Monday 22 March 2010

Predicting Novel Metabolic Pathways in Genome-Scale Networks.

Stefan Schuster

Need systematic methods.

Decompose network into pathways. Need stoichiometries and reversibility.
Elementary modes are minimum sets of enzymes that can operate at steady state. Enzymes are weigted by their flux. Clipping of complete space for irreversibility to create a pyramid in high dimensions.

A connected route is not enough to infer a process occurs at steady state you need to do elementary mode analysis. Bioinformatics Schuster paper and discussion with other group. Other group defending graph theoretical models.

All metabolites that are involved in more than 4 reactions are specified as external metabolites.

Genome Research 19 2009 1872-1883 can work on sub-systems of the complete genome model so long as you perform a consistency check.

Modularity reduces the combinatorial problem of checking networks for all elementary modes. Not consistent if in the complete model the reagants used in the synthesis of a product are generated by the subsequent reactions of the product.

Cell Signalling in Space and Time

Kholodenko

Src gradients are the same as Belousov Zhabotinsky. Bistable switches - hysteresis - different paths for the up and down cycles.

Problems with kinase cascades in large cells as these signals are opposed by phosphatases in the cytoplasm. Munoz-Garcia PLoS Comp Biol 5 e1000330

Solution to the French Flag problem for size does the signal propagate to the nucleus? If it does not then the cell is too large.

Moonlighting signalling the same as moonlighting enzymes. Wave front propagation as this makes fast signalling - reaction diffusion much faster than diffusion.

Different signals are obtained from different temporal profiles even if they use the same kinase pathway Marshall C.J. Cell 80 179-185. 1995.

Modelling does not always help understanding why is there different transient activity and not persistent vs transient actvity? Nakakuki et al Cell in Press


Analysing Genome Scale Metabolic Models

David Fell

Create the soichiometry matrix for the complete metablism from the reaction lists from the annotated genomes. Then you can work out the null space matrix for the steady states (actually superset of steady states) which corresponds to the matrix which when multiplied by the stoichiometric matrix will give you the vertical matrix of the velocities (all zero at the steady states).
Calculate complete picture of fluxes at steady state by measuring some fluxes, and using these as multiplier of the  Flux analysis. Can simplify the matrix for those rows that are the same in the null space. Can also remove rows where they are all zero as these are dead enzymes in the steady state. Then you can add extra constraints - Pallson's Flux Balance Analysis.

When using a cost function it is called linear programming.

[Mathematically you could use Lagrangian Multipliers to test cost function and where it intersects with model space.]

  • Transport - need to find which molecules can be exchanged with the environment - genome analysis should reveal this.
  • Need to check for mass balance
  • Remove dead reactions
  • ScrumPy tests the stoichiometry matrix
  • Use PRIAM to improve annotation
  • Use simulated annealing to sample EC classification when it is ambiguous


Wednesday 1 July 2009

Network based prediction of human tissue specific metabolism

Tomer Shlomi****

Metabolism is the essence of life. Metab0lic diseases are very common and cancer has abnormal metabolic phenotype. Metabolic networks are the best understood networks (??! what about moon-lighting and localisation and modification and allostery and ...).
Nodes are substrates edges are reactions.
Modelling Network Function
  • Kinetics - small scales need few nodes need ks
  • Topological analysis - graph theory - abstract
  • Boolean
  • Constraints based models - optimisation theory - steady state behaviour.
Constraint based modelling from Palsson (how is this related to MCA of Fell?)
Treat as series of black boxes with input (medium and nutrients) output is the biomass, we have chemical limits from thermodynamics and the matter constraints.

Multicellular means the different cell metabolisms pass inputs and outputs to each other through the medium between cells. Combine metabolic data with other data which is tissue specific - gene expression and single cell metabolic flux data. Concentrated on gene expression. RNA and protein levels are not well correlated because of translation rate and other factors. Protein level not well correlated to flux depends on degradation allostery etc.
Bilu 2006
Schuster et al
Fong et al

Look for consistency between flux and the gene expression data based on paths and outputs to check those paths are on that produce the outputs. Can have high expression but the control is at the metabolite level (Fell says this is the general case).

http://www.cs.technion.ac.il/~tomersh

The Comparative Analysis Reveals Indepdendence of Developmental Process During Early Development in Frogs

Eugenia Maria del Pino ****

All flora and fauna on the Galapagos had to get there by drifting as they are 1000 km from the mainland. There are no frogs in the Galapagos - barriers to many species. There are 443 species of frogs in Ecuador. High species diversity 10-20% of all vertebrate species.

Can only study the development of frogs that breed in captivity. Model organism frog is Xenopus laevis - development better known than that of human beings. This is from South Africa. Three types of frogs studied in Ecuador including foam-nesting and marsupial. Foam-nesting frog develops rapidly eggs deposited in foam away from water. Different predations and environmental pressures that need rapid development.

Frogs beat a jelly into foam with the fertilised eggs inside. Two days after laying the tadpoles hatch. The embryos are white so they are camoflaged. Terrestrial nesting frogs and marsupial frogs develop more slowly. Dendrobatid frogs - terrestrial nesting include poison arrow frogs are territorial. Find a nest and then calls a female. Embryos cared for by father for 20 days tadpoles attach to the father who transports them back to the water. Pouch in marsupial frogs is on the back - male pushes the eggs into the pouch with his feet. Incubation is for four months then the female gives birth by opening the pouch. Physiology is similar to pregnancy. These are large eggs and can be 1cm in diameter whereas xenopus is 1.2 mm. Have an embryonic disk which is similar to chick development.

Development starts at the dorsal blastocore lip which grows to encircle the yolk plug and then the embryo will surround it. The egg has a polarity from the start - it knows where dorsal is. Needs to change from spherical to elongated. Brachyury is an important gene for development. Work carried our by Spemann-Mangold to identify organiser that if it is not properly located causes twining.

Pino paper


Goosecoid expressed in the dorsal blastocore lip. Identified as an organiser gene. Then Lim1 also identified so that there are manu different genes for the dorsal and even the ventral side of the egg.

Slow egg development means the slow elongation of the Archenteron. Lim1 and Brachyury are important genes. Developed antibodies for transcription factors Lim1 and Brachyury show they are nuclear. Slow development extension of the notochord only occurs after closure of the blastocore. So development is related to evolutionary properties. There is less pressure to form the elongated mobile tadpole and so elongation is delayed.

Computational comparative approaches are valid but there is a lack of molecular data in more exotic species.

Tuesday 30 June 2009

New Challenges and Opportunities in Newtwork Biology

Trey Ideker *****

Richard Karp and Lee Hood are his science inspiration.
Wants to reconstruct pathways from multiple biological sources.

If you know something about a pathway you can perturb it systematically in silico and in vivo/vitro. Start was the Galactose metabolism pathway and its transcriptional control. 3 core genes. 1000 other genes were perturbed in the mutants. Transcriptional interactions were available first, now we have protein-protein interactions and we also had KEGG/metabolism.
Coloured networks - states, expression profiles, phosphorylation etc. extract sub-networks from the main network based on colour - pathway extraction.

http://linkinghub.elsevier.com/retrieve/pii/S0092867408009525
Used phenotype for colouring the invastion of HIV.

What is the goal for the next 10 years? (What do networks actually mean?). Cell is a hairball inside in terms of its interactions.

Developed PathBLAST and NetworkBLAST using the analogy of genome assembly for network assembly. Align the orthologues between species to align the network. Works at the protein family level - removes paralogues. They are not the best algorithms now.

Moves to gene relation networks based on synthetic lethality or phenotype changes for double mutants. Gene interaction networks are logical networks (Booleans). Very little overlap between physical and genetic networks. Gene interactions are between subgraphs that are physical interactions.

http://www.cytoscape.org

ChiP-chip looking at DNA damage caused by stress and DNA repair. Finding where a TF binds by immunoprecipitation - identify the fragments that were bound to the immunoglobulin purified TFs. Problems of drift in that binding might not be significant - non-functional binding. Validated byt checking downstream loss in gene deletions but only 10% of those sites identified from ChiP-chip are verified. Spurious interactions take place close to telomeres closer than 25 kilo-base-pairs. This is condition specific so adding rapamycon can perturb it. Possibly indicates sequestration effects that make the TFs non-functional.

Using it for disease classification. Diagnose breast cancer metastasis using expression profiles. 300 patients 1/3 of whom became metastatic. Area under ROC is 65%. Heterogeneity is a problem. Little overlap between different gene sets from different studies.
Look at sequentiality or connections in protein-protein interactions.

http://www.nature.com/msb/journal/v3/n1/full/msb4100180.html


Cancer might be the perturbation out of homeostasis.

http://www.nature.com/nbt/journal/v27/n2/full/nbt.1522.html

Web 09

Bioinformatics in an Undergraduate Programme


Kam Dahlquist, Murli Nair *****

Part of curriculum development movement - the need to use problem based learning and the need to have a progressive curriculum that follows what will happen in real life. How to build problems and hands on learning into biology curriculum including statistics. Based around the ideas from the http://bioquest.org frameworks.

Online reports about the problems of including quantitative and computational skills into the biology curriculum.
"Bioinformatics is Biology and we cannot just have 2 pages in the textbook" Murli Nair, IUSB.
http://genomebiology.com/2008/9/12/114 All Biologists are Bioinformaticians