CG020 Genomika Bi7201 Základy genomiky 10. Systémová biologie 10. Systems biology Kamil Růžička Funkční genomika a proteomika rostlin, Mendelovo centrum genomiky a proteomiky rostlin, Středoevropský technologický institut (CEITEC), Masarykova univerzita, Brno kamil.ruzicka@ceitec.muni.cz, www.ceitec.muni.cz §What is systems biology §System theory §Omics §Reductionism vs. holism §Networks §Modular concept §Regulation of gene expression – example task for systems biology §Gene regulation X->Y §Transcriptional network of E. coli §Negative autoregulatory networks §Robustness of negative autoragulatory networks §(Positive autoregulatory networks) § § § Přehled •fashionable catchword? •a real new (philosophical) concept? •new discipline in biology? •just biology? • • What is systems biology http://www.psb.ugent.be/images/stories/psb/home/building_colorised.png •fashionable catchword? •a real new (philosophical) concept? •new discipline in biology? •just biology? • • What is systems biology http://www.psb.ugent.be/images/stories/psb/home/building_colorised.png http://www.ceitec.eu/programs/genomics-and-proteomics-of-plant-systems/ •The behavior of a system depends on: • •Properties of the components of the system •The interactions between the components • Systems theory •The behavior of a system depends on: • •Properties of the components of the system •The interactions between the components • Systems theory Forget about reductionism, think holistically. ὅλος [hol'-os] – greek. all, the whole, entire, complete Systems biology • •Systems theory and theoretical biology are old • •Experimental and computational possibilities are new meeting of old and new Ludwig von Bertalanffy (1901-1972) msRKuTxT3vpASBVH sK8rEh6vIGs8L7eaks0b6PxyQtHSfyHkmp Qffs+v35leo5GGLA205qFb14dL9rsU3CxDWxOL2fZI5Siam5PTzTH5vbhjRG6eeLiqmw4WDqsdM= Omics-revolution shifts paradigm to large systems - Integrative bioinformatics - (Network) modeling Two roots of systems biology Palsson 2007 Associated disciplines ¢Genomics ¢Epigenomics ¢Transcriptomics ¢Translatomics / Proteomics ¢Interactomics ¢Metabolomics ¢Fluxomics ¢NeuroElectroDynamics ¢Phenomics ¢Biomics ¢ Associated disciplines ¢Genomics ¢Epigenomics ¢Transcriptomics ¢Translatomics / Proteomics ¢Interactomics ¢Metabolomics ¢Fluxomics ¢NeuroElectroDynamics ¢Phenomics ¢Biomics ¢ Jozef Mravec’s term: multidimensional biology How I understand systems biology ¢Genetics – you have one or few RNA processing genes where you show their importance in protoxylem development ¢Functional genomics – you find in e.g protoxylem expression profiles numerous RNA processing genes and demonstrate which are important for protoxylem developments ¢Systems biology – based on obtained large scale data you propose model how genes (and/or other components) collectively regulate protoxylem development How I understand systems biology ¢Good biology – you explain why just some genes regulate protoxylem development (sorry for aphorisms) Reconstructed genome-scale networks Species #Reactions #Genes Reference Escherichia coli 2077 1260 Feist AM. et al. (2007), Mol. Syst. Biol. Saccharomyces cerevisiae 1175 708 Förster J. et al. (2003), Genome Res. Bacillus subtilis 1020 844 Oh YK. et al. (2007), J. Biol. Chem. Lactobacillus plantarum 643 721 Teusink B. et al., (2006), J. Bio. Chem. Human 3673 1865 Duarte NC. et al., (2007), PNAS Arabidopsis Arabidopsis Interactome Mapping Consortium (2011), Science Reconstructed genome-scale networks Species #Reactions #Genes Reference Escherichia coli 2077 1260 Feist AM. et al. (2007), Mol. Syst. Biol. Saccharomyces cerevisiae 1175 708 Förster J. et al. (2003), Genome Res. Bacillus subtilis 1020 844 Oh YK. et al. (2007), J. Biol. Chem. Lactobacillus plantarum 643 721 Teusink B. et al., (2006), J. Bio. Chem. Human 3673 1865 Duarte NC. et al., (2007), PNAS Arabidopsis Arabidopsis Interactome Mapping Consortium (2011), Science Complexity of cellular networks in E. coli Sometimes the things are different than we just think •Gene expression networks: based on transcriptional profiling and clustering of genes • •Protein-protein interaction networks (Y2H, TAP etc). • •Metabolic networks: network of interacting metabolites through biochemical reactions. Reconstruction of networks from -omics for systems analysis Reconstruction of networks from -omics for systems analysis •Gene expression networks: based on transcriptional profiling and clustering of genes • •Protein-protein interaction networks (Y2H, TAP etc). • •Metabolic networks: network of interacting metabolites through biochemical reactions. How to simplify. Modularity concept. Lets e.g. assume that transcription and translation is one module. E. coli Generation time 20 min Description of gene regulation Description of gene regulation [units [time-1] Description of gene regulation [units [time-1] Description of gene regulation cells grow protein is degraded + [units [time-1] Description of gene regulation 1. Steady state – ustálený stav t Yst Y 2. Production of Y stops: (log => ln [.CZ]) 2. Production of Y stops: (log => ln [.CZ]) Large 𝛼 → rapid degradation Stable proteins (most of E. coli proteins) τ – cell generation Stable proteins τ – cell generation Response time is one generation. 3. Production of Y starts from zero ß t 3. Production of Y starts from zero 3. Production of Y starts from zero Y grows almost linearly initially (magic) 3. Production of Y starts from zero ¢Response time: The same response time as in case 2. Response time does not depend on production rate! 3. Production of Y starts from zero ¢Response time: Not many degradation mechanisms in E. coli (energy consuming). Perhaps in plants? Networks http://www.igentribe.com/wp-content/uploads/2011/02/facebook_network.jpg node (CZ: uzel) thread (CZ: hrana) Transcriptional network of E. coli http://biologos.org/uploads/static-content/louis_fig_4_2.jpg 420 nodes, 520 edges How may self-edges? (CZ: samohrana?) Likelihood of the self-edge Autoregulation is a network motif http://biologos.org/uploads/static-content/louis_fig_4_2.jpg 420 nodes, 520 edges. 40 self-edges! Autoregulation is a network motif E. coli:40 autoregulatory loops: 36 negative, 4 positive http://2011.igem.org/wiki/images/0/0c/Negative_promoter_general.png positive regulation negative regulation Negative autoregulatory loop is best described by Hill’s function ß(Y) Y Negative autoregulatory loops Hill’s function ß1/2 ß(Y) Y n ßmax K Negative autoregulatory loops Hill’s function ßmax ß(Y) K maximum production rate ß1/2 Y n steepness (Hill’s function) concentration of Y needed for 50 % repression Negative autoregulatory loops Hill’s function ßmax ß(Y) K maximum (initial) production rate ß1/2 Y n Negative autoregulatory loops ß synthesis rate – stochastic noise ß(Y) Y ß may vary by 10 - 30 % (other parameters stable) (stochastický ruch) Negative autoregulatory loops Hill’s coeficcient ß1/2 ß(Y) Y n K •varies between 1 – 4, the higher the steeper •important factor: multimerization Negative autoregulatory loops K – repression coefficient ß1/2 ß(Y) Y n K •depends on chemical bonds between Y and its binding sites •a point mutation can increase K ~10 times ATG Y Y NNNNN concentration of Y needed for 50 % repression Positive autoregulatory loops Hill’s function ß1/2 ß(Y) Y n ßmax K Back to simple regulation production (ß) Y Yst Example: if ß=0 (no production) Y Yst=0 production (ß) removal > production Y Yst production > removal production (ß) Y Yst the longer the distance, the faster the change ¢Therefore more difficult to come to Yst with time Yst Y t production (ß) Y Yst Autoregulation vs. simple production Comparison ¢Lets assume that these values are the same: ¢ 1. Yst ¢ 2. 𝛼 ¢ ¢ Comparison ¢Lets assume that these values are the same: ¢ 1. Yst ¢ 2. 𝛼 ¢ ¢Lets put it in one graph. production (ß) Y Yst ß ßmax in such case, always ßmax>ß Autoregulation vs. simple production Autoregulation vs. simple production production (ß) Y Yst ß ßmax the distance always more far => the reactions are faster Response time was confirmed indeed faster (~5 times) time Y simple regulation negative autoregulation Alon 2007 Cases of sharp curve Y Yst ß ßmax Cases of sharp curve Y Yst ß ßmax Cases of sharp curve production (ß) Y Yst ß ßmax Fluctuations in synthesis or removal don’t change much. Cases of sharp curve production (ß) Y Yst≈K ß ßmax Yst depends only on K – on protein-DNA binding properties. Conclusions ¢Negative autoregulation ¢speeds up response time ¢is robust (for 𝛼, ß) => basically on/off ¢bypasses stochastic noise production (ß) Y Yst ß ßmax Conclusions ¢The model explains why negative autoregulation is a common network motif in E. coli. production (ß) Y Yst ß ßmax Conclusions ¢The model explains why negative autoregulation is network motif. ¢ ¢We will not avoid mathematics in biology. Positive autoregulation leads to slower response Positive autoregulation leads to higher variation extreme case => increasing cell-cell variability • less extreme case Positive autoregulation leads to higher variation extreme case Strong variation: -=> differentiation of cells into 2 populations (development) -=> memory for maintaining gene expression (development) -helps with maintaining mixed phenotype for better response to changing environment §Source literature §http://www.youtube.com/watch?v=Z__BHVFP0Lk and further – excellent talks about systems biology from Uri Alon (Weizman Institute) §Rosenfeld N, Negative autoregulation speeds the response times of transcription networks. J Mol Biol. 2002 Nov 8;323(5):785-93. – experimental testing of the data §Alon U. Network motifs: theory and experimental approaches. Nat Rev Genet. 2007 Jun;8(6):450-61. Review about the same. §Alon, U. (2006). An Introduction to Systems Biology: Design Principles of Biological Circuits (Chapman and Hall/CRC). §Palsson, B.Ø. (2011). Systems Biology: Simulation of Dynamic Network States (Cambridge University Press). Most common textbook about systems biology, § § § § Literature §For enthusiasts § §Zimmer (2009). Microcosm- E Coli & the New Science of Life (Vintage) (popular scientific book about E. coli as model organism and what you probably didn’t know) §Albert-László Barabási (2005) V pavučině sítí. (Paseka) (znamenitá kniha o matematice sítí, dynamicky se rozvíjejícím oboru od předního světového vědce) §PA052 Úvod do systémové biologie, Přednášky. Fakulta Informatiky MU §http://sybila.fi.muni.cz/cz/index - obor na fakultě informatiky. § Reductionism vs. holism Stochastic noise (stochastický ruch) Flux balance analysis (FBA) C:\Work\tmp1.JPG A B C Constraints set bounds on solution space, but where in this space does the “real” solution lie? 8.JPG FBA: optimize for that flux distribution that maximizes an objective function (e.g. biomass flux) – subject to S.v=0 and αj≤vj≤βj Thus, it is assumed that organisms are evolved for maximal growth -> efficiency! Flux balance analysis (FBA) C:\Work\tmp1.JPG A B C Constraints set bounds on solution space, but where in this space does the “real” solution lie? 8.JPG FBA: optimize for that flux distribution that maximizes an objective function (e.g. biomass flux) – subject to S.v=0 and αj≤vj≤βj Thus, it is assumed that organisms are evolved for maximal growth -> efficiency! PA052 Úvod do systémové biologie Metagenomics ¢