spring 2017 – epigenetics and systems biology lecture ...2014) systems-based technologies in...

Spring2017–EpigeneticsandSystemsBiologyLectureOutline–SystemsBiologyMichaelK.Skinner–Biol476/576CUE418,10:35-11:50am,Tuesdays&ThursdaysJanuary10,12,17&19,2017Weeks1and2

SystemsBiology

• HistoryandDefinitions• Reductionism/GeneticDetermination• Holism/Emergentism/HomeostasisorRobustness• RevolutionaryandEvolutionarySystemsBiology• NetworksandComputationalBiology• BasicMolecularandCellularComponents

RequiredReading

AntonyPM,BallingR,VlassisN.(2012)Fromsystemsbiologytosystemsbiomedicine.CurrOpinBiotechnol.23(4):604-8.Knepperetal.(2014)Systemsbiologyversusreductionismincellphysiology.AmJPhysiolCellPhisiol307:C308-C309.

BackgroundBookReferences

JamesA.Marcum(2009)TheConceptualFoundationsofSystemsBiology,NovaSciencePublishers,Inc.EberhardVoit(2012)AFirstCourseinSystemsBiology,GarlandScienceCapraandLuisi(2014)TheSystemsViewofLife,CambridgeUniversityPress.

LiteratureNobileMS,CazzanigaP,TangherloniA,BesozziD.(2016)Graphicsprocessingunitsin

bioinformatics,computationalbiologyandsystemsbiology.BriefBioinform.Jul8.pii:bbw058.[Epubaheadofprint]

TuraevD,RatteiT.(2016)Highdefinitionforsystemsbiologyofmicrobialcommunities:metagenomicsgetsgenome-centricandstrain-resolved.CurrOpinBiotechnol.Jun;39:174-81.

NiuF,WangDC,LuJ,WuW,WangX.(2016)Potentialsofsingle-cellbiologyinidentificationandvalidationofdiseasebiomarkers.JCellMolMed.20(9):1789-95

PettaS,ValentiL,BugianesiE,TargherG,BellentaniS,BoninoF;SpecialInterestGrouponPersonalisedHepatologyoftheItalianAssociationfortheStudyoftheLiver(AISF);SpecialInterestGrouponPersonalisedHepatologyofItalianAssociationforStudyofLiverAISF.(2016)A"systemsmedicine"approachtothestudyofnon-alcoholicfattyliverdisease.DigLiverDis.48(3):333-42.

BouchardC.(2015)AdaptationtoAcuteandRegularExercise:FromReductionistApproachestoIntegrativeBiology.ProgMolBiolTranslSci.135:1-15.

deVargasRoditiL,ClaassenM.(2015)Computationalandexperimentalsinglecellbiologytechniquesforthedefinitionofcelltypeheterogeneity,interplayandintracellulardynamics.CurrOpinBiotechnol.34:9-15.

BennerSA,KaralkarNB,HoshikaS,LaosR,ShawRW,MatsuuraM,FajardoD,MoussatcheP.(2016)AlternativeWatson-CrickSyntheticGeneticSystems.ColdSpringHarbPerspectBiol.1;8(11).

NemhauserJL,ToriiKU.(2016)Plantsyntheticbiologyformolecularengineeringofsignallinganddevelopment.NatPlants.2;2:16010.

ShahE.(2016)Ataleoftwobiographies:themythandtruthofBarbaraMcClintock.HistPhilosLifeSci.38(4):18.

PezzuloG,LevinM.(2016)Top-downmodelsinbiology:explanationandcontrolofcomplexlivingsystemsabovethemolecularlevel.JRSocInterface.13(124).pii:20160555.

KesićS.(2016)Systemsbiology,emergenceandantireductionism.SaudiJBiolSci.23(5):584-91.MargineanuDG.(2016)Neuropharmacologybeyondreductionism-Alikelyprospect.Biosystems.

141:1-9.LoorJJ,Vailati-RiboniM,McCannJC,ZhouZ,BionazM.(2015)TRIENNIALLACTATION

SYMPOSIUM:Nutrigenomicsinlivestock:Systemsbiologymeetsnutrition.JAnimSci.93(12):5554-74.

ListaS,KhachaturianZS,RujescuD,GaraciF,DuboisB,HampelH.(2016)ApplicationofSystemsTheoryinLongitudinalStudiesontheOriginandProgressionofAlzheimer'sDisease.MethodsMolBiol.1303:49-67.

HilbertA.(2016)WeightStigmaReductionandGeneticDeterminism.PLoSOne.15;11(9):e0162993.

CasaneD,RétauxS.(2016)EvolutionaryGeneticsoftheCavefishAstyanaxmexicanus.AdvGenet.95:117-59.

Arribas-AyllonM.(2016)Aftergeneticization.SocSciMed.159:132-9.BuecherB.(2016)Colorectaladenomatouspolyposissyndromes:Geneticdeterminism,clinical

presentationandrecommendationsforcare.BullCancer.2016Feb;103(2):199-209.ScheidV.(2016)Holism,ChineseMedicineandSystemsIdeologies:RewritingthePasttoImagine

theFuture.In:AtkinsonS,MacnaughtonJ,RichardsJ,authors;WhiteheadA,WoodsA,editors.TheEdinburghCompaniontotheCriticalMedicalHumanities.Edinburgh(UK):EdinburghUniversityPress;2016Jun.Chapter3.WellcomeTrust–FundedMonographsandBookChapters.

VogtH,HofmannB,GetzL.(2016)Thenewholism:P4systemsmedicineandthemedicalizationofhealthandlifeitself.MedHealthCarePhilos.19(2):307-23.

TordayJS.(2015)Acentraltheoryofbiology.MedHypotheses.85(1):49-57.Díaz-MuñozSL,BoddyAM,DantasG,WatersCM,BronsteinJL.(2016)Contextualorganismality:

Beyondpatterntoprocessintheemergenceoforganisms.Evolution.70(12):2669-2677.

Alanis-LobatoG,MierP,Andrade-NavarroMA.(2016)EfficientembeddingofcomplexnetworkstohyperbolicspaceviatheirLaplacian.SciRep.22;6:30108.

ScarpaJR,JiangP,LosicB,etal.(2016)SystemsGeneticAnalysesHighlightaTGFβ-FOXO3DependentStriatalAstrocyteNetworkConservedacrossSpeciesandAssociatedwithStress,Sleep,andHuntington'sDisease.PLoSGenet.8;12(7):e1006137.

AdachiN,SendaT,HorikoshiM.(2016)Uncoveringancienttranscriptionsystemswithanovelevolutionaryindicator.SciRep.16;6:27922.

VersluisF,vanEschJH,EelkemaR.(2016)SyntheticSelf-AssembledMaterialsinBiologicalEnvironments.AdvMater.28(23):4576-92.

IgamberdievAU,Shklovskiy-KordiNE.(2016)Computationalpowerandgenerativecapacityofgeneticsystems.Biosystems.2016Apr-May;142-143:1-8.

KaufmanCK,MosimannC,FanZP,etal.(2016)Azebrafishmelanomamodelrevealsemergenceofneuralcrestidentityduringmelanomainitiation.Science.29;351(6272):aad2197.

OlivierBG,SwatMJ,MonéMJ.(2016)ModelingandSimulationTools:FromSystemsBiologytoSystemsMedicine.MethodsMolBiol.1386:441-63.

AmirkhahR,FarazmandA,WolkenhauerO,SchmitzU.(2016)RNASystemsBiologyforCancer:FromDiagnosistoTherapy.MethodsMolBiol.1386:305-30.

ChenJY,ShenQS,ZhouWZ,etal.(2015)Emergence,RetentionandSelection:ATrilogyofOriginationforFunctionalDeNovoProteinsfromAncestralLncRNAsinPrimates.PLoSGenet.15;11(7):e1005391.

Eloundou-MbebiJM,KükenA,OmranianN,etal.(2016)Anetworkpropertynecessaryforconcentrationrobustness.NatCommun.19;7:13255.

VickovicS,StåhlPL,SalménF,etal.(2016)Massiveandparallelexpressionprofilingusingmicroarrayedsingle-cellsequencing.NatCommun.14;7:13182.

HeF,MurabitoE,WesterhoffHV.(2016)Syntheticbiologyandregulatorynetworks:wheremetabolicsystemsbiologymeetscontrolengineering.JRSocInterface.13(117).pii:20151046.

MestekBoukhibarL,BarkoulasM.(2016)Thedevelopmentalgeneticsofbiologicalrobustness.AnnBot.117(5):699-707.

LiS,TodorA,LuoR.(2015)Bloodtranscriptomicsandmetabolomicsforpersonalizedmedicine.ComputStructBiotechnolJ.31;14:1-7.

BeckS,LeeBK,KimJ.(2014)Multi-layeredglobalgeneregulationinmouseembryonicstemcells.CellMolLifeSci.Sep17.[Epubaheadofprint].DOI10.1007/s00018-014-1734-9

WylesSP,FaustinoRS,LiX,TerzicA,NelsonTJ.(2014)Systems-BasedTechnologiesinProfilingtheStemCellMolecularFrameworkforCardioregenerativeMedicine.StemCellRev.Sep14.[Epubaheadofprint].DOI10.1007/s12015-014-9557-5

BaslanT,HicksJ.(2014)Singlecellsequencingapproachesforcomplexbiologicalsystems.CurrOpinGenetDev.10;26C:59-65.

PathakRR,DavéV.(2014)Integratingomicstechnologiestostudypulmonaryphysiologyandpathologyatthesystemslevel.CellPhysiolBiochem.33(5):1239-60.

ConwayO'BrienE,PrideauxS,ChevassutT.(2014)Theepigeneticlandscapeofacutemyeloidleukemia.AdvHematol.2014:103175.

SchlichtingCD,WundMA(2014)Phenotypicplasticityandepigeneticmarking:anassessmentofevidenceforgeneticaccommodation.Evolution.Mar;68(3):656-72.

WoodJG,HelfandSL.(2013)Chromatinstructureandtransposableelementsinorganismalaging.FrontGenet.4;4:274.

MasonCE,PorterSG,SmithTM.(2014)Characterizingmulti-omicdatainsystemsbiology.AdvExpMedBiol.799:15-38.

RosaA,BrivanlouAH.(2013)Regulatorynon-codingRNAsinpluripotentstemcells.IntJMolSci.11;14(7):14346-73.

ClarkeJ,PenasC,PastoriC,etal.(2013)Epigeneticpathwaysandglioblastomatreatment.Epigenetics.8(8):785-95.

KyrtopoulosSA.(2013)MakingsenseofOMICSdatainpopulation-basedenvironmentalhealthstudies.EnvironMolMutagen.54(7):468-79.

LovejoyDA,Barsyte-LovejoyD.(2013)Systemsapproachestogenomicandepigeneticinter-regulationofpeptidehormonesinstressandreproduction.ProgBiophysMolBiol.113(3):375-86.

FrebourgT.(2014)Thechallengeforthenextgenerationofmedicalgeneticists.HumMutat.35(8):909-11.

GeorgesA,CambisanoN,AharizN,KarimL,GeorgesM.(2013)Agenomescanconductedinamultigenerationalpedigreewithconvergentstrabismussupportsacomplexgeneticdeterminism.PLoSOne.23;8(12):e83574.

DiGiulioM.(2013)Theoriginofthegeneticcode:matterofmetabolismorphysicochemicaldeterminism?JMolEvol.77(4):131-3.

ParrottR,SmithRA.(2014)Defininggenesusing"blueprint"versus"instruction"metaphors:effectsforgeneticdeterminism,responseefficacy,andperceivedcontrol.HealthCommun.29(2):137-46.

YanQ.(2014)Frompharmacogenomicsandsystemsbiologytopersonalizedcare:aframeworkofsystemsanddynamicalmedicine.MethodsMolBiol.1175:3-17.

TobinAB,PrihandokoR.(2014)Replyto"Lettertotheeditor:'Systemsbiologyversusreductionismincellphysiology'".AmJPhysiolCellPhysiol.1;307(3):C310.

KnepperMA,RaghuramV,BradfordD,etal.(2014)Lettertotheeditor:"Systemsbiologyversusreductionismincellphysiology".AmJPhysiolCellPhysiol.1;307(3):C308-9.

WinquistRJ,MullaneK,WilliamsM.(2014)Thefallandriseofpharmacology--(re-)definingthediscipline?BiochemPharmacol.1;87(1):4-24.

BoseB.(2013)Systemsbiology:abiologist'sviewpoint.ProgBiophysMolBiol.113(3):358-68.WolkenhauerO,GreenS.(2013)Thesearchfororganizingprinciplesasacureagainst

reductionisminsystemsmedicine.FEBSJ.280(23):5938-48.GareA.(2013)OvercomingtheNewtonianparadigm:theunfinishedprojectoftheoreticalbiology

fromaSchellingianperspective.ProgBiophysMolBiol.113(1):5-24.VogtH,UlvestadE,EriksenTE,GetzL.(2014)Gettingpersonal:cansystemsmedicineintegrate

scientificandhumanisticconceptionsofthepatient?JEvalClinPract.Oct14.doi:10.1111/jep.12251.[Epubaheadofprint]

WesterhoffHV,BrooksAN,SimeonidisE,etal.(2014)Macromolecularnetworksandintelligenceinmicroorganisms.FrontMicrobiol.22;5:379.

WernerHM,MillsGB,RamPT.(2014)CancerSystemsBiology:apeekintothefutureofpatientcare?NatRevClinOncol.11(3):167-76.

KompanichenkoV.(2014)Emergenceofbiologicalorganizationthroughthermodynamicinversion.FrontBiosci(EliteEd).1;6:208-24.

GoransonHT,CardierB.(2013)Atwo-sortedlogicforstructurallymodelingsystems.ProgBiophysMolBiol.113(1):141-78.

Fernandez-LeonJA.(2014)Robustnessasanon-localizablerelationalphenomenon.BiolRevCambPhilosSoc.89(3):552-67.

GutiérrezJ,MaereS.(2014)Modelingtheevolutionofmolecularsystemsfromamechanisticperspective.TrendsPlantSci.19(5):292-303.

GohT,VoβU,FarcotE,BennettMJ,BishoppA.(2014)Systemsbiologyapproachestounderstandtheroleofauxininrootgrowthanddevelopment.PhysiolPlant.151(1):73-82.

AgarwalS.(2013)Systemsapproachesinunderstandingevolutionandevolvability.ProgBiophysMolBiol.113(3):369-74.

KlinmanJP,KohenA.(2014)EvolutionaryAspectsofEnzymeDynamics.JBiolChem.31;289(44):30205-30212.

WeinreichDM,LanY,WylieCS,HeckendornRB.(2013)Shouldevolutionarygeneticistsworryabouthigher-orderepistasis?CurrOpinGenetDev.23(6):700-7.

BarrickJE,LenskiRE.(2013)Genomedynamicsduringexperimentalevolution.NatRevGenet.14(12):827-39.

SoyerOS,O'MalleyMA.(2013)Evolutionarysystemsbiology:whatitisandwhyitmatters.Bioessays.35(8):696-705.

Burgess-HerbertSL,EulingSY.(2011)Useofcomparativegenomicsapproachestocharacterizeinterspeciesdifferencesinresponsetoenvironmentalchemicals:challenges,opportunities,andresearchneeds.ToxicolApplPharmacol.15;271(3):372-85.

BolouriH.(2014)Modelinggenomicregulatorynetworkswithbigdata.TrendsGenet.30(5):182-91.

CivelekM,LusisAJ.(2014)Systemsgeneticsapproachestounderstandcomplextraits.NatRevGenet.15(1):34-48.

LeisersonMD,EldridgeJV,RamachandranS,RaphaelBJ.(2013)NetworkanalysisofGWASdata.CurrOpinGenetDev.23(6):602-10.

KooCL,LiewMJ,MohamadMS,SallehAH.(2013)Areviewfordetectinggene-geneinteractionsusingmachinelearningmethodsingeneticepidemiology.BiomedResInt.2013;2013:432375.

MarjoramP,ZubairA,NuzhdinSV.(2014)Post-GWAS:wherenext?Moresamples,moreSNPsormorebiology?Heredity(Edinb).112(1):79-88.

MurthyD,AttriKS,GokhaleRS.(2013)Network,nodesandnexus:systemsapproachtomultitargettherapeutics.CurrOpinBiotechnol.24(6):1129-36.

CullP.(2013)BIOCOMPUTATION:somehistoryandprospects.Biosystems.112(3):196-203.PajoroA,BiewersS,DougaliE,etal.(2014)The(r)evolutionofgeneregulatorynetworks

controllingArabidopsisplantreproduction:atwo-decadehistory.JExpBot.65(17):4731-45.NajafiA,BidkhoriG,BozorgmehrJH,KochI,Masoudi-NejadA.(2014)Genomescalemodelingin

systemsbiology:algorithmsandresources.CurrGenomics.15(2):130-59.HarroldJM,RamanathanM,MagerDE.(2013)Network-basedapproachesindrugdiscoveryand

earlydevelopment.ClinPharmacolTher.94(6):651-8.NilssonE,ZhangB,SkinnerMK.(2013)Genebionetworksthatregulateovarianprimordialfollicle

assembly.BMCGenomics.23;14:496.AntonyPM,BallingR,VlassisN.(2012)Fromsystemsbiologytosystemsbiomedicine.CurrOpin

Biotechnol.23(4):604-8.HansenJ,IyengarR.(2013)Computationasthemechanisticbridgebetweenprecisionmedicine

andsystemstherapeutics.ClinPharmacolTher.93(1):117-28.ChalanconG,RavaraniCN,BalajiS,Martinez-AriasA,AravindL,JothiR,BabuMM.(2012)

Interplaybetweengeneexpressionnoiseandregulatorynetworkarchitecture.TrendsGenet.28(5):221-32.

HeJC,ChuangPY,Ma'ayanA,IyengarR(2012)Systemsbiologyofkidneydiseases.KidneyInt.81(1):22-39.

ChiaNY,NgHH.(2012)Stemcellgenome-to-systemsbiology.WileyInterdiscipRevSystBiolMed.4(1):39-49.

ZhaoS,IyengarR.(2012)Systemspharmacology:networkanalysistoidentifymultiscalemechanismsofdrugaction.AnnuRevPharmacolToxicol.10;52:505-21.

AntezanaE,MironovV,KuiperM.(2012)TheemergenceofSemanticSystemsBiology.NBiotechnol.2012Nov16.[Epubaheadofprint]

MorelliLG,UriuK,AresS,OatesAC.(2012)Computationalapproachestodevelopmentalpatterning.Science.13;336(6078):187-91

MitchellW,MatsumotoS.(2011)Large-scaleintegratedsuper-computingplatformfornextgenerationvirtualdrugdiscovery.CurrOpinChemBiol.15(4):553-9.

OrchardS.(2012)Molecularinteractiondatabases.Proteomics.12(10):1656-62.HoyleRB,AvitabileD,KierzekAM.(2012)Equation-freeanalysisoftwo-componentsystem

signallingmodelrevealstheemergenceofco-existingphenotypesintheabsenceofmultistationarity.PLoSComputBiol.8(6):e1002396

BorensteinE.(2012)Computationalsystemsbiologyandinsilicomodelingofthehumanmicrobiome.BriefBioinform.2012Nov;13(6):769-80.

AderemA.(2005)Systemsbiology:itspracticeandchallenges.Cell.20;121(4):511-3.HoodL.(2003)Systemsbiology:integratingtechnology,biology,andcomputation.MechAgeing

Dev.124(1):9-16.HoodL,HeathJR,PhelpsME,LinB.(2004)Systemsbiologyandnewtechnologiesenable

predictiveandpreventativemedicine.Science.306(5696):640-3.KirschnerMW.(2005)Themeaningofsystemsbiology.Cell.121(4):503-4.KitanoH.(2002)Systemsbiology:abriefoverview.Science.295(5560):1662-4.KitanoH.(2002)Computationalsystemsbiology.Nature.14;420(6912):206-10.KitanoH.(2004)Biologicalrobustness.NatRevGenet.5(11):826-37.KitanoH.(2004)Cancerasarobustsystem:implicationsforanticancertherapy.NatRevCancer

4(3):227-35.KitanoH.(2007)Towardsatheoryofbiologicalrobustness.MolSystBiol.3:137.Lisacek,F.andAppel,R.D.(2007),SystemsBiology.PROTEOMICS,7:825–827.MihajloMesarovic,SreeN.Sreenath(2006)BeyondtheFlatEarthPerspectiveinSystemsBiology.

BiologicalTheoryWinter.Vol.1,No.1:33–34.O'MalleyMA,DupréJ.(2005)Fundamentalissuesinsystemsbiology.Bioessays27(12):1270-6.SorgerPK.(2005)Areductionist'ssystemsbiology:opinion.CurrOpinCellBiol.17(1):9-11.SotoAM,SonnenscheinC.(2004)Thesomaticmutationtheoryofcancer:growingproblemswith

theparadigm?Bioessays.26(10):1097-107.WesterhoffHV,PalssonBO.(2004)Theevolutionofmolecularbiologyintosystemsbiology.Nat

Biotechnol.22(10):1249-52.SneppenK,KrishnaS,SemseyS.(2010)Simplifiedmodelsofbiologicalnetworks.AnnuRev

Biophys.9;39:43-59.GavinAC,MaedaK,KühnerS.(2010)Recentadvancesinchartingprotein-proteininteraction:

massspectrometry-basedapproaches.CurrOpinBiotechnol.Oct8.[Epubaheadofprint].LittleJW.(2010)Evolutionofcomplexgeneregulatorycircuitsbyadditionofrefinements.Curr

Biol.14;20(17):R724-34.MalyIV.(2009)Introduction:apracticalguidetothesystemsapproachinbiology.MethodsMol

Biol.500:3-13.

From Systems Biology to Systems BiomedicinePaul MA Antony, Rudi Balling and Nikos Vlassis

Available online at www.sciencedirect.com

Systems Biology is about combining theory, technology, and

targeted experiments in a way that drives not only data

accumulation but knowledge as well. The challenge in Systems

Biomedicine is to furthermore translate mechanistic insights in

biological systems to clinical application, with the central aim of

improving patients’ quality of life. The challenge is to find

theoretically well-chosen models for the contextually correct

and intelligible representation of multi-scale biological

systems. In this review, we discuss the current state of Systems

Biology, highlight the emergence of Systems Biomedicine, and

highlight some of the topics and views that we think are

important for the efficient application of Systems Theory in

Biomedicine.

Address

Luxembourg Centre for Systems Biomedicine, University of

Luxembourg, Luxembourg

Corresponding author: Antony, Paul MA ([email protected])

Current Opinion in Biotechnology 2012, 23:604–608

This review comes from a themed issue on Systems biology

Edited by Jens Nielsen and Sang Yup Lee

For a complete overview see the Issue and the Editorial

Available online 24th November 2011

0958-1669/$ – see front matter, # 2011 Elsevier Ltd. All rights

reserved.

http://dx.doi.org/10.1016/j.copbio.2011.11.009

IntroductionComplex interactions in biological systems are an increas-

ingly important component of translational research. Sys-

tems Biology is a unifying term that groups

interdisciplinary scientific efforts that focus on the analysis

of complex interactions in biological systems [1]. For the

past decade, the term has been widely used in a variety of

biomedical contexts. The amalgamation of two opposing

concepts, namely, holism and reductionism is currently

contributing to conceptual advances in Systems Biology

[2�,3]. The holistic approach is based on the idea that

complex systems cannot be fully understood by studying

isolated parts. One of the main ambitions of holistic Sys-

tems Biology is the understanding of emergent properties

such as robustness [4]. The reductionists focus on modules

within bigger systems and can be described as modern

physiologists working on molecular explanations for bio-

logical events. There is increasing consent that it is coun-

terproductive to engage in ongoing potentially divisive

arguments comparing holism with reductionism, and it is


becoming clear that both reductionists and holists, need to

connect molecular and clinical scales. Systems Biomedi-

cine can be considered as an emergent branch of Systems

Biology that allows zooming through the multiple scales of

life and disease by combining reductionist and holistic

approaches. A widely applied research concept is a cycle

composed of theoretical analysis, computational model-

ling, and experimental validation of model hypotheses,

which drives the refinement of computational and exper-

imental models.

Systems Biomedicine: Scope and bottlenecksThe ultimate goal in Systems Biomedicine is to apply

mechanistic insights to clinical application and to improve

patients’ quality of life. This aim, however, raises the

need to solve some important questions on how to organ-

ise a framework supporting these interdisciplinary

research efforts [5,6�]. Recognition of a clinical, mechan-

istic, or theoretical problem is the first step. Formulating a

testable hypothesis is the second. Such a hypothesis can

be considered as a model that requires further validation

in biological and clinical trials before clinical application.

However, such interdisciplinary research can easily stag-

nate. To avoid stagnation, theoretical, biological, and

clinical researchers need to share their current knowl-

edge, aims, and visions for the future (Figure 1).

The human genome has now been completely

sequenced, and many genetic variants and epigenetic

modifications have been identified. Recent improve-

ments in technology have facilitated the collection of

unprecedented amounts of data at various biological

and clinical levels. A comprehensive understanding of

all this information is, however, lagging behind the

accumulation of data, and we are only beginning to

understand the extent to which phenotypic traits and

their modes of emergence are regulated at multiple levels

of biological complexity [7,8�].

A classical approach for studying emergent properties

from genetic or environmental factors is to perturb the

factor of interest and to observe the resultant phenotypic

changes. An important technology-driven development

in this field is the shift from phenotypic endpoint studies

to time-resolved phenotypic profiling [9]. It is, however,

potentially dangerous and misleading to expect relation-

ships of the type ‘one gene – one protein – one function –one phenotype’. Frequently, phenotypic changes do not

clearly reveal genetic perturbations due to extensive

buffering driven by network robustness [10]. Thus, pre-

dictions of phenotypic changes need to consider both

perturbation properties and network context. The effect

www.sciencedirect.com

mailto:[email protected]

http://www.sciencedirect.com/science/journal/09581669/23/4



http://www.sciencedirect.com/science/journal/09581669

From Systems Biology to Systems Biomedicine Antony, Balling and Vlassis 605

Figure 1

Problem recognition

Model = Hypothesis

Biological validation

Clinical validation

Clinical application

Inte

rdis

cipl

inar

y re

sear

ch p

rogr

ess

Specialist research progress

Biological models

Clinical trials

Knowledge base

Theoretical modelsPatient,clinicalreality

Current Opinion in Biotechnology

Systems Biomedicine: Scope, scale, and critical transitions in research progress.

of certain drugs in such contexts is often not easily

predicted. Such predictions would require both a suffi-

cient understanding of molecular, cellular, and clinical

scales, and a sufficient understanding of emergent proper-

ties that connect the different scales, for example, of heart

function. In a pioneer study, Fink and Noble [11] demon-

strated that mathematical models are beneficial for unra-

velling the complex interactions of pharmacodynamics in

the heart: They predicted that embedding detailed cel-

lular-scale models into anatomically correct organ-scale

models would enable reconstructions of cardiac dysfunc-

tions, providing a tool for connecting cellular-scale exper-

imental data to clinical-scale applications.

In the context of Western medicine, molecular under-

standing of pathogenic events is a fundamental aim and

driving force. In addition, modern Systems Biomedicine

highlights the importance of a higher level of biological

organisation. A much older discipline with a very similar

interest in higher order biological organisation is Oriental

medicine. Sasang constitutional medicine, the Korean

traditional medicine, seeks patient-specific analysis and

treatment. The current Western trend of recognizing the

potential value of predictive, personalised, preventive,

and participatory (P4) medicine [12] represents a timely

convergence between Western and Oriental medical dis-

ciplines [13,14]. However, many challenges will need to

be overcome before this convergence translates to actual

clinical outcomes.

Diseases and modelsDiseases can be roughly categorised into those that are due

to environmental factors and those due to genetic factors.


The latter can range from single causal factors to modifying

risk factors and multifactorial diseases. Furthermore, the

onset and progression of many diseases is influenced by a

combination of both environmental and genetic factors

[15,16]. In monogenic diseases, a mutation in a single gene

causes the disease. Such diseases are relatively easy to trace

by classical Mendelian inheritance patterns, and recent

research has been directed towards exploiting such pat-

terns in the context of whole family DNA sequencing [17].

On the contrary, multigenic or complex diseases are those

in which progression is believed to be influenced by the

collective action of several genes and environmental fac-

tors. Examples include epilepsy, Parkinson’s disease, and

cancer. The multifactorial nature of such diseases renders

making predictions relating to their onset and progression a

very challenging task.

In an effort to facilitate more accurate predictions of

disease onset and progression, researchers often utilise

‘models’ of the disease. The term ‘model’ is used in

biology in different ways, but it often means an idealised

representation of an empirical system [18]. Models can be

viewed as artefacts that allow casting of a natural phenom-

enon in a formal language such as mathematics, physics,

or computer science. Modelling, the process that com-

bines the identification and collection of confident pre-

liminary knowledge and the identification of potential

mechanistic explanations, is a central tool in Systems

Biology. As George Box said ‘all models are wrong, but

some are useful’ [19].

To predict ‘simple’ diseases such as those with Men-

delian inheritance patterns we probably do not need


606 Systems biology

complex models; often, we only need to determine the

allelic makeup of a single gene, which usually amounts

to a low-complexity computational test. The length of

the polyglutamine stretch in the Huntingtin gene, for

example, allows prediction of the approximate age of

onset of Huntington’s disease. While additional disease-

modifying factors exist, they exert only a minor influ-

ence on the age of onset [20].

On the contrary, simple models are probably inadequate

for facilitating predictions relating to multifactorial com-

plex diseases such as epilepsy. It has been shown that

single-nucleotide polymorphism burden in ion channels

holds little clinical predictive power for epilepsy [21��],thus validating the complex multigenic inheritance of

epileptic channelopathy and the principle of compensa-

tory effects in complex systems. Individualised disease

prediction appears to require an understanding of how

genetic variants contribute to risk in each given individual

[22,23]. In epilepsy, different ion channel variants might

potentiate or suppress similar membrane currents and

thereby influence neuronal firing patterns and pathologi-

cal phenotypes. Furthermore, even physically non-inter-

acting channel proteins could modulate distant channels

via changes in the transmembrane potential. Appreciation

of such mechanisms and modifying factors within a

personalised approach would require complex models,

but may also yield valuable information with far-ranging

implications for future research aimed at translating geno-

mic profiling into risk prediction [21��,24].

The prototypical complex model in Systems Biology is a

network, whose semantics may vary depending on the

particular application and degree of approximation. A net-

work model provides a view of a biological system as a

graph dynamical system, which is defined by the coordi-

nated action of several nodes and corresponding edges, and

their local dynamics [25]. This allows for the study of the

global properties of dynamical biological systems via cor-

responding concepts in the phase space. Commonly used

concepts are stable equilibrium states, for example, health

and disease [26,27], and limit cycles, for example, oscil-

lations of the cell cycle [28��]. The appropriate application

of these modelling concepts to biological systems can

strongly enhance predictive power [29,30]. Large-scale

complex networks can be difficult to analyse though,

and recently, a line of research has emerged that focuses

on the use of mathematical theorems and results from

computer science in an effort to mitigate this complexity.

In particular, a variety of theoretical and experimental

results demonstrate a relationship between the structural

properties of a network and properties of the phase space of

the corresponding dynamical system [31–35].

In general, three main types of modelling strategies can

be distinguished. Modelling can be top-down, from

clinical signs to molecular processes; bottom-up, from


molecular processes to clinical phenotypes; or middle-

out, starting from an intermediate scale [36]. Bottom-up

approaches are useful when achievements from Molecu-

lar Biology allow for the formulation of hypotheses on

emergent properties. Middle-out approaches typically

start from the basic unit of life, the cell [37]. Since living

cells coordinate the flow of spatiotemporal information

between molecular and organic scales, middle-out

approaches to modelling seem to be especially apt for

multiple-scale models; the components incorporated into

these models may include the genetic code, transcription,

splicing, RNA interference, translation, post-translational

modification, protein complexes, organelles and compart-

ments, metabolism, cellular function, cell to cell inter-

actions, tissue function, organ function, system response

to environmental factors, health status, and quality of life.

Examples of such multi-scale models can be found in the

Virtual Cell, Physiome Project, and Virtual Physiological

Human initiative [37–39].

Complex models possess the freedom to capture subtle

details of the system of interest; however, some models

trade off precision for simplicity and represent qualitative

features of the system. For example, a Boolean network

can be used for modelling gene expression in a qualitative

manner, by assuming that a gene can be either ‘on’ or ‘off’,

a simplified assumption that seems to work well in

practice [40]. While complex models would seem a better

choice in terms of predictive power, such models may

lead to ‘good’ results for the wrong reasons. A ‘good fit’

between a model and observed system behaviour does

not guarantee any realism of parameter values or model

structure [41,42].

It can be argued that biological processes may be inher-

ently stochastic [43,44]. This calls for models that explicitly

incorporate uncertainty. The prototypical paradigms here

are probability theory and Bayesian analysis. The latter, in

particular, allows for the incorporation of process stochas-

ticity and model parameter uncertainty within the same

framework. For parameter estimation, the modeller places

a prior probability distribution over the space of model

parameters, and then uses the existing sensorial evidence

to infer a posterior distribution over these parameters,

thereby ‘learning’ a model from data [45]. The field of

machine learning addresses questions related to the effi-

cient learning of models based on observed data [46,47].

With regard to process stochasticity, even if certain bio-

logical processes ultimately prove to be truly deterministic

(as Einstein said, ‘I, at any rate, am convinced that He does

not throw dice’), it may still be useful to treat some of these

as stochastic; stochasticity is often a convenient mathemat-

ical surrogate of our ignorance [48,49].

One important guideline however is to keep models as

simple as possible while adding as much complexity as

is needed in order to test hypotheses of relevance


From Systems Biology to Systems Biomedicine Antony, Balling and Vlassis 607

for translational research. Furthermore, it is crucial to

incorporate considerations arising from Systems Theory,

Molecular Biology, Physiology, and Medicine into disease

models. An interdisciplinary approach accelerates the

route of mechanistic insights to clinical application and

the improvement of patients’ quality of life.

References and recommended readingPapers of particular interest, published within the period of review,have been highlighted as:

� of special interest�� of outstanding interest

1. Auffray C, Noble D: Origins of systems biology in WilliamHarvey’s masterpiece on the movement of the heart and theblood in animals. Int J Mol Sci 2009, 10:1658-1669.

2.�

Brenner S: Sequences and consequences. Philos Trans R SocLond B Biol Sci 2010, 365:207-212.

This publication constitutes an outstanding and critical view of the scopeand status quo of systems biology, incorporating the concepts of bothholism and reductionism.

3. Gatherer D: So what do we really mean when we say thatsystems biology is holistic? BMC Syst Biol 2010, 4:22.

4. Kitano H: Systems biology: a brief overview. Science 2002,295:1662-1664.

5. Molina F, Dehmer M, Perco P, Graber A, Girolami M, Spasovski G,Schanstra JP, Vlahou A: Systems biology: opening new avenuesin clinical research. Nephrol Dial Transplant 2010, 25:1015-1018.

6.�

Kitano H, Ghosh S, Matsuoka Y: Social engineering for virtual‘big science’ in systems biology. Nat Chem Biol 2011,7:323-326.

This publication highlights the social challenges involved in big interdis-ciplinary research and discusses the implementation of a potential frame-work for successful projects.

7. Hall SS: Revolution postponed. Sci Am 2010, 303:60-67.

8.�

Tay S, Hughey JJ, Lee TK, Lipniacki T, Quake SR, Covert MW:Single-cell NF-kappaB dynamics reveal digital activation andanalogue information processing. Nature 2010, 466:267-271.

In this paper, the authors study the spatiotemporal cell populationbehaviour triggered by high-throughput microfluidic perturbations, thatis, analogue gene expression triggered by digital NF-kB activation.

9. Neumann B, Walter T, Heriche JK, Bulkescher J, Erfle H, Conrad C,Rogers P, Poser I, Held M, Liebel U et al.: Phenotypic profiling ofthe human genome by time-lapse microscopy reveals celldivision genes. Nature 2010, 464:721-727.

10. Noble D: Neo-Darwinism, the modern synthesis and selfishgenes: are they of use in physiology? J Physiol 2011,589:1007-1015.

11. Fink M, Noble D: Pharmacodynamic effects in thecardiovascular system: the modeller’s view. Basic ClinPharmacol Toxicol 2010, 106:243-249.

12. Galas D, Hood L: Systems biology and emerging technologieswill catalyze the transition from reactive medicine topredictive, personalized, preventive and participatory (P4)medicine. IBC 2009, 6:1-4.

13. Noble D: Could there be a synthesis between Western andOriental medicine, and with Sasang constitutional medicine inparticular? Evid Based Complement Alternat Med 2009,6(Suppl. 1):5-10.

14. Noble D: Systems biology, the Physiome Project and orientalmedicine. J Physiol Sci 2009, 59:249-251.

15. Knip M, Veijola R, Virtanen SM, Hyoty H, Vaarala O, Akerblom HK:Environmental triggers and determinants of type 1 diabetes.Diabetes 2005, 54(Suppl. 2):S125-S136.

16. Virgin HW, Todd JA: Metagenomics and personalizedmedicine. Cell 2011, 147:44-56.


17. Roach JC, Glusman G, Smit AF, Huff CD, Hubley R, Shannon PT,Rowen L, Pant KP, Goodman N, Bamshad M et al.: Analysis ofgenetic inheritance in a family quartet by whole-genomesequencing. Science 2010, 328:636-639.

18. Odenbaugh J: Models in biology. In Routledge Encyclopedia ofPhilosophy. Edited by Craig E. Routledge: London; 2009. ISSN:04719332.

19. Box J: Science and statistics. J Am Stat Assoc 1976:791-799.

20. Walker FO: Huntington’s disease. Lancet 2007, 369:218-228.

21.��

Klassen T, Davis C, Goldman A, Burgess D, Chen T, Wheeler D,McPherson J, Bourquin T, Lewis L, Villasana D et al.: Exomesequencing of ion channel genes reveals complex profilesconfounding personal risk assessment in epilepsy. Cell 2011,145:1036-1048.

This publication discusses the low predictive power of ion channelmutations with regard to epilepsy and highlights the importance ofunderstanding the context-dependant functional downstream effectsof such ion channel mutations for improving personal risk assessment.

22. Consortium GP: A map of human genome variation frompopulation-scale sequencing. Nature 2010, 467:1061-1073.

23. Manolio TA, Collins FS, Cox NJ, Goldstein DB, Hindorff LA,Hunter DJ, McCarthy MI, Ramos EM, Cardon LR, Chakravarti Aet al.: Finding the missing heritability of complex diseases.Nature 2009, 461:747-753.

24. Choi M, Scholl UI, Ji W, Liu T, Tikhonova IR, Zumbo P, Nayir A,Bakkaloglu A, Ozen S, Sanjad S et al.: Genetic diagnosis bywhole exome capture and massively parallel DNA sequencing.Proc Natl Acad Sci USA 2009, 106:19096-19101.

25. Ma’ayan A: Introduction to network analysis in systemsbiology. Sci Signal 2011, 4:tr5.

26. Devor A, Sakadzic S, Saisan PA, Yaseen MA, Roussakis E,Srinivasan VJ, Vinogradov SA, Rosen BR, Buxton RB, Dale AMet al.: ‘‘Overshoot’’ of O2 is required to maintain baseline tissueoxygenation at locations distal to blood vessels. J Neurosci2011, 31:13676-13681.

27. del Sol A, Balling R, Hood L, Galas D: Diseases as networkperturbations. Curr Opin Biotechnol 2010, 21:566-571.

28.��

Ferrell JE Jr, Tsai TY, Yang Q: Modeling the cell cycle: why docertain circuits oscillate? Cell 2011, 144:874-885.

This publication discusses how cell cycle oscillations may arise. In theirtutorial, the authors elaborate on how to adapt the complexity of modelsto the complexity of the biological systems being studied, and commenton the extent to which the theory of non-linear systems can assist theunderstanding of observed events.

29. Kauffman S: Homeostasis and differentiation in randomgenetic control networks. Nature 1969, 224:177-178.

30. Huang S, Ernberg I, Kauffman S: Cancer attractors: a systemsview of tumors from a gene network dynamics anddevelopmental perspective. Semin Cell Dev Biol 2009, 20:869-876.

31. Cournac A, Sepulchre JA: Simple molecular networks thatrespond optimally to time-periodic stimulation. BMC Syst Biol2009, 3:29.

32. Shah NA, Sarkar CA: Robust network topologies for generatingswitch-like cellular responses. PLoS Comput Biol 2011,7:e1002085.

33. Shiraishi T, Matsuyama S, Kitano H: Large-scale analysis ofnetwork bistability for human cancers. PLoS Comput Biol 2010,6:e1000851.

34. Thomas R: On the relation between the logical structure ofsystems and their ability to generate multiple steady states orsustained oscillations. In Springer Series in Synergetics, vol. 9.Edited by Della Dora J, Demongeot J, Lacolle B. Berlin: SpringerVerlag; 1981:180-193.

35. Remy E, Ruet P: From minimal signed circuits to the dynamicsof Boolean regulatory networks. Bioinformatics 2008,24:i220-i226.

36. Majumder D, Mukherjee A: A passage through systems biologyto systems medicine: adoption of middle-out rational


608 Systems biology

approaches towards the understanding of therapeuticoutcomes in cancer. Analyst 2011, 136:663-678.

37. Walker DC, Southgate J: The virtual cell—a candidate co-ordinator for ‘middle-out’ modelling of biological systems.Brief Bioinform 2009, 10:450-461.

38. Kohl P, Noble D: Systems biology and the virtual physiologicalhuman. Mol Syst Biol 2009, 5:292.

39. Hunter PJ, Borg TK: Integration from proteins to organs: thePhysiome Project. Nat Rev Mol Cell Biol 2003, 4:237-243.

40. Shmulevich I: In Probabilistic Boolean Networks: The Modelingand Control of Gene Regulatory Networks. Edited by Greenspan E.SIAM; 2009. ISBN: 978-0-898716-92-4.

41. Scheffer M: Critical Transitions in Nature and Society. PrincetonUniversity Press; 2009.

42. Nurse P: Life, logic and information. Nature 2008, 454:424-426.

43. Ideker T, Dutkowski J, Hood L: Boosting signal-to-noise incomplex biology: prior knowledge is power. Cell 2011,144:860-863.


44. Li GW, Xie XS: Central dogma at the single-molecule level inliving cells. Nature 2011, 475:308-315.

45. Tarantola A: Popper, Bayes and the inverse problem. Nat Phys2006, 2:492-494.

46. Hastie T, Tibshirani R, Friedman J: The Elements of StatisticalLearning: Data Mining, Inference, and Prediction. edn 2.Springer Series in Statistics. Springer Science+BusinessMedia, LLC; 2009 :ISBN: 978-0-387-84857-0, doi:10.1007/b94608_2.

47. Koller D, Friedman N: Probabilistic Graphical Models, Principlesand Techniques. MIT Press; 2009.

48. Dompere KK: In Fuzziness and Approximate Reasoning:Epistemics on Uncertainty, Expectation and Risk in RationalBehavior (Studies in Fuzziness and Soft Computing), edn 1. Editedby Kacprzyk J. Berlin Heidelberg: Springer; 2009. ISBN: 13: 978-3540880868.

49. Russell SJ, Norvig P: Artificial Intelligence: A Modern Approach.Prentice Hall; 2009.


Letter to the editor: “Systems biology versus reductionism in cell physiology”

Mark A. Knepper, Viswanathan Raghuram, Davis Bradford, Chung-Lin Chou, Jason D. Hoffert,and Trairak PisitkunEpithelial Systems Biology Laboratory, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda,Maryland

TO THE EDITOR: The following is a response to the editorialcomment of Prihandoko and Tobin (15) about our recent paperin American Journal of Physiology-Cell Physiology (2), whichaddresses a key question in modeling of signaling networks:How to assign the protein kinases (from the entire 521-memberkinome list) that are responsible for each measurable phosphor-ylation event in a given cell type. In our study, we usedvasopressin-stimulated phosphorylation of the water channelprotein, aquaporin-2, at serine-256 as an example because of itsimportance to the physiology of collecting duct principal cells.We thank Prihandoko and Tobin for their thorough and wellthought out summary of our paper. We write now to provideadditional clarification regarding the epistemological approach,which was based on a systems biological framework ratherthan on reductionist principles. Understanding the two ways ofdoing experiments is aided by a bit of history.

Attention to the problem of how to make practical scientificinferences from scientific observations peaked in 19th centurywith John Stuart Mill’s book “A System of Logic” (12; seechapter “Of the Four Methods of Scientific Inquiry”). Mill’swork described several approaches built from two fundamentalmethods, viz. the “method of difference” and the “method ofagreement.” From the viewpoint of modern biology, the formermethod is the basis of reductionist approaches and becamedominant in the 20th century. The latter method is the basis ofthe newly resurgent systems biology approach. We can con-ceptualize the method of difference as the standard hypothesis-driven experiment in which a given variable is altered andanother variable is observed. This approach thrived because ithas often been feasible to make the targeted measurementsneeded and because statistical methods were developed early inthe 20th century by Fisher and others to analyze such data (14).However, reductionist approaches have drawn fire in recentyears because of perceived bias in publication (7). Critics claimthat positive results from reductionist experiments are publish-able (often whether true-positive or false-positive), while negativeresults are not. In addition, the statistical approach to analysis ofreductionist data draws conclusions one experiment at a time, anddoes not generally utilize prior information to draw conclusions(4, 14), a problem that is circumvented in systems biologicalapproaches. The latter, roughly equivalent to Mill’s method ofagreement, looks broadly for correlations in comprehensivedata sets and builds models based on these correlations. Com-prehensive methodologies including large-scale proteomics,DNA microarrays, and “next generation” DNA sequencinghave only recently become feasible because of the availabilityof genome-wide sequence data needed for mapping. Thus,biological approaches based on Mill’s method of agreement

(systems biology approaches), heretofore impractical, have inthe 21st century become feasible. Concomitantly, statisticalmethodologies for analysis of comprehensive data sets havefollowed, e.g., the use of Bayesian statistics. Our study (2)utilized the systems approach as summarized in the next twoparagraphs. The commentary (15) appeared to retell the storythat we presented as a series of separately interpreted reduc-tionist experiments, thus losing the major message of ourpaper, viz. that Bayes’ theorem can be used to integratemultiple imperfect data sets to provide deeper, stronger con-clusions than could be expected without data integration.

Our previous study in AJP-Cell (5) showed, using massspectrometry, that protein kinases are low fidelity enzymes andwhen combined with prior observations (11) suggested thatprotein kinases gain specificity in the cell chiefly throughfactors that cause them to colocalize with specific substrates.From this and other studies, it was already clear that we canrely only on very general specificity constraints, basicallywhether they phosphorylate tyrosines or serines/theonines, andwhether the latter are basophilic, acidophilic, or proline-di-rected. Thus, the question of what protein kinase(s) phosphor-ylate serine-256 of aquaporin-2 was not answerable simply bylooking at the amino acid sequence surrounding it. Moreinformation was needed. To address the question, we inte-grated prior information from several sources using Bayes’theorem to rank all 521 kinases in the rat genome with regardto the probability that they phosphorylate serine-256 of aqua-porin-2 in the rat inner medullary collecting duct (IMCD). Thisincluded data gleaned from prior large-scale (proteomic ortranscriptomic) experiments in the IMCD. This Bayes’ ap-proach allowed us to utilize data, which in isolation did notanswer the question, but narrowed the choices. For example,transcriptomics experiments divided the 521 protein kinasegenes into those that were expressed in IMCD and those thatwere not detectable, and thus were unlikely to play a regulatoryrole regardless of kinase specificity. Use of Bayes’ theorem tointegrate information from many sources is not new; it wasused for example to establish the conclusion that smoking isharmful to health in the 1950s (3). However, as far as we cantell, the use of Bayes’ theorem to integrate multiple data sets incell physiology is novel and it is therefore surprising that it wasnot explicitly discussed in the Prihandoko and Tobin commen-tary.

Using the Bayesian integration of prior data as a launchingpoint, our study (2) addressed whether addition of inhibitordata could sharpen the Bayesian estimates. Protein kinaseinhibitors have been used in physiology for many decades,always with tacit recognition that they inhibit multiple kinasesin addition to the nominal target kinase. Now, the InternationalCentre for Kinase Profiling (ICKP, http://www.kinase-screen.mr-c.ac.uk/kinase-inhibitors) has provided profiling data for manycommonly used protein kinase inhibitors. This comprehensive

Address for reprint requests and other correspondence: M. A. Knepper,National Institutes of Health, Bldg. 10, Rm. 6N307, 10 Center Dr., MSC-1603,Bethesda, MD 20892-1603 (e-mail: [email protected]).

Am J Physiol Cell Physiol 307: C308–C309, 2014;doi:10.1152/ajpcell.00175.2014.Letter to the Editor

http://www.ajpcell.orgC308

by 10.220.33.3 on Decem

ber 21, 2016http://ajpcell.physiology.org/

Dow

nloaded from

http://www.kinase-screen.mrc.ac.uk/kinase-inhibitors%29

http://www.kinase-screen.mrc.ac.uk/kinase-inhibitors%29

mailto:[email protected]

http://ajpcell.physiology.org/

data set identifies which kinases are and which kinases are notinhibited by a given small-molecule kinase inhibitor, and esti-mates the percentage of kinase activity remaining for relevantinhibitor concentrations. The ICKP data give new life to the useof inhibitors in physiological experiments by its comprehen-sive nature. It allowed phosphorylation data from immunoblot-ting of IMCD suspensions to be integrated with prior datausing Bayes’ theorem, thereby significantly improving dis-crimination among candidate kinases involved in aquaporin-2phosphorylation at serine-256. The overall Bayes’ analysisshows that the conventional wisdom, that protein kinase Aphosphorylates this site in the collecting duct cell, is not anybetter supported by the data than roles for several otherbasophilic protein kinases including calcium/calmodulin-de-pendent protein kinase 2� (Camk2d) and protein kinase B-�(Akt1). In fact, the top ranked protein kinase in the Bayes’analysis, calcium/calmodulin-dependent protein kinase 2�, wasshown in mass spectrometry experiments to be as potent inphosphorylating aquaporin-2 in vitro as was protein kinase A,or more so.

In summary, our paper used a systems biological approachinvolving application of Bayes’ theorem to integrate multipledata sets. Such an approach appears to be new to cell physi-ology and appears to provide significant advantages for certainphysiological problems such as the assignment of kinases tophosphorylation sites. We as authors recognize that the onus ison us to provide a persuasive argument for the systems ap-proach. It may indeed be difficult for many biologists toembrace systems biology after a 100 years of reductionism.Toward that end, we invite the interested reader to view ourprevious writings about systems biology in AJP- Cell (8, 9) aswell as recent articles by others in this journal (1, 6, 10, 13, 16).

ACKNOWLEDGMENTS

Present address of T. Pisitkun: Faculty of Medicine, Chulalongkorn Uni-versity, Bangkok, Thailand. Supported by NHLBI Division of IntramuralResearch (Project ZO1-HL-001285).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

M.A.K. drafted manuscript; M.A.K., V.R., D.B., C.-L.C., J.D.H., and T.P.edited and revised manuscript; M.A.K., V.R., D.B., C.-L.C., J.D.H., and T.P.approved final version of manuscript.

REFERENCES

1. Beard DA, Vendelin M. Systems biology of the mitochondrion. Am JPhysiol Cell Physiol 291: C1101–C1103, 2006.

2. Bradford D, Raghuram V, Wilson JL, Chou CL, Hoffert JD, KnepperMA, Pisitkun T. Use of LC-MS/MS and Bayes’ theorem to identifyprotein kinases that phosphorylate aquaporin-2 at Ser256. Am J PhysiolCell Physiol 307: C123–C139, 2014.

3. Cornfield J, Haehszel W, Hammond EC, Lilienfield AM, ShimkinMB, Wynder EL. Smoking and lung cancer: recent evidence and adiscussion of some questions. J Natl Cancer Inst 22: 173–203, 1959.

4. Davidoff F. Standing statistics right side up. Ann Intern Med 130:1019–1021, 1999.

5. Douglass J, Gunaratne R, Bradford D, Saeed F, Hoffert JD, SteinbachPJ, Knepper MA, Pisitkun T. Identifying protein kinase target prefer-ences using mass spectrometry. Am J Physiol Cell Physiol 303: C715–C727, 2012.

6. George CH, Parthimos D, Silvester NC. A network-oriented perspectiveon cardiac calcium signaling. Am J Physiol Cell Physiol 303: C897–C910,2012.

7. Ioannidis JP. Why most published research findings are false. PLoS Med2: e124, 2005.

8. Knepper MA. Systems biology in physiology: the vasopressin signalingnetwork in kidney. Am J Physiol Cell Physiol 303: C1115–C1124, 2012.

9. Knepper MA. Proteomic pearl diving versus systems biology in cellphysiology. Focus on “Proteomic mapping of proteins released duringnecrosis and apoptosis from cultured neonatal cardiac myocytes.” Am JPhysiol Cell Physiol 306: C634–C635, 2014.

10. Kohli P, Bartram MP, Habbig S, Pahmeyer C, Lamkemeyer T,Benzing T, Schermer B, Rinschen MM. Label-free quantitative pro-teomic analysis of the YAP/TAZ interactome. Am J Physiol Cell Physiol306: C805–C818, 2014.

11. Linding R, Jensen LJ, Ostheimer GJ, van Vugt MA, Jorgensen C,Miron IM, Diella F, Colwill K, Taylor L, Elder K, Metalnikov P,Nguyen V, Pasculescu A, Jin J, Park JG, Samson LD, Woodgett JR,Russell RB, Bork P, Yaffe MB, Pawson T. Systematic discovery of invivo phosphorylation networks. Cell 129: 1415–1426, 2007.

12. Mill JS. A System of Logic: Ratiocinative and Inductive. Honolulu:University Press of the Pacific, 1891.

13. Moore NM, Nagahara LA. Physical Biology in Cancer. 1. Cellularphysics of cancer metastasis. Am J Physiol Cell Physiol 306: C78–C79,2014.

14. Nuzzo R. Scientific method: statistical errors. Nature 506: 150 –152,2014.

15. Prihandoko R, Tobin AB. Challenges of assigning protein kinases to invivo phosphorylation events. Focus on “Use of LC-MS/MS and Bayes’theorem to identify protein kinases that phosphorylate aquaporin-2 atSer256.” Am J Physiol Cell Physiol 307: C121–C122, 2014.

16. Welch GR, Clegg JS. From protoplasmic theory to cellular systemsbiology: a 150-year reflection. Am J Physiol Cell Physiol 298: C1280–C1290, 2010.

Letter to the Editor

C309LETTER TO THE EDITOR

AJP-Cell Physiol • doi:10.1152/ajpcell.00175.2014 • www.ajpcell.org

by 10.220.33.3 on Decem

ber 21, 2016http://ajpcell.physiology.org/

Dow

nloaded from

http://ajpcell.physiology.org/

1/9/17

1

1/9/17

2

www.skinner.wsu.edu


SystemsBiology


RequiredReading




SystemsBiology

Defini0onHistoryTheoryParadigmShi8Parameters

Systemsbiologyisacomprehensivequan0ta0veanalysisofthemannerinwhichallthecomponentsofabiologicalsysteminteractfunc0onallyover0me.Suchananalysisisexecutedbyaninterdisciplinaryteamofinves0gatorsthatisalsocapableofdevelopingrequiredtechnologiesandcomputa0onaltools.Inthismodel,biologydictateswhatnewtechnologyandcomputa0onaltoolsshouldbedeveloped,and,oncedeveloped,thesetoolsopennewfron0ersinbiologyforexplora0on.Thus,biologydrivestechnologyandcomputa0on,and,inturn,technologyandcomputa0onrevolu0onizebiology.

1/9/17

3

“systemsbiologyisthestudyofanorganism,viewedasanintegratedandinterac1ngnetworkofgenes,proteinsandbiochemicalreac1onswhichgiverisetolife”(Hood2005).

SystemsBiologyTheory

Evolu0onarySystemsBiology-Extensionofclassicalbiologyparadigmwithnew technology

Revolu0onarySystemsBiology-Newparadigmshi8inbiologywithaltered

perspec0veoncausalrela0onshipsandsystems

Evolu0onarySystemsBiologyHistory

Systemsbiologyextensioncurrentparadigmandhistoryofbiologywithnewtechnology300BCAristotle,Systemhas4proper0esorcauses:Material,Formal,Efficient,Teleological200ADGalen(RomanPhysician),Teleologicalimportantroleinorganismfunc0on1500sFernel,Systema0capproachAnatomy1600sHarvey,Physiology,CellBiology,Circula0on1700sNewton,Physicsleadstomechanis0cdeterminismtoexplainsystems

LaMe\rie,DefineBiologicalMachine(egClock)1800sBernard,Fatherphysiologyandintegra0onbiologicalsystems(milieuinterieur)1900sCannon,Biologicalequalibriumandhomeostatsis

-DiscoveryDNA/Structure/Genes(MolecularBiology) -Computa0onalBiology(non-equalibriumthermodynamicsandkine0csmetabolism)

2000s-GenomeSequence -OmicsTechnology

1/9/17

4

Evolu0onarySystemBiologyDefini0onsExtensionoftradi0onalbiologicalparadigmMarcKirchner2005“Systemsbiologyisthestudyofthebehaviorofcomplexbiologicalorganiza0onand

processesintermsofthemolecularcons0tuents”WesterhoffandAlberghina2005Systemsbiologyis“nothingbutgoodoldphysiology”orthatis“molecularbiologyclaiming

addi0onalmoney”Sorger2005“Systembiologyaimistobuildnumericalmodelsofbiologicalprocessesandtestthe

modelsexperimentally”

Revolu0onarySystemsBiologyHistoryJanSmuts(1870-1950),SouthAfrica,Defined-Holism(Tendencyinnaturetoformwholes

thataregreaterthanthesumofthepartsthroughcrea0veevolu0on)AlfredWhitehead(1861-1947),USA,Defined-Organisms(Philosophyoforganismto

explainthecomplexityofnaturalprocesses-includingbiologicalorganisms)LudwigvonBertalanffy(1901-1972),Austria,Defined-Disequalibrium(Biological

organismsareopensystems,whichrespondtochangesinenvironment,suchthatdisequalibriumisstateoflivingorganismandequalibriumisdeath)

NorbertWiener(1894-1964),USA,Defined-Cyberne3cs(Applica0onmathema0cstoexplainbiologicalmechanisms)

JosephWoodger(1894-1981),UK,Defined-Bauplan(Bauplanastheessen0alstructuralplanormorphologyofanorganismbodyplan,egvertebrates)

ConradWaddington(1905-1975),Scotland,Defined-Epigene3cs(Discusslater)WalterElsasser(1904-1991),Hungarian,Defined-Biotonic(Lawsnotreducibletophysical

orchemicallaws)1980sTheore0calBiologyHolism(ElsasserandLaszlo)(Bu\erflyEffect)

ChaosTheory(Mathema0calapproachcomplexsystems)1990sHighthroughputsequencingandexpansionepigene0carea2000sSequencegenomeandtranscriptome(Omicstechnologies)

1/9/17

5

Revolu0onaryDefini0onsforSystemsBiologyLeroyHood(2005)“Theinter-rela0onshipsofalltheelementsinasystemratherthanstudyingthemoneata

0me”MethodologicalApproach-1)  Developsimpledescrip0ve,graphical,ormathema0calmodelofhowsystemfunc0ons2)  Iden0fyanddefinethevariouscomponentsofthesystemandtheirstate(egomics)3)  Disturbthesystemwithexternalperturba0onanddocumentchangesinthecomponents4)  Integra0onofthetwodatasetsfromstep3andcomparisontomodelinstep15)  Adjustmodelun0lharmonyorconjunc0onexistsbetweendataandmodel

HiroakiKitano(2002)Fourfactorsforcomprehensivesystemsbiologydefini0on1)  SystemStructure,organiza0onofcomponents(macromolecules,genes,cells,0ssuesetc2)  SystemDynamics,interac0onsbetweenorrela0onshipsofthevarioushierarchicallevels

over0me3)  SystemsControlMethod,regulatorymechanismsinvolvedinthemaintenanceofthe

organiza0onalhierarchy4)  SystemsDesignMethod,hierarchicalorganiza0onwithspecificproper0esandmanipulate

Reduc0onismTheviewthattheul0matescien0ficunderstandingofarangeofphenomenaistobe

gainedexclusivelyfromlookingatthecons0tuentsofthesephenomenaandtheirproper0es

OntologicalReduc0onism

Thatcomplexphenomenaarereducibletoordeterminablebysimpleren00esandforcesthatcomposethem(eggene0cdeterminism)and(bo\om-uporupwardcausa0on)

MethodologicalReduc0onism

Reducingwholestopartsandexplainingthehigherlevelsintermsofloweronesastheul0matedirec0onforallscien0ficresearch(egphysics)

EpistemologicalReduc0onism

Reduc0onofscien0ficknowledge,whetherintermsoftheories,laws,orexplana0ons,fromahigherleveloforganiza0ontothatofalowerormorebasicone

The fall and rise of pharmacology--(re-)defining the discipline? Winquist RJ, Mullane K, Williams M. Biochem Pharmacol. 2014 Jan 1;87(1):4-24. Abstract Pharmacology is an integrative discipline that originated from activities, now nearly 7000 years old, to identify therapeutics from natural product sources. Research in the 19th Century that focused on the Law of Mass Action (LMA) demonstrated that compound effects were dose-/concentration-dependent eventually leading to the receptor concept, now a century old, that remains the key to understanding disease causality and drug action. As pharmacology evolved in the 20th Century through successive biochemical, molecular and genomic eras, the precision in understanding receptor function at the molecular level increased and while providing important insights, led to an overtly reductionistic emphasis. This resulted in the generation of data lacking physiological context that ignored the LMA and was not integrated at the tissue/whole organism level.

However, concerns regarding the disconnect between basic research efforts and the approval of new drugs to treat 21st Century disease tsunamis, e.g., neurodegeneration, metabolic syndrome, etc. has led to the reemergence of pharmacology, its rise, often in the semantic guise of systems biology. Against a background of limited training in pharmacology, this has resulted in issues in experimental replication with a bioinformatics emphasis that often has a limited relationship to reality. The integration of newer technologies within a pharmacological context where research is driven by testable hypotheses rather than technology, together with renewed efforts in teaching pharmacology, is anticipated to improve the focus and relevance of biomedical research and lead to novel therapeutics that will contain health care costs.

1/9/17

6

Neuropharmacology beyond reductionism - A likely prospect. Margineanu DG. Biosystems. 2016 Mar;141:1-9.

Abstract Neuropharmacology had several major past successes, but the last few decades did not witness any leap forward in the drug treatment of brain disorders. Moreover, current drugs used in neurology and psychiatry alleviate the symptoms, while hardly curing any cause of disease, basically because the etiology of most neuro-psychic syndromes is but poorly known. This review argues that this largely derives from the unbalanced prevalence in neuroscience of the analytic reductionist approach, focused on the cellular and molecular level, while the understanding of integrated brain activities remains flimsier. The decline of drug discovery output in the last decades, quite obvious in neuropharmacology, coincided with the advent of the single target-focused search of potent ligands selective for a well-defined protein, deemed critical in a given pathology. However, all the widespread neuro-psychic troubles are multi-mechanistic and polygenic, their complex etiology making unsuited the single-target drug discovery. An evolving approach, based on systems biology considers that a disease expresses a disturbance of the network of interactions underlying organismic functions, rather than alteration of single molecular components. Accordingly, systems pharmacology seeks to restore a disturbed network via multi-targeted drugs. This review notices that neuropharmacology in fact relies on drugs which are multi-target, this feature having occurred just because those drugs were selected by phenotypic screening in vivo, or emerged from serendipitous clinical observations. The novel systems pharmacology aims, however, to devise ab initio multi-target drugs that will appropriately act on multiple molecular entities. Though this is a task much more complex than the single-target strategy, major informatics resources and computational tools for the systemic approach of drug discovery are already set forth and their rapid progress forecasts promising outcomes for neuropharmacology.

Overcoming the Newtonian paradigm: the unfinished project of theoretical biology from a Schellingian perspective. Gare A. Prog Biophys Mol Biol. 2013 Sep;113(1):5-24 Abstract Defending Robert Rosen's claim that in every confrontation between physics and biology it is physics that has always had to give ground, it is shown that many of the most important advances in mathematics and physics over the last two centuries have followed from Schelling's demand for a new physics that could make the emergence of life intelligible. Consequently,

. This history is used to identify and defend a fragmented but progressive tradition of anti-reductionist biomathematics. It is shown that the mathematico-physico-chemical morphology research program, the biosemiotics movement, and the relational biology of Rosen, although they have developed independently of each other, are built on and advance this anti-reductionist tradition of thought. It is suggested that understanding this history and its relationship to the broader history of post-Newtonian science could provide guidance for and justify both the integration of these strands and radically new work in post-reductionist biomathematics.

Holism(Revolu0onarySystemsBiology)

Thelivingworldconsistsinarealitythatcanbeunderstoodonlyinitsglobalandinseparableunity.Thewholeisfundamental,notanyonelevel.Thewholeisgreaterthanthesumofitspartsorofitslevels.

OntologicalHolism

Puingtogetherthepartswillnotproducethewholes(suchaslivingsystems)oraccountfortheirproper0esandbehaviors.Downwardcausa0onclaimsthathigherorderen00esdeterminecausallytheproper0esorbehavioroflower-levelen00es.

MethodologicalHolism

Thatlifecanonlybeunderstoodbystudyingitasawhole.Theworldisdisorderedanditrecognizedthateachhierarchicallevelrequiresitsownresearchstrategynotreducibletothemethodologicalstrategybelowit.

EpistemologicalHolism

Complexwholesareconsiderednottobeunderstandablefromthemereknowledgeofthebehaviorofthepartsinisola0on;onlyproper0esofthesystemasawholemayofferunderstanding.

The new holism: P4 systems medicine and the medicalization of health and life itself. Vogt H, Hofmann B, Getz L. Med Health Care Philos. 2016 Jun;19(2):307-23.

Abstract The emerging concept of systems medicine (or 'P4 medicine'-predictive, preventive, personalized and participatory) is at the vanguard of the post-genomic movement towards 'precision medicine'. It is the medical application of systems biology, the biological study of wholes. Of particular interest, P4 systems medicine is currently promised as a revolutionary new biomedical approach that is holistic rather than reductionist. This article analyzes its concept of holism, both with regard to methods and conceptualization of health and disease. Rather than representing a medical holism associated with basic humanistic ideas, we find a technoscientific holism resulting from altered technological and theoretical circumstances in biology. We argue that this holism, which is aimed at disease prevention and health optimization, points towards an expanded form of medicalization, which we call 'holistic medicalization': Each person's whole life process is defined in biomedical, technoscientific terms as quantifiable and controllable and underlain a regime of medical control that is holistic in that it is all-encompassing. It is directed at all levels of functioning, from the molecular to the social, continual throughout life and aimed at managing the whole continuum from cure of disease to optimization of health. We argue that this medicalization is a very concrete materialization of a broader trend in medicine and society, which we call 'the medicalization of health and life itself'. We explicate this holistic medicalization, discuss potential harms and conclude by calling for preventive measures aimed at avoiding eventual harmful effects of overmedicalization in systems medicine (quaternary prevention).

1/9/17

7

Gene0cDeterminism

Theviewthatgenes(genotype)causetraits(phenotype)

Gene0cdeterminismalsoreferredtoas-Gene0cism,Gene0cEssen0alismandGene0cFatalism

StrongGene0cDeterminism-genotype“always”dictatesphenotypeWeakGene0cDeterminism-genotype“some0mes”dictatesphenotype,also

poten0alsorpredisposi0onsClassicalGene0cs(Mendel)toMolecularGene0cs(DNA)toMolecularBiology

Two applications of network-based analyses of GWAS. (a) GWAS analysis computes the association between a SNP and case/control, reporting a P-value for each SNP. (b) Casual gene identification is the problem of identifying a single causal gene (circled in red) for the phenotype from a larger locus of candidate genes that is significantly associated with the phenotype. (c) Causal network identification is the problem of finding a group of interacting genes (e.g. a signaling pathway or protein complex) containing SNPs that distinguish cases and controls.

1/9/17

8

Systems Genetic Analyses Highlight a TGFβ-FOXO3 Dependent Striatal Astrocyte Network Conserved across Species and Associated with Stress, Sleep, and Huntington's Disease. Scarpa JR, Jiang P, Losic B, Readhead B, Gao VD, Dudley JT, Vitaterna MH, Turek FW, Kasarskis A. PLoS Genet. 2016 Jul 8;12(7):e1006137.

Network-specific pathology and functional characterization of CN Thistle2 module. (A,B) Differential connectivity analysis reveals network-level alterations (light purple) that were not observed by previous differential expression analysis in the same cohort1 (dark purple). (B) Venn diagrams depict the number of genes identified by differential connectivity (light purple) and differential expression analyses (dark purple), as well as their overlap. (C) CN modules showing enrichment for previously published cell-type specific gene signatures identified by FACS (F) and in situ hybridization (I) experiments. Fisher’s exact test odds ratios are plotted only for modules with P < 0.05, two-sided, Bonferroni corrected. (D) Circos plot depicting FOXO3 as the top TF associated with Thistle2 in CN; rings are numbered 1 (outermost) to 5 (innermost). TF binding site enrichment scores are depicted in rings 2, 3, and 4 (Z score, Fisher’s score, and Composite Rank, respectively). Ring 5 depicts the differential expression profile of each TF in HD (-log10(P)). Blue histogram height (ring 1) reflects the cumulative scores of each TF based upon rings 2–5, with taller heights depicts greater relevance to Thistle2.

A genome scan conducted in a multigenerational pedigree with convergent strabismus supports a complex genetic determinism. Georges A, Cambisano N, Ahariz N, Karim L, Georges M. PLoS One. 2013 Dec 23;8(12):e83574

Abstract A genome-wide linkage scan was conducted in a Northern-European multigenerational pedigree with nine of 40 related members affected with concomitant strabismus. Twenty-seven members of the pedigree including all affected individuals were genotyped using a SNP array interrogating > 300,000 common SNPs. We conducted parametric and non-parametric linkage analyses assuming segregation of an autosomal dominant mutation, yet allowing for incomplete penetrance and phenocopies. We detected two chromosome regions with near-suggestive evidence for linkage, respectively on chromosomes 8 and 18. The chromosome 8 linkage implied a penetrance of 0.80 and a rate of phenocopy of 0.11, while the chromosome 18 linkage implied a penetrance of 0.64 and a rate of phenocopy of 0. Our analysis excludes a simple genetic determinism of strabismus in this pedigree.

Weight Stigma Reduction and Genetic Determinism. Hilbert A. PLoS One. 2016 Sep 15;11(9):e0162993.

Abstract One major approach to weight stigma reduction consists of decreasing beliefs about the personal controllability of-and responsibility for-obesity by educating about its biogenetic causes. Evidence on the efficacy of this approach is mixed, and it remains unclear whether this would create a deterministic view, potentially leading to detrimental side-effects. Two independent studies from Germany using randomized designs with delayed-intervention control groups served to (1) develop and pilot a brief, interactive stigma reduction intervention to educate N = 128 university students on gene × environment interactions in the etiology of obesity; and to (2) evaluate this intervention in the general population (N = 128) and determine mechanisms of change. The results showed (1) decreased weight stigma and controllability beliefs two weeks post-intervention in a student sample; and (2) decreased internal attributions and increased genetic attributions, knowledge, and deterministic beliefs four weeks post-intervention in a population sample. Lower weight stigma was longitudinally predicted by a decrease in controllability beliefs and an increase in the belief in genetic determinism, especially in women. The results underline the usefulness of a brief, interactive intervention promoting an interactionist view of obesity to reduce weight stigma, at least in the short term, lending support to the mechanisms of change derived from attribution theory. The increase in genetic determinism that occurred despite the intervention's gene × environment focus had no detrimental side-effect on weight stigma, but instead contributed to its reduction. Further research is warranted on the effects of how biogenetic causal information influences weight management behavior of individuals with obesity.

After geneticization. Arribas-Ayllon M. Soc Sci Med. 2016 Jun;159:132-9.

Abstract The concept of geneticization belongs to a style of thinking within the social sciences that refers to wide-ranging processes and consequences of genetic knowledge. Lippman's original use of the term was political, anticipating the onerous consequences of genetic reductionism and determinism, while more recent engagements emphasise the productivity and heterogeneity of genetic concepts, practices and technologies. This paper reconstructs the geneticization concept, tracing it back to early political critiques of medicine. The argument is made that geneticization belongs to a style of constructionist thinking that obscures and exaggerates the essentializing effects of genetic knowledge. Following Hacking's advice, we need a more literal sense of construction in terms of 'assembly' to give a clearer account of the relationship between processes and products. Using the 'assemblage' concept to explore the social ontology of genetics, the paper reviews three areas of the empirical literature on geneticization - disease classification, clinical practice and biosociality - to show that a new style of thinking has appeared within the social sciences. In the final assessment, the conditions that gave rise to geneticization are now obsolete. While it may serve as a useful ritual of debate, conceptually geneticization offers a limited account of the heterogeneity of socio-technical change.

1/9/17

9

Epigene0cs

Waddington(1940s)coinedtermtodescribeenvironment-geneinterac0onsthatpromotephenotype.

Non-gene0cfactorsinthecontrolofdevelopmentalprocessesandphenotype(?an0-

gene0cdeterminism)ArtRiggs(1996),definedas“mito0callyand/ormeio0callyheritablechangesingene

func0onthatcannotbeexplainedbychangesinDNAsequence”Epigene0csrepresentsformanysystemsbiologistsapromiseforcontrolofbiological

phenomenaunfulfilledbygene0cdeterminism(Silverman2004)

Epigenetics

Molecular factors/processes around the DNA that regulate genome activity, independent of DNA sequence, and are mitotically stable

Epigenetic Mechanisms of Gene Regulation

-DNA Methylation-Histone Modification-Chromatin Structure-DNA Organization into Domains (eg Loops)-Nuclear Compartmentalization (eg nuclear matrix)-Noncoding functional RNAs

1/9/17

10

MechanismandEmergence

Mechanism-Glennan2002-“isacomplexsystemthatproducesthatbehaviorbytheinterac0onof

anumberofparts,wheretheinterac0onsbetweenpartscanbecharacterizedbydirect,invariant,changerela0nggeneraliza0ons”

Machamer,Darden,Craver2000-“arein00esandac0vi0esorganizedsuchthatthey

areproduc0veofregularchangesfromstartorset-uptofinishortermina0oncondi0ons”(AtoBtoC)

Mechanismsareespeciallyopentoinves0ga0onpar0cularlythroughexperimenta0on

Emergence.Complexsystemsdisplayproper0es,o8encalled“emergentproper0es,”thatarenotdemonstratedbytheirindividualpartsandcannotbepredictedevenwithfullunderstandingofthepartsalone.Forexample,understandingtheproper0esofhydrogenandoxygendoesnotallowustopredicttheproper0esofwater.Lifeisanexampleofanemergentproperty.ItisnotinherentinDNA,RNA,proteins,carbohydrates,orlipidsbutisaconsequenceoftheirac0onsandinterac0ons.Acomprehensiveunderstandingofsuchemergentproper0esrequiressystems-levelperspec0vesandcannotbegleanedfromsimplereduc0onistapproaches.

“Whatisthedifferencebetweenalivecatandadeadone?Onescien0ficanswerissystemsbiology.Alivecatistheemergentbehaviorofthesystemincorpora0ngthoseparts.”

1/9/17

11

Emergence of biological organization through thermodynamic inversion. Kompanichenko V. Front Biosci (Elite Ed). 2014 Jan 1;6:208-24.

Emergence, Retention and Selection: A Trilogy of Origination for Functional De Novo Proteins from Ancestral LncRNAs in Primates. Chen JY, Shen QS, Zhou WZ, et al. PLoS Genet. 2015 Jul 15;11(7):e1005391.

De novo protein-coding genes originating from lncRNAs. (A) Computational pipeline for ab inito identification and meta-analysis of de novo genes in the hominoid lineage. (B) Number of de novo genes on the phylogenetic tree, with the branch length proportional to the divergence time. (C) Stacked histogram showing the percentage of de novo gene orthologs that also show expression in chimpanzee or rhesus macaque. (D) Boxplot showing relative expression levels of the transcripts and their nearby regions corresponding to de novo genes (orthologs) in human (chimpanzee or macaque). The nearby regions are defined as upstream and downstream regions with equal length to the corresponding genes. For each region, the relative expression was calculated by normalizing the expression level of this region with the sum of the expression levels of the genic region and the nearby regions. (E) Percentage of splicing junctions with supporting RNA-Seq reads in human, chimpanzee and rhesus macaque. (F) For each pair of tissues, Spearman correlation coefficients were computed separately, and the extent of tissue-specific differences in de novo gene expressions are shown (based on the color scale). Dotted lines highlight parallel comparisons between two different species.

Contextual organismality: Beyond pattern to process in the emergence of organisms. Díaz-Muñoz SL, Boddy AM, Dantas G, Waters CM, Bronstein JL. Evolution. 2016 Dec;70(12):2669-2677.

The cooperation‐conflict space is useful to visualize and evaluate potentially organismal interactions. Panel (A) illustrates organismality space (after Queller and Strassmann 2009) and some of the potential paths (numbered 1–4) organisms can move through under changing ecological contexts, such as development, resource availability, population size, and species interactions. In Panel (B), we provide examples of movement across organismal space in honey bee colonies (blue) and groups of microbial cells (red). In both examples, the cloud plot depicts the movement over “organismality space” and the labels represent the context that facilitates this change. The shading around the points is meant to convey the possibility of small changes in cooperation‐conflict in any context.

1/9/17

12

HomeostasisvsRobustness

Homeostasis-ClaudeBernard(1800s)-“internalmilieu’sconstancy”Cannon(1939)-“steadystatesinthebody…..acondi0onthatmayvary,butis

rela0velyconstant”Miglani(2006)-“amechanismforpromo0ngthestabilityofphenotypicexpressionof

agenotypewhengrownoverawiderangeofenvironments”

Illustration of environmental influences and the effect of perturbations on inner dynamics. In (A), two environments are shown (rich and minimal media). Plots adapted from (Freilich et al., 2010). In (B), a current state of an internal control can be modified by small or large perturbations (thick black arrows) pushing the agent–internal dynamics within the current boundary of attraction or far from it. NN, neural network. See main text for further details.

Robustness.Biologicalsystemsmaintainphenotypicstabilityinthefaceofdiverseperturba0onsimposedbytheenvironment,stochas0cevents,andgene0cvaria0on.Robustnesso8enarisesthroughposi0veandnega0vefeedbackloopsandotherformsofcontrolthatconstrainagene’soutput.Thisfeedbackinsulatesthesystemfromfluctua0onsimposedonitbytheenvironment.Posi0vefeedback,ingeneral,enhancessensi0vity,whereasnega0vefeedbackcandampennoiseandrejectperturba0ons.Robustnessisaninherentpropertyofallbiologicalsystemsandisstronglyfavoredbyevolu0on.

Robustnessasanorganiza3onalprincipleRobustnessenablesthesystemtomaintainitsfunc0onali0esagainstexternalandinternalperturba0ons.Thispropertyhasbeenwidelyobservedacrossmanyspecies,fromthelevelofgenetranscrip0ontothelevelofsystemichomeostasis.

1/9/17

13

1/9/17

14

Illustration of redundancy (A) and distributed robustness (B). Plots show a hypothetical organization in which an upstream signal from the upper white circles is processed by a number of intermediate components (dark circles) to a downstream effector (lower white circles).

1/9/17

15

Modularity.Afurthercharacteris0cofcomplexsystemsistheirmodularity.Mul0pleusefuldefini0onsofamoduleexist.Toanengineer,amoduleisafunc0onalunit,acollec0onofpartsthatinteracttogethertoperformadis0nctfunc0on.Suchamodulewouldhavedis0nctinputs,thingsitissensi0veto,andoutputs,thingsitcontrols.Toabiologist,amoduleinanetworkisasetofnodesthathavestronginterac0onsandacommonfunc0on.Modularitycancontributetobothrobustnessoftheen0resystem,byconfiningdamagetoseparableparts,andtoevolu0on,bysimplyrewiringmodules.Furthermore,modularitydecreasestheriskoffailureofthesystembypreven0ngthespreadofdamageinonepartofthenetworkthroughouttheen0renetwork.

1/9/17

16

1/9/17

17

Evolutionary Systems Biology Revolutionary Systems Biology

Ecosystem

Populations

Organisms

Physiology

Organ Systems

Organs

Tissues

Cells

Organelles

Macromolecules

DNA

Ecosystem

Populations

Organisms

Physiology

Organ Systems

Organs

Tissues

Cells

Organelles

Macromolecules

DNA

1/9/17

18


Ecosystem

Populations

Organisms

Physiology

Organ Systems

Organs

Tissues

Cells

Organelles

Macromolecules

DNA

Ecosystem

Populations

Organisms

Physiology

Organ Systems

Organs

Tissues

Cells

Organelles

Macromolecules

DNA

Linear Stasis Homeostasis Mechanism Genetic Determinism Reductionism

Nonlinear Robustness Synergism Emergence Epigenetics Holism

Actual Systems Biology

Ecosystem

Populations

Organisms

Physiology

Organ Systems

Organs

Tissues

Cells

Organelles

Macromolecules

DNA

Top-down models in biology: explanation and control of complex living systems above the molecular level. Pezzulo G, Levin M. J R Soc Interface. 2016 Nov;13(124).


SystemsBiology


RequiredReading




1/9/17

19


Ecosystem

Populations

Organisms

Physiology

Organ Systems

Organs

Tissues

Cells

Organelles

Macromolecules

DNA

Ecosystem

Populations

Organisms

Physiology

Organ Systems

Organs

Tissues

Cells

Organelles

Macromolecules

DNA

Linear Stasis Homeostasis Mechanism Genetic Determinism Reductionism

Nonlinear Robustness Synergism Emergence Epigenetics Holism

Actual Systems Biology

Ecosystem

Populations

Organisms

Physiology

Organ Systems

Organs

Tissues

Cells

Organelles

Macromolecules

DNA

Computa0onalBiology

•  Mathema0calmodeling•  Datasetanalysistodevelopmodels

1/9/17

20

Computa0onalModels

•  ModelScope(mathema0calelements)

•  ModelStatement(equa0ons)

•  SystemState(dynamic,snapshot)

•  Variables,ParametersandConstants

•  ModelBehavior(environmentalandinternalprocesses)

•  ModelAssignment(biologydescribedmathema0cal)

•  DataIntegra0on(omicsdata)

1/9/17

21

Synthetic biology and regulatory networks: where metabolic systems biology meets control engineering. He F, Murabito E, Westerhoff HV. J R Soc Interface. 2016 Apr;13(117).

An unbranched metabolic pathway under hierarchical regulation. (a) The first enzyme is regulated through both transcriptional repression and allosteric activity inhibition by the end product. Enzymes in other steps might also be regulated through gene expression (in dashed arrows), but this is not explicitly considered here. TF denotes transcription factor. (b) The hierarchical supply–demand representation of the pathway (a). The lower part represents the classical metabolic supply–demand system, in which only the metabolic regulation (in this case allosteric inhibition) is considered. The letter ‘X’ denotes the penultimate product xn, still in pathway. The supply is catalysed by enzyme Es (i.e. e1 here or enzymes stemming from an entire operon). (c) Illustration of the steady-state properties of a supply–demand system in terms of changes in the flux, intermediate concentration and elasticity coefficients.

1/9/17

22

1/9/17

23

ParameterEs0ma0ons

•  Regression(minimumofthefunc0on)•  Es0mators(distancemeasure)•  Maximumlikelihoodes0ma0on(Gaussiannoise)•  Iden0fiability(landscapeinparameterspace)•  Bootstrapping(samplingandnoisydata)•  CrossValida0on(modelfiingandpredic0on)•  BaysianParameterEs0ma0on(parameternotfixed,randomvariables)•  LocalandGlobalOp0miza0on•  MachineLearningAlgorithms(simula0ons)

(Mathema0ca/Matlab/SystemsBiologyMarkupLanguage,SBML)

Patterning with activator-inhibitor systems. (A) Local activation and lateral inhibition generates spatially heterogeneous patterns. (B) Interactions between black and yellow pigment cells produce Turing patterns in zebrafish skin. Mutual inhibition between them functions as self-activation for the yellow cells. Each yellow cell activates distant black cells. Therefore, inhibition of the yellow cell by the black cell works as a lateral inhibition. (C) Different modeling approaches to spontaneous pattern formation.

1/9/17

24

Patterning with signaling gradients. (A) Schematic of early fruit fly embryo showing the maternal gradient of Bicoid protein at cycle 13 that directs the formation of precise target gene domains such as hunchback and knirps. (B) Proposed gene regulatory network showing cross-regulation of target genes (9). The four genes are also under control of Bicoid and other players. t, time.

ParameterEs0ma0ons

•  Regression(minimumofthefunc0on)•  Es0mators(distancemeasure)•  Maximumlikelihoodes0ma0on(Gaussiannoise)•  Iden0fiability(landscapeinparameterspace)•  Bootstrapping(samplingandnoisydata)•  CrossValida0on(modelfiingandpredic0on)•  BaysianParameterEs0ma0on(parameternotfixed,randomvariables)•  LocalandGlobalOp0miza0on•  Gene0cAlgorithms(simula0ons)


ParameterEs0ma0ons



1/9/17

25

ParameterEs0ma0ons



MachineLearningModeling

•  Largedatasetwithmanipula0ons

•  Testdatasetwithknownoutcomesparameters(learningdataset)

•  Mathema0calAlgorithmdevelopmentfromtrainingset

•  RefineAlgorithmdevelopmentwithlargedataset

•  FinalAlgorithmshouldbecorrectwithtrainingsetandrevealnewbiologyinsight

Bifurcation diagrams for the deterministic reaction rate equations. The diagrams are constructed using XPPAUT for equations (1)–(13) and the parameter values given in Results. Numbers of reporter protein molecules produced are plotted against the natural logarithm of the external signal ln(k14/k15), in the a) autoregulated and b) constitutive cases, showing a bistable and graded response respectively. Bold lines denote stable solutions and dashed lines denote unstable solutions. doi:10.1371/journal.pcbi.1002396.g003

1/9/17

26

Methods of information theory and algorithmic complexity for network biology. Zenil H, Kiani NA, Tegnér J. Semin Cell Dev Biol. 2016 Mar;51:32-43.

Abstract We survey and introduce concepts and tools located at the intersection of information theory and network biology. We show that Shannon's information entropy, compressibility and algorithmic complexity quantify different local and global aspects of synthetic and biological data. We show examples such as the emergence of giant components in Erdös-Rényi random graphs, and the recovery of topological properties from numerical kinetic properties simulating gene expression data. We provide exact theoretical calculations, numerical approximations and error estimations of entropy, algorithmic probability and Kolmogorov complexity for different types of graphs, characterizing their variant and invariant properties. We introduce formal definitions of complexity for both labeled and unlabeled graphs and prove that the Kolmogorov complexity of a labeled graph is a good approximation of its unlabeled Kolmogorov complexity and thus a robust definition of graph complexity.

1/9/17

27

ParameterEs0ma0ons



Networks

•  Modules•  Nodes•  Clusters•  Interactomes

1/9/17

28

1/9/17

29

1/9/17

30

1/9/17

31

(a) A network of 750 nodes was generated by means of the PS model, with target average node degree 2m = 10, scaling exponent γ = 2.75 and network temperature T = 0. The network is embedded to the hyperbolic plane H2 An external file that holds a picture, illustration, etc. Object name is srep30108-m31.jpg with LaBNE to reveal the angular position of the nodes in the hyperbolic circle containing the network. (b) Finally, the radial coordinates of the nodes are assigned, so that they resemble the rank of each node according to its degree. By the colour of the nodes, which highlights their angular coordinates, one can note that the embedding by LaBNE is rotated by some degrees with respect to the actual node angular coordinates obtained with the PS model. This does not impact the hyperbolic, distance-dependent connection probabilities, because distances are invariant under rotations. Edges in the raw embedding by LaBNE are not shown for clarity.

Note that to embed a network G to H2, the truncated spectral decomposition of L is used. This gives the closest approximation to the eigen-decomposition by a matrix λk+1 of rank λ k + 1 and ensures that the computational complexity of LaBNE is O(N2).

Efficient embedding of complex networks to hyperbolic space via their Laplacian. Alanis-Lobato G, Mier P, Andrade-Navarro MA. Sci Rep. 2016 Jul 22;6:30108.

1. Feedbackisanessen0alpartofmolecularnetworks.Itallowsthecelltoadjusttherepertoireoffunc0onalproteinstocurrentneeds.

2. AFLisprimarilycharacterizedbyitssign:nega0vefeedbackformaintaining

homeostasis,posi0vefeedbackforobtainingultrasensi0vityormul0plestablestatesofthecellularcomposi0on.

3. Nega0vefeedbackcancauseoscilla0onsifsignalpropaga0onaroundtheFLis

sufficientlyslow.HighHillcoefficients,addi0onalposi0veFLs,orsaturateddegrada0onfacilitatesoscilla0onsinanega0veFL.

4. Posi0vefeedbackcancomefromstrongself-ac0va0onofagene,frommutual

repressionbetweenproteins,orbyautocataly0cprocesses.Inallcasesonecanobtainbistabilityifreac0onsinvolvesomesortofcoopera0vity.

5. Metabolismofsmallmoleculesischaracterizedbyasepara0onofscales.Typically,

theintracellularpoolofavailablesmallmoleculesismuchsmallerthanthetotalamountofsmallmoleculesconsumedduringonecellgenera0on.

6. Combina0onsofFLsinsmall-moleculeuptakeandmetabolismcanresultinnew

behavioralfeaturesthataresignificantlydifferentfromasimplesumofthebehaviorsofsingleloops.

SummaryPoints

1/9/17

32

1/9/17

33

Transcription network in bacteria

NetworkApplica0onExamples(a) Gene regulatory network for Drosophila gap genes, showing relationship between input genes (Bcd, Cad, Hb, Tll) and output genes (Kni,Hb,Kr,Gt). (After figure 1 of Papatsenko and Levine (2011)). (b) Concentration of Gap genes along the anterior posterior axis of the embryo. Model was fitted to this data. Hb, hunchback; Gt, giant; Kr, Kruppel; Kni, Knirps.

1/9/17

34

The (r)evolution of gene regulatory networks controlling Arabidopsis plant reproduction: a two-decade history.

A schematic of the network perturbations of one neural degenerative network over the 20 weeks of the progression of this disease in a mouse model. The red nodes indicate mRNAs that have become disease perturbed as compared with the brain transcripts of normal mice. The spreading of the disease-perturbed networks at the three different times points is striking – indicating the progressive disease perturbation of this neurodegenerative network.

Systems biology and gene networks in neurodevelopmental and neurodegenerative disorders. Parikshak NN, Gandal MJ, Geschwind DH. Nat Rev Genet. 2015 Aug;16(8):441-58.

1/9/17

35

Use of comparative genomics approaches to characterize interspecies differences in response to environmental chemicals:

challenges, opportunities, and research needs.

Interspecies comparisons and human relevance — a modified parallelogram approach for integrating in vivo, in vitro, and computational approaches in interspecies extrapolation of toxicity perturbation. The parallelogram approach proposed by Nesnow (2004) and referred to by the National Research Council ( National Research Council, 2006) is modified here by the incorporation of computational comparative genomics approaches. Using rat and human as examples: 1) a rat network perturbation model is developed based on in vivo data; 2) the rat and human networks are computationally compared; 3) differences and similarities found by the interspecies network comparison are tested via human in vitro assays (e.g., primary human cell lines); 4) quantified in vitro perturbations are mapped back to the compared networks; and, 5) human in vivo outcomes are inferred. In addition, rat in vivo assays, driven by network-based hypotheses or otherwise (as represented by the white arrows), can inform the rat network model and the compared network model.

1/9/17

36

ColinMacilwain

SystemsBiology:EvolvingintotheMainstream

CellVolume144,Issue62011839-841h\p://dx.doi.org/10.1016/j.cell.2011.02.044

ColinMacilwainSystemsBiology:EvolvingintotheMainstreamCellVolume144,Issue62011839-841h\p://dx.doi.org/10.1016/j.cell.2011.02.044

spring 2017 – epigenetics and systems biology lecture ...2014) systems-based technologies in...

Documents