Cellularstatetransitions(i.e.frompluripotenttocommitted,replicativetoquiescent,etc.)requirethe coordinatedregulationofthousandsofgenes.Therapeuticallyharnessingthesetransitionsholdsgreat promiseforhumanhealth;?forinstance,autologousstemcelltherapyhasbeensuccessfullyusedin regenerativemedicineandcancertreatments,amongothers.Whilesomeofthekeyregulatoryswitchesare known,thefieldlacksasystems-levelunderstandingofthegenomiccircuitsthatcontrolthesetransitions, informationthatiscriticalforinformedclinicalintervention.Here,wewilldevelopanintegratedcomputational frameworktoidentifycoregeneregulatorycircuitsfromlargegenenetworksandpredicttheirdynamicsand regulatoryfunctionswithouttheneedofdetailednetworkkineticparameters.Advancesingenomicsprofiling technologyhaveenabledthemappingofgeneregulatorynetworks,thuswenowhavethecapacitytoidentify combinatorialinteractionsamonggenesandthemasterregulatorsofstatetransitions.Somesystemsbiology approacheshavesimulatedthedynamicsofageneregulatorycircuit,buttraditionalmethodssufferfromtwo keyissues.First,thereisnorationalruletochooseanappropriatesetofregulatorgenesinalargenetworkto model.Second,sinceitishardtodirectlymeasuremostnetworkkineticparametersfromexperiment, modelingresultsarebasedonasetofguessedparametersthatcanbelessthanoptimal,limitingthe applicationofmathematicalmodelingtolargesystemsandthepredictionpowerofsystemsbiology.To addresstheseissues,werecentlydevelopedalgorithmnamedrandomcircuitperturbation(RACIPE).RACIPE generatesanensembleofcircuitmodels,eachofwhichcorrespondstoadistinctsetofrandomkinetic parameters,anduniquelyidentifiesrobustfeatures,suchasclustersofstablegeneexpressionstates,by statisticalanalysis.WewillfurtherenhanceRACIPEalgorithmsforlargesystemsandnewdataanalysistools usingmachinelearning.Thisapproachwillconvertatraditionalnonlineardynamicsproblemintoadata analysisproblem,anessentialstepforextendingtheapplicationofgenecircuitmodelingtolargesystems.It alsoprovidesanovelstrategytointegratetop-downgenomicsapproacheswithbottom-upmathematical modeling.Thealgorithmswillbetestedandrefinedusingliterature-basedgenenetworks,publicgenomics data,anddatafromcollaboration,withafocusoncelldifferentiationindevelopmentalprocessesandstate transitionsinoncogenesis.Successoftheprojectwillresultinacomprehensivetoolkitthatwillunveilthegene regulatorymechanismofcellulardecision-makinginanycellofinterest.Thealgorithmicdevelopmentis expectedtohaveabroadimpactonnotonlybasicresearchinsystemsbiologybutalsoshedlighton therapeuticinterventioningenomicmedicine.
TOPUBLICHEALTH Mosthumancomplexdiseasesaremultigenic,butourunderstandingoftheinteractionsbetweengenesin healthystatesandhowtheyaredisruptedindiseasestatesislacking.Computationalalgorithmswillbe designedtoconstructandmodelgeneregulatorycircuitsofhighercomplexitytofillanimportantgapinour understandingofthemaintenanceandtransitionsamongcellularstatesandtheirabnormalbehaviorin disease.Inthisproposal,wewilldevelopasuiteofsystems-biologytoolstoextendgenecircuitmodelingto largesystemsandintegrategenomicsdatarelevanttofundamentalcellbiology,cancer,andotherhuman diseases.
| Kohar, Vivek; Lu, Mingyang (2018) Role of noise and parametric variation in the dynamics of gene regulatory circuits. NPJ Syst Biol Appl 4:40 |
