ounnagan au Time Time mLZ DDIIN Time lemory 499585704 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 30 Prior Links Use If the Auto Model checkbox is selected the model selection procedure will be performed at the initialization If the Inference checkbox is selected the prior links will be used not only for the model selection but also will be imbedded in the selected model for final reconstruction 8 Oct 07 NETI12 D inWork Software Setup WWW_RE demodata MAPkDistorted bi
17. from the Toolbar or the Menu Item FilelSavelSave Analysis Ctrl S The results are saved as a list of links in the text file importable into Microsoft Excel 8 Oct 07 NETI12 D InWork Software Setup WWW_RE demodata WMAPKDistorted bin File Run Window Help i e eo mM b M ers Target Prior Kernel Background Times Exponent Inverse Polynom 1 0 1 0 0 op aow CEEE E E nao ooaaangmnao Time Time Self Regulation K pl P KKK I KPase PP K KKPase PP KK i A i y ib WE Bee y IZ a ENTYTIITITITT P KKK KK NS SG Kinetics Kinetics Kinetics 1 266 NP b um 1 194 NP H uu xz2xXocs900s59 CoaDaADANHD Time Time Kinetics e e e e e Mi anaana 0 0o22a230ue Time lemory 515937296 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 26 Save Experiments NETI12 D InWork Software Setup WWW_RE demodata WMAPKDistorted bin File Run Window Help b BNX 5 H2 Kernel Background Times z ge Exponent Inverse Polynom K RE i E The whole experiment 2 la 1 results parameters other amm eS 4N iensesuuas penseaange settings can be saved Self Regulation using the Menu Item FilelSavelSave P_KKK_P_ iE NE 3 Experiment KPas PPK E x KKPase_PP_KK Ctrl W in the internal binary
18. m m Ww 7 Tters Target Fit Model Prior Kernel Background Times _Exponent Inverse Polynom To open Network 1 2 Simulator select the amm tom E oon 0425245mn02m9m Menu Item RunlSimulator E Self Regulation v Kinetics e e BREE v e P Kkk p Kg KKK PP EK B fk KPase PFK FKKPase P KK m EEEE EEEE E KKPase_PP_KK KKPase PP_KK_K KPase_P_K e e P_KKK_KK E1 y v KPase E2 Ul as PP_K KKK PK P_KKK PP c w 12 Tasi L3 Kinetics Kinetics om ae 00 ao 09 naaa Nna E E E E oog2 3 a20nae Time Time Time lemory 495728256 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 39 Institut Curie Network Simulator Window Simulator NETI File Run Help To start simulations press the Run Simulations button from the Toolbar or select the Menu Item RunlRun Simulations F5 De sid To simulate data the following parameters P Wm File Simulate Kinetics Rate Law First Rate Range Overflow Times v 3 100 1 000 r Network Structure Nodes 20 Random Scale Free p Link 0 05 7 Time Step D 01 3 Out Times Table 50 v Permute Rows T Ls Table v
19. to 1005 Target Model Prior a wy e reconstruct the network Links Template E l s FL U 48 Ge t z The number of correctly Polynom rz 2 recovered prior links is ponen 333228344 um 34 WAS counted after the number of Exponent Inverse iterations N D defined by the db number of prior links 2 in Inverse ca this case i If NI is not sufficient to PP KK B Kec 2 7 KPase PF K FKKPase P KK unambiguously identify the KkPase PP KK KKPase best model the NI is Pe os increased by one and the FS P_KKK_KK El D KPase E2 procedure starts again PPK x E PK P_KKK 3 PP kk P E Novikov E Barillot E Model UIT E selection in the reconstruction of 0 x2 181 788 NP 22 ooa2ammnaa nara M 3 3 4 S d 8 regulatory networks from time Time Time Time series data submitted to Algorithms Memory 499505704 for Molecular Biology 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 34 Institut Curie Model Selection Result Kernel The model selection procedure after testing the eight models defined in the Template table selects the best model and copies the identification of this model into the tables Kernel and Times of the Model tab Kernel table 8 Oct 07 D O oftware Setup RE demodata WAPkKDistorted b a File Run Window Help i o o mm D W
20. 057275 0 11357458 o 0052 1 23317372 6 5E 6 2 21276279 1 10000000 2 25559811 1 3500 4 78586511 2 30000000 0 00514494 2 72000000 0 01561012 7 6850t 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 5 32821062 3 0E 6 6 40613020 2 9500t 0 0 1 0 2 11799427 1 00000000 3 71790448 2 0000C 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 6 81570453 3 5E 6 6 12453913 2 9999t 0 0 1 0 0 0 1 0 0 0 1 0 2 93667162 1 5E 6 7 56427929 4 00000000 8 97975389 5 0E 6 2 85189337 1 35000000 2 18929163 1 10000000 184055799 9 5E 6 4 30187251 2 25E 5 0 00577453 2 69000000 0 01566717 7 615E Model Prior Kernel Background Times Exponent Intensity values with the corresponding standard errors SE aK iia E SE column can be hidden using the SE button from the local toolbar i 2828882865 E EE Kt ce KKPe lp KAKCeIOR HI Ded ORI ND PR ER IRR TR UU 3 uv 3 s 5 X 3 Y E z 8E l z 2 g Memory 500330760 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 7 Institut Curie Plot View The number of the graph columns and the height of the graph rows can be adjusted using the corresponding controls 8 Oct 07 NETI12 D inWork Software Setup WWW_RE demodata MAPkDistorted b
21. 6 K 1 21922989 E 0 00993045 E 9 74E 06 K j P 1 18169628 0 0599925 1 14199093 0 057234 1 0437543 0 0529535 1 2 K 0 05952 1 06178469 0 052231 1 00851474 0 048417 K 1 45E 04 0 00250212 1 32E 04 0 00260262 1 28E 04 O ay 7 l gt PK 8 16E 05 0 oo 849E04 420005 000607996 3 06E04 KPase PPKK 33 o ou o J ou o oo 0 0447 7 Sun o n EE 0 0n ee Vy 4 4 1 00E 06 3 50E 06 9 56E 05 4 50E 06 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie Table View F NETI12 D InWork Software Setup WWW_RE demodata MAPKDistorted bin BAR File Run Window Help i S d mmp pj Fit meam Table v Iters 1005 Target er i t 5 SE x11 8 1 11111 41 1111 09828184 1 5E 6 3 12373743 1 5E 6 398 3 11788321 1 49999999 3 37839618 1 4999 87 11 42E 4 0 00262817 1 35E 4 0 00248207 1 305E 736785888 1 5E 6 6 44455290 3 0E 6 7 81802572 4 0000C 1 09694903 0 05830650 J1 21132341 0 05454449 1 01075772 0 0509 0 03432677 0 0016565 0 10067342 0 0050945 0 17432846 0 00797 1 30193776 6 0E 6 0 00139286 6 95E 5 0 00424773 2 055E 1 24208287 0 05989750 1 02269297 0 05873800 1 17772457 0 0561 0 00106066 5 7E 5 0 01391098 7 06E 4 0 04633620 0 0021 0 0 1 0 1 63421023 8 50000000 0 00152726 8 5E 5 0 12199318 0 0059775 0 11166633 0 0
22. Functions NETI12 D InWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help i d d mm WW Target Polynom The background functions Kinetics e e can be defined similar to tomm mm onnnananan the kernel functions ERN E E Kt ce KKPe lp KAKCe DR HI Ded ORDER ERE CABINE ERR Kinetics Kinetics aa 0 an ETT 3 02 xgs anj 06 b 0522425258589 Time amLz 052328253877989m9 Time lemory 500330760 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie Inference Model Definition Characteristic Times Characteristic times for the Exponential and Inverse kernel functions can be set using the corresponding spinner boxes The characteristic times are defined as a 7 where a is taken from the table and T is the measurement time range The number of times can be increased using the button or can be decreased using the button 8 Oct 07 NETI12 D nWork Software Setup WWW_RE demodata MAPKDistorted bin File Run Window Help Ek i old N iN Fit ters 100 2 J Target Model Prior Kernel Background Times Inverse 0 9 0 2 c E mu TP Exponent Self Regulation e ae KK gs zR m E ies d ta pace ase ET e E EB KXiCe HIP OH KPOP FI
23. GWIFREN HA WET Y Kinetics 09noz22587m3m9 Kinetics 03 esasesonaga Time om conaanonam Time 0o 0522425258889 Time Kinetics L Kinetics e P p es 2 8 R e LJ ao ooanaaa rs w Time 0o eoasasasonao Time Time 19 coenssaseneao v lemory 500330760 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 16 Inference Model Definition Characteristic Times Selection 9 NETI12 D InWork Software Setup WWW_RE demodata MAPKDistorted bin DER File Run Window Help i S d mH Hw Fit Iters 1 00 2 Target Model Prior Kernel Background Times Exponent Inverse Kinetics e e The number of characteristic times should not be smaller than the number of terms in the kernel functions 000 it e2asoeseonam oesaa3a22 ne 025n22242658Gov99g EC E KEROKI OKH KPOP Hi Ded ORI EER NBI TR Kinetics If the number of characteristic times is larger than the number of kernel K H terms all possible combinations of Hd idi U rUVazxasanuu 5235498287 characteristic times are a consequently substituted in the kernel function and fitting is performed for each tested link The combination ensuring the best fitness value 32 is preserved for this link 03 Kinetics e p rm gd m m e B Kinetics Uu
24. NETI Network Inference Version 1 2 User Manual Copyright C 2004 2007 Institute Curie All rights reserved NETI download page Author s Eugene Novikov Institut Curie E mail 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie System Requirements NETI is written in Java and C It runs on Windows platforms 95 98 Me NT 2000 XP and needs the Java Runtime Environment JRE to be installed 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie Installation NETI can be downloaded from the NETI download page Click NETI Setup 1 2 exe to start the NETI 1 2 installer and follow the instructions NETI 1 2 installation creates a Curie NETI 1 2 folder in the list of Programs of the Windows Start menu This new folder contains the following entries NETI 1 2 starts Network inference software User Manual is a user manual pdf file Uninstall NETI will remove NETI from your computer Installation procedure may also create a NETI 1 2 icon on your Desktop Installation procedure asks about the default size of the JVM Java Virtual Machine memory allocation pool It is recommended to set it as large as possible but not larger than the amount of available RAM 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 3 Institut Curie Network Inference 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie Processing Window NETI12 Experiment 1 bin
25. e kernel function can be A Differential selected El e Ee Ct HOH OFICOFIP KAPOD Hb EIKT BEER RRR E First order linear H us M ordinary differential i uen ease NAD Ma aaa WWE equations i Z 12 100 Sas E d 5 02 E ET EEEIEE AN Y Time Time Time lemory 500330760 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 12 Institut Curie Inference Model Definition Exponential Model Using the popup menu of the Model table the default representation for the kernel function can be selected Exponential kernel function 8 Oct 07 NETI12 D InWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help i oeb mH W Target Fit Iters 100 3 Model Prior Kernel Background Times Exponent Polynom 0 0 Inverse Exponent Kinetics e e 000000 0ux2x2o059680mgm9 Self Regulation Inverse Polynom Differential X 4 of E E KR OO OIH HD OI K Ded ORI ND GPS ER PABBA TR Kinetics 03 0o nanasa NAD a oenaam a 000000 nanasa nao Time ceansansa nas Time A Sn 7 e e e a Kinetics e ti Kinetics e e rm Bs 2 8 m eme B R um Coon naDaDR RMA Time esaeg2g 02 nao Time anLZ 052328253877989m9 Time Memory 500330760 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 13 Inference Model De
26. ems Biology 8 Oct 07 3 Simulator NETI SEE File Run Help bu 2 File ate Ls Table v Kinetics n Metwork Structure aelaw First v Nodes Rate Range 1 Random Scale Free v Y Overflow 100 p Link TH Times 1 000 Time Step D 01 a a4 E Out Times sois Permute Rows Table oz Noise Type Proportional Noise to Signal 0 05 Seed i NETI 1 2 Manual Copyright C 2004 2007 42 Institut Curie Artificial Networks III Simulator NETI Number of Time Steps Fle Run Help Times and Time Step to i m File Simulate generate idealistic time C Kinetics gt Network Structure series A RatPww First v Nodes 201 Rate Range Random Scale Free Overflow 0 05 3 Out Times defines i A J sampling frequency mes 1 0002 TmeStep 0017 In this example the original Out Times 502 Permute Rows 1000 point time series are Table Y converted into 20 intervals of 50 points At each interval the output time point is randomly selected Generation of the permuted data i e when node labels are randomly assigned to generated Noise Type Proportional tim es eri es Noise to Signal 0 05 Seed al 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 43 Institut Curie Additive Statistical Noise Model for the standard deviation of the additive noise It can be constant proportional
27. finition Power Factors Power factors for any term of any kernel function can be adjusted using the corresponding spinner boxes If the power factors differ from zero for two or three functions these functions are combined in a product For example the kernel function defined by the given combination of the power factors takes the form w t m expt rc bx H rie YF Tus expt tta x j Novikov E Barillot E Model selection in the reconstruction of regulatory networks from time series data submitted to Algorithms for Molecular Biology 8 Oct 07 NETI12 D InWork Software Setup WWW_RE demodata MAPkDistorted bin File Run Window Help EJP b BN Ww Fit Iters 100 4 Target gt m ModeMspgior Kernel Backgrourt i es Exponent 0 3 Inverse 1 6 Polynom xL D o 15 Self Regulation v ay EC EE Kt OO OKH HD OIDP H Ded ORDER ERI PHBA TR Ki tics e e ood 0005 tom 22a458mm0mg Time xx e m gd os e B Kinetics 03 050mz33a4532887 889ms Time nasasanay 0o ceansansa nas Time Kinetics Kinetics e rm gd 8S m e B 02 am gt 0o22a498mmm Time 10 eanaaes 802 n a o Time Memory 526075936 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 14 Inference Model Definition Background
28. in File Run Window Help Aolo mmy p Fit tes 1008 Target BEND Model Prior Kernel Background Times Exponent Polynom co lt gt Kinetics Kinetics Kinetics e e 0nx22zxz2o098n0mgm90 Kinetics nananana tom Coa DaDA NH aa 0s an EIE E 02 xgs anj 16 Kinetics b 0522425258589 Time S wid 8 anos E Xa 10 0522325898989 00 ceasaensou oo Time 10 oessanseo nu oo Time Q9 00 2249mn8Smm Time v Memory 500330760 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie Inference Model Definition Number of Terms The checkbox Self Regulation allows for self feedback loops Number of terms in the kernel function The number of terms can be increased using the button or can be decreased using the button Novikov E Barillot E Model selection in the reconstruction of regulatory networks from time series data submitted to Algorithms for Molecular Biology 8 Oct 07 NETI12 D inWork Software Setup WWW_RE demodata MAPKDistorted bin File Run Window Help i Sd d mH Hw 1002 Target Model Prior Kernel Background Tithe Exponent Kinetics 09noz2587msm Kinetics 03 esasesonaga Time om conaa3anaoo
29. n File Run Window Help K m y gt w Fit Iters 100 gt Target Nodel Prior Lin Template From To KKK E1_KKK KPase PP K Inference Auto Model Randomize P kkk p fd KKK KPase PP K KKPase PP KK PP KK K P KKK KK KPase PP_K PK pp 1 194 NP 181 788 NP Graph v PP B e KKPase P KK KKPase KPase P K El E2 KKK 116 22 Kinetics tmm anaana NA Kinetics e e Kinetics n P E E Time Kinetics ConDADAN AH Time w 020 tc vis E Xa a mLZ DDIIN Time lemory 499585704 NETI 1 2 Manual Copyright C 2004 2007 Institut Curie 3l Models Library NETI12 D InWork Software Setup WWW_RE demodata WMAPKDistorted bin File Run Window Help ie o mM M rFit ters Target Model Prior yd 4 Template Model The models to be tested zx are defined in the Exponent Exponent Template table from Exponent Inverse the Prior tab Inverse Inverse Kinetics P kkk p Rai FKK K KPase PP K FKKPase P KK om KKPase_PP_KK KKPase nanana PP_KK_K KPase_P_K P_KKK_KK E1 Ly v KPase E2 apa ware E Ul E PPK mf yo PK P_KKK z z zm pp f E E 2 a Novikov E Barillot E Model i i S pna T selectio
30. n in the reconstruction of 7 181 788 NP 22 EETITITITI Mm azsasanas Ui uzasosanas regulatory networks from time Time Time Time series data submitted to Algorithms emory 499585704 for Molecular Biology 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 32 Institut Curie Models Template NETI12 D InWork Software Setup WWW_RE demodata WMAPKDistorted bin The model is selected Fie Run Window Help x from the three iW e e m mx Fit implemented models 00 O ED Exponential Inverse E Se Polynomial Links Template s We T YgModel E i 7 Fo Polynom E UN The number of terms for o s x ms BEE SSESSSESES each model is defined by Polynom nananana UL Uasesansa the number of Power em Factors separated by gwen spaces E 1 T PP The corresponding PKA HEB HK ok characteristic times a 4 o KK ase PK should also be separated P KKK KK E1 by spaces KPase E2 KKK 8 Only the selected models l i EN will be used in testing J N P O x2 181 788 NP 22 ooa2ammnaa nara DDIIN Time Time Time lemory 499585704 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 33 Institut Curie Adaptive Model Selection 5 NETI12 D InWork Software Setup WWW_RE demodata MAPKDistorted bin At initialization the FS File Run Window Help procedure uses e mx Ww consequently each of the m defined models
31. p WWW_RE demodata MAPkDistorted bin File Run Window Help H oldm HN W Fit Iters 100 3 Target Model Prior Kernel Background Times 3 Exponent Inverse Polynom i Using the popup menu of nmm 1 the Model table the Inverse y 0000 00000 iit PA 4 Polynom oe22asose0nao coes2aaa2280 nam cen2asanso nao default representation for GA ee the kernel function can be 3 T selected EOE EB KEE OD OHP KKP HEP KIP SPH AT ERO FHE ERS 8 Polynomial kernel us gt K i b UM function err rr rr ee i ui CEF Sr SE E E E Pase pP KK Time Time A m Pek T Fes rke 12 Mopase KK ase PK un 2 i fw i a 0a ao ao 19 025a2220252588285w 052258655958nnvymm9 05235850959588m7238m89 Time Time Time lemory 500330760 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 11 Institut Curie Inference Model Definition Differential Model NETI12 D inWork Software Setup WWW_RE demodata WMAPkKDistorted bin File Run Window Help i be M NY Fit Iters M Target Model Prior Kernel Background Times 4 Exponent Inverse Polynom E Using the popup menu of J the Model table the i tom i Inverse eao2saseon ao 0252225235G6570829 052348502588a7989m9 default representation for APR th
32. run If the properties of the network is x Wr FEB Re P_KK defined by the external file e g CN ASS IN KKPase SBML network structure kinetic laws and kinetic parameters remained unchanged ER ER ER n Kinetics 05 Kinetics Kinetics Kinetics e e Different prior links are also generated at each run The number of prior links is defined by the corresponding spinner box e a uvnxozossso0nsSsms Time lemory 522614552 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 51 Institut Curie Measured Statistical Characteristics The collected statistical characteristics are the averaged dependencies on the total number of links of ey criterion Positive Predictive Value PPV TP TP FP Sensitivity Se TPHTP FN The results in the log window The selected models at each run model numbering corresponds to the Template table Novikov E Barillot E Model selection in the reconstruction of regulatory networks from time series data submitted to Algorithms for Molecular Biology 8 Oct 07 NETI12 Simulated 2 bin File Run Window Help ei So EN W Fit Iters Model Prior Links Template To E1_KKK KKK E1_KKK E1 IP KKK E1_KKK E2 P KKK bpoooooo wv
33. to signal or proportional to the square root of signal Noise to signal level for the additive statistical noise This noise is finally added to each data point Seed for random number generator selection 1 as a seed will initiate the random generator with automatically or randomly chosen seed 8 Oct 07 Simulator NETI File Run Help i m File Simulate DER ES Table v r Kinetics r Network Structure Rate Law First v Nodes 20 Rate Range ig Random Scale Free Overflow 100 2 p Link 0 05 3 Times 1 000 Time Step D 01 50 Permute Rows Noble og v a Out Times Noise Type Proportional ise Lo Signal 0 05 Seed 1 NETI 1 2 Manual Copyright C 2004 2007 44 Institut Curie SBML Modules I Simulator NETI DER Simulator can import Fie Run Heb some intermediate data i E typically time series Fle simulate E IL EET generated by SBML RICH modules iia XML The structure of the network can also be imported It allows to OutTmes soc PermiteRows compare the structure Table o used in data generation with the structures obtained by the inference algorithm The structure can be defined by the xml or txt files Noise Type Proportional Noise to Signal 0 05 Seed 1 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 45 Institut Curie SBML Mod
34. ules II 7 Simulator NETI Imported model MAPK File Run Help ibe 9 D InWork Software Setup PK_BIOMDO000000009 xml Om OTT oats 55 J LKK PP_KK PaK iy N W h KKPase_P_KK iN INN N aN KKPase Idealistic time series generated in JDesigner These can be further a s Sel distorted by non s gon Xs homogeneous time a 0s loise to Signal 000 aoo 000 sampling and adding a statistical noise Emulation finished 8 Oct 07 NETI 1 2 Manual Copyright C 2004 2007 46 Institut Curie Export Data To send the generated data in the Processing Window use the Send Data button from the Toolbar or the Menu Item FilelSend Data Alt To save the generated data in the text file use the Menu Item FilelSave Data Ctrl Shift O To save the generated structure in the text file use the Menu Item FilelSave Data Ctrl Shift U 8 Oct 07 Simulator NETI Fie Run Help A gt m 9 E Simulate 2 XML ote D InWork Software Setup Structure PK BIOMDODDO000009 xml P KKK P Kl KPase PP K ase PP KK Jl y AS iv P KKK KK NN i ao NA Graph v 08 j pum Type j Proportional Noise to Signal Seed Simulation finished NETI 1 2 Manual Copyright C 2004 2007 Institut Curie
Download Pdf Manuals
Related Search