GRAPHICAL RULE BASED MODELING OF BIOCHEMICAL NETWORKS

• US 20070212719A1
• Filed: 03/12/2007
• Published: 09/13/2007
• Est. Priority Date: 03/10/2006
• Status: Abandoned Application

• Current Assignee: Los Alamos National Security LLC (Government of the United States of America)
• Original Assignee: Los Alamos National Security LLC (Government of the United States of America)
• Inventors: Faeder, James, Hlavacek, William, Blinov, Michael
• Application Number:
  US11/685,114
• Publication Number:
  US 20070212719A1
• Time in Patent Office: Days
• US Class Current: 435/6
• CPC Class Codes: G16B 20/00 ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
• CPC Class Codes: G16B 20/30 Detection of binding sites or motifs
• CPC Class Codes: G16B 5/00 ICT specially adapted for modelling or simulations in systems biology, e.g. gene-regulatory networks, protein interaction networks or metabolic networks
• CPC Class Codes: G16C 20/10 Analysis or design of chemical reactions, syntheses or processes

Abstract

Formalized graphical reaction rules and conventions accounting for chemical states and binding or reaction states are provided for modeling complex biological systems such as signal transduction pathways. A system model is derived by defining typed attributed graphs which delimit molecular entities and their possible states. Graph transformation rules defining a class of potential reactions are defined and applied to the graphs and all new graphs that subsequently arise as a result of graph transformation. In one embodiment, a model is generated through the use of graph-rewriting rules which are associated with rate laws and applied iteratively to a seed set of chemical species graphs until a termination condition occurs.

Claims

1. A method for generating a biochemical reaction network, comprising:

identifying the components and states of molecular entities within a system;

generating molecular entity graphs representing the components and states of said molecular entities;

formulating graph transformation rules representing the potential reactions and products of said molecular entities; and

applying said graph transformation rules iteratively to said molecular entity graphs until no new graphs are generated or a termination condition is satisfied to provide a network of biochemical reactions.

2. A method as recited in claim 1, wherein said molecular entity comprises a plurality of molecular entities.

3. A method as recited in claim 1, wherein said molecular entity graphs comprise nodes representing components of said molecular entities and edges representing associations between one of said components with another component.

4. A method as recited in claim 3, wherein said molecular entity graphs further comprise edge labels.

5. A method as recited in claim 3, wherein said nodes representing said components of said molecular entity graphs comprise variable and fixed attributes.

6. A method as recited in claim 5, wherein said fixed attributes of said nodes of said molecular entity graphs are attributes selected from the group of attributes consisting essentially of molecular weight, sequence and annotation sources.

7. A method as recited in claim 6, wherein said variable attribute of said nodes comprises phosphorylation status.

8. A method as recited in claim 1, wherein said graph transformation rules comprise graph rewriting rules.

9. A method for generating a model of complex biochemical networks, comprising:

specifying an initial set of molecular entities within a system;

identifying possible physical and chemical states of molecules, components and complexes of said molecular entities to provide a set of chemical species;

representing graphically said molecules, components, complexes and states of said chemical species;

specifying graph transformation rules for classes of products and reactions of said chemical species;

applying said graph transformation rules on an initial set of graphs of said chemical species to produce a first generation set of graphs; and

applying iteratively said graph transformation rules on said first generation set and upon on all subsequent generations of sets of graphs produced by the subsequent application of said graph transformation rules on the subsequent generations of sets of graphs to produce a model.

10. A method as recited in claim 9, further comprising an output function associated with a group of chemical species.

11. A method as recited in claim 9, wherein said graph transformation rules comprise graph rewriting rules.

12. A method as recited in claim 11, wherein said graph rewriting rules comprise:

the addition and removal of intra-molecular edges;

the addition and removal of inter-molecular edges, the change of values of variable attributes; and

the replacement of a molecular entity or set of molecular entities with another molecular entity or set of molecular entities having the same components.

13. A method as recited in claim 9, wherein said graph transformation rules comprise reaction rules of group graphs representing reactants, group graphs representing products and a rate law.

14. A method as recited in claim 13, further comprising pattern graphs where components, molecular entities and molecular complexes are associated with variable attributes of said molecular entities.

15. A method as recited in claim 14, wherein said graph transformation rules further comprise function evaluation rules.

16. A method as recited in claim 15, wherein said function evaluation rule comprises:

matching said chemical-species graphs with said pattern graphs; and

calculating a value of the output function of said function evaluation rule.

17. A method as recited in claim 9, wherein said initial set of graphs comprises chemical species graphs matched with identical pattern graphs.

18. A method as recited in claim 9, wherein the application of said graph transformation rules comprises:

applying a reaction rule to said initial set of chemical species to identify a reactant group of chemical species corresponding to each reactant group graph;

replacing subgraphs of each reactant species matching the reactant group graphs with the corresponding product group graphs to define a group of product species; and

checking a current list of chemical species and adding members of said group of product species to the list that are not already present.

19. A method as recited in claim 9, further comprising:

associating a mathematical function with a group of chemical species.

20. A method as recited in claim 9, further comprising:

archiving graphs and graph transformation rules in a database in computer readable form.

21. A computer system for modeling biochemical networks, comprising:

a computation device configured for receiving data input;

a software program to operate said computation device, said program performing the operations of;

graphically representing molecules, components and chemical states of an initial set of molecular entities within a system;

applying specified graph transformation rules for classes of products and reactants of molecular entities within said system upon a seed set of graphs to produce a first generation set of graphs;

iteratively applying graph transformation rules to subsequent generations of graphs until a termination condition occurs; and

forming a model of molecular entities within said system; and

a user interface allowing the user to view computed model results and provide graphical and text input to said computation device.

22. A computer system for modeling biochemical networks as recited in claim 21, wherein said data input to said computation device comprises graphs and graph transformation rules obtained from a database of graphs and graph transformation rules.

23. A computer system for modeling biochemical networks as recited in claim 21, wherein said software is configured to simulate network activity.