ROM P P ETRI ETRI N N ETS ETS TO TO P AR TIAL D D IF NTIAL E E QUATI - PowerPoint PPT Presentation

CFSB W ORKSHOP , B ALLIOL C OLLEGE , M ARCH 19 TH 2012 PN & BioModel Engineering F RO ROM P P ETRI ETRI N N ETS ETS TO TO P AR TIAL D D IF NTIAL E E QUATI TIONS ARTIAL IFFERENTIAL AND BE AND BEYOND - B IO - IO M OD EL E E NGIN RING FO FOR M M ULTI ULTI - SC ALE S S YS YSTEMS B B IOL OGY - - ODEL NGINEE EERIN SCALE IOLOGY Monika Heiner Brandenburg University of Technology Cottbus David Gilbert Brunel University, Uxbridge/London monika.heiner@tu-cottbus.de March 2012

O UTLINE PN & BioModel Engineering � B ACKGROUND -> modelling, what for ? -> how many model types do we need ? -> some case studies � F RAMEWORK -> unifying paradigms: QPN - SPN - CPN � C OLOUR AND M UTLI - SCALE S YSTEM -> replication -> encoding space � S UMMARY & O UTLOOK -> open problems -> next steps monika.heiner@tu-cottbus.de March 2012

PN & BioModel Engineering B ACKGROUND monika.heiner@tu-cottbus.de March 2012

M ODELS IN S YSTEMS B IOLOGY PN & BioModel Engineering NG = FO FORMAL KN LEDGE RE MOD MODELLI LLING KNOWLE REPRE RESENTAT ATION formalizing wetlab understanding experiments observed behaviour natural model model biosystem validation (knowledge) predicted analysis model-based behaviour simulation experiment design MODEL VA VALIDATION = = CO CONFIDENCE INCR EASE MODE INCREASE monika.heiner@tu-cottbus.de March 2012

M ODELS IN S YSTEMS B IOLOGY PN & BioModel Engineering NG = FO FORMAL KN LEDGE RE MODELLI MOD LLING KNOWLE REPRE RESENTAT ATION DESCRIPTIVE DESCRIPTIV formalizing wetlab understanding experiments observed behaviour natural model model biosystem validation (knowledge) predicted analysis model-based EXPLANATO TORY behaviour simulation experiment design PREDICTIV PR EDICTIVE MODEL VA VALIDATION = = CO CONFIDENCE INCR EASE MODE INCREASE monika.heiner@tu-cottbus.de March 2012

M ODELS IN S YNTHETIC B IOLOGY PN & BioModel Engineering NG = BL EPRINT FO FOR SY STEM CON MOD MODELLI LLING BLUEP SYSTEM CONSTR TRUCTIO UCTION construction model synthetic design (blueprint) biosystem desired verification behaviour predicted observed validation behaviour behaviour verification RELIABLE AND AND RO ROBUST ENGINEERING ENGINEERING RE RES VE VERIFIED MODELS MODELS RE REQUIRE monika.heiner@tu-cottbus.de March 2012

PN & BioModel Engineering W HAT KIND OF MODEL SHOULD BE USED ? ( BIOCHEMICAL NETWORKS ) monika.heiner@tu-cottbus.de March 2012

N ETWORK R EPRESENTATIONS , E X 1 PN & BioModel Engineering ? R R eceptor eceptor S e.g. 7-TM e.g. 7-TM R R cell m cell m em em brane brane C � � � � � � � I tyrosine tyrosine � � � � R R R as as as T shc shc shc AdC AdC AdC AdC yc yc yc yc SO SO SO S S S � � � kinase kinase � � � � � � � � � R R R as as as N cAM P Akt Akt Akt cAM cAM P P heterotrim heterotrim eric eric grb2 grb2 grb2 R R R ac ac ac R R R ap1 ap1 ap1 G G -protein -protein G G G EF EF EF A A A A A TP TP TP TP cA cA cA cA cA M M M M M P P P P P ? PI-3 PI-3 PI-3 R R R af-1 af-1 af-1 cA cA cA M M M P P P cA cA cA cA M M M M P P P P M K K K PKA PKA PKA PKA cA cA cA cA M M M M P P P P PAK PAK PAK B-R B-R B-R af af af cA cA cA cA cA M M M M M P A P A P A P A P M M M M P P P P E Y PKA PKA PKA PKA T S I M M M EK EK EK PD PD PD PD E E E E L M M M EK1,2 EK1,2 EK1,2 I L ER ER ER K1,2 K1,2 K1,2 B A A ER ER ER K1,2 K1,2 K1,2 cytosol cytosol S M Y M M M KP KP KP R L A O N ? N transcription transcription transcription F A O factors factors factors I T U C E X E nucleus nucleus monika.heiner@tu-cottbus.de March 2012

N ETWORK R EPRESENTATIONS , E X 2 PN & BioModel Engineering READABILITY CAUSALITY ? E ? R U T E C U R T S U Q I N U monika.heiner@tu-cottbus.de March 2012

B IO N ETWORKS , SOME P ROBLEMS PN & BioModel Engineering � knowledge -> PROBLEM 1 -> uncertain -> growing, changing -> distributed over independent data bases, papers, journals, . . . � various, mostly ambiguous representations -> PROBLEM 2 -> verbose descriptions -> diverse graphical representations -> contradictory and / or fuzzy statements � network structure -> PROBLEM 3 -> tend to grow fast -> dense, apparently unstructured -> hard to read MODELS ARE PATCHWORKS FULL OF ASSUMPTIONS monika.heiner@tu-cottbus.de March 2012

B IO N ETWORK R EPRESENTATIONS SHOULD BE PN & BioModel Engineering � readable & unambigious -> fault avoidant model construction � various abstraction levels � locality - causality - concurrency � compositional � executable -> to experience the model, spec. causality � analysable, with unifying power -> formal = mathematical representations -> high-level description for various analysis approaches � as simple as possible -> how many model types do we need ? monika.heiner@tu-cottbus.de March 2012

M ODELLING = A BSTRACTION PN & BioModel Engineering � hierarchical organisation of components -> model variables genes, molecules, organelles, cells, tissues, organs, organisms � functionality of atomic events chemical reactions with/out stoichiometry, conformational change, transport, . . . � time qualitative versus quantitative models � individual vs population behaviour � (hierarchical) space � observables � shape and volume of components � biosystem development monika.heiner@tu-cottbus.de March 2012

B IO N ETWORKS PN & BioModel Engineering .. . 2 NAD + + 2 H 2 O -> 2 NADH + 2 H + + O 2 ARE NETWORKS 2 NAD + 2 NADH OF BIOCHEMICAL 2 H + 2 H 2 O REACTIONS O 2 hyper-arcs . . . NATURALLY NADH NAD + 2 2 EXPRESSIBLE AS 2 H + r1 P ETRI NETS 2 H 2 O O 2 monika.heiner@tu-cottbus.de March 2012

P LACES , T RANSITIONS - S OME B IO I NTERPRETATIONS PN & BioModel Engineering � places -> model variables -> (bio-) chemical compounds -> proteins -> protein conformations -> complexes -> genes, . . . etc. . . . in different locations � transitions -> atomic events -> (stoichiometric) chemical reaction -> complexation / decomplexation -> phosphorylation / dephosphorylation -> conformational change -> transport step, . . . etc. . . . in different locations monika.heiner@tu-cottbus.de March 2012

PN & BioModel Engineering B IO P ETRI NETS - S OME E XAMPLES monika.heiner@tu-cottbus.de March 2012

E X 1 - Glycolysis and Pentose Phosphate Pathway PN & BioModel Engineering [Reddy 1993] 4 Ru5P Xu5P 5 S7P E4P 6 7 8 2 NADPH 2 GSSG GAP F6P R5P 2 3 1 2 NADP + 4 GSH ?? ?? ?? ?? 9 10 11 12 Gluc G6P F6P FBP GAP 13 14 ATP ADP ATP ADP NAD + DHAP + Pi 15 NAD + NADH NADH ATP ADP ATP ADP 20 19 18 17 16 Lac Pyr PEP 2PG 3PG 1,3-BPG monika.heiner@tu-cottbus.de March 2012

E X 1 - Glycolysis and Pentose Phosphate Pathway PN & BioModel Engineering [Reddy 1993] 4 Ru5P Xu5P 5 S7P E4P 6 7 8 2 NADPH 2 GSSG GAP F6P R5P 2 3 1 2 NADP + 4 GSH -> INT -> TION ? ? INTERPRETA TATIO 9 10 11 12 Gluc G6P F6P FBP GAP 13 14 ATP ADP ATP ADP NAD + DHAP + Pi 15 NAD + NADH NADH ATP ADP ATP ADP 20 19 18 17 16 Lac Pyr PEP 2PG 3PG 1,3-BPG monika.heiner@tu-cottbus.de March 2012

E X 1 - Glycolysis and Pentose Phosphate Pathway PN & BioModel Engineering [Reddy 1993] Xu5P 4 [Heiner 1998] E4P S7P Ru5P 6 7 8 ATP . . . GSSG NADPH GAP F6P [Koch, 5 R5P 2 2 1 3 Heiner 2010] ADP 2 2 2 NADP+ GSH Pi F6P Gluc FBP GAP 12 9 10 11 13 G6P 14 NAD+ ATP ATP ADP ADP DHAP Pi 15 NADH NAD+ NADH ATP ADP ATP ADP 20 19 18 17 16 1,3-BPG Lac Pyr 2PG PEP 3PG monika.heiner@tu-cottbus.de March 2012

E X 2 - A POPTOSIS IN M AMMALIAN C ELLS PN & BioModel Engineering Fas-Ligand Apoptotic_Stimuli s7 Procaspase-8 FADD Bax_Bad_Bim Apaf-1 Bcl-2_Bcl-xL s8 BidC-Terminal CytochromeC Bid s1 dATP/ATP s9 s6 s10 Mitochondrion s5 Caspase-8 Procaspase-3 s2 (m20) Caspase-9 Caspase-3 s13 s11 Procaspase-9 s3 DFF CleavedDFF45 (m22) s12 DFF40-Oligomer s4 DNA DNA-Fragment [H EINER , K OCH , WI LL 2004] [GON 2003] monika.heiner@tu-cottbus.de March 2012

E X 3 - B IOSENSOR PN & BioModel Engineering tf positive feedback TF|S TF + S TF|S tf phzMS PhzMS PCA PYO [G ILBERT , H EINER , R OSSER , F ULTON , G U , T RYBILO 2008] monika.heiner@tu-cottbus.de March 2012

E X 3 - B IOSENSOR PN & BioModel Engineering TF TF expression 1 TF degradation 2 1' pfb POSITIVE signal 4 3 FEEDBACK TFS disassociation TFS association 5 TFS degradation TFS reporter reporter expression 6 reporter degradation 7 precursor response 8 9 response degradation response production [G ILBERT , H EINER , R OSSER , F ULTON , G U , T RYBILO 2008] monika.heiner@tu-cottbus.de March 2012