System, method and apparatus for causal implication analysis in biological networks

• US 8,594,941 B2
• Filed: 11/19/2004
• Issued: 11/26/2013
• Est. Priority Date: 11/26/2003
• Status: Expired due to Fees

• Current Assignee: Selventa, Inc.
• Original Assignee: Selventa, Inc.
• Inventors: Chandra, Dundee Navin, Segaran, Suresh Toby, Kightley, David, Sun, Justin, Pratt, Dexter
• Primary Examiner(s): SIMS, JASON M
• Application Number:
  US10/992,973
• Publication Number:
  US 20050165594A1
• Time in Patent Office: 3,294 Days
• Field of Search: None
• US Class Current: 702/19
• CPC Class Codes: G16B 40/00 ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
• 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: G16B 5/20 Probabilistic models

Abstract

Described are methods, systems and apparatus for hypothesizing a biological relationship in a biological system. A database of biological assertions is provided consisting of biological elements, relationships among the biological elements, and relationship descriptors characterizing the properties of the elements and relationships. A biological element may be selected from the database and a logical simulation may be performed within the biological database, from the selected biological element, through relationship descriptors, along a path defined by potentially causative biological elements to discern a biological element hypothetically responsible for the change in the selected biological element. The logical simulation may be either a backward logical simulation, performed upstream through the relationship descriptors to discern a hypothetical responsible biological element, or a forward logical simulation, performed downstream through the relationship descriptors to discern the extent to which the perturbation generates the observed change in the selected biological element.

Claims

1. Apparatus for hypothesizing a biological relationship, comprising:

a processor;

an electronic data store holding a database comprising a multiplicity of nodes representative of biological elements, and relationship descriptors describing causal relationships between nodes, the causal relationships defining how an increase or decrease in activity or abundance of one node has been shown, or is suspected, to cause an increase or decrease in the activity or abundance of another node, the database being non-specific in that at least some relationship descriptors define casual relationships that are not known or suspected to be relevant to any particular biological context of interest; and

computer memory holding computer program instructions which, when executed by the processor, hypothesize the biological relationship from the collection of non-specific relationship descriptors by executing one of;

a reverse logical simulation, and a forward logical simulation, the computer program instructions executable;

(a) for a reverse logical simulation;

(i) to select a set of target nodes;

(ii) to associate to the set of target nodes a set of perturbations, wherein each target node of the set of target nodes has an associated perturbation, and wherein the set of perturbations represent one of;

experimentally-observed perturbations and theoretical perturbations;

(iii) with respect to at least one particular source node that is hypothetically responsible for the set of perturbations, to traverse causal relationships in the database in a reverse direction, from an effect to a cause of each causal relationship, to identify a set of one or more paths from each of the target nodes back to the particular source node, the set of paths comprised of casual relationships of which at least one such causal relationship is not known or suspected to be relevant to the particular biological context of interest;

(iv) based on results obtained from the traversing step, to define, as a hypothesized biological relationship, the particular source node, together with the one or more casual relationships that comprise the set of paths that link the particular source node to the set of target nodes; and

(v) to compute one or more scores associated with the hypothesized biological relationship based on the set of perturbations and the set of paths;

(b) for a forward logical simulation;

(i) to select a set of source nodes;

(ii) to associate to the set of source nodes a set of perturbations, wherein each source node of the set of source nodes has an associated perturbation, and wherein the set of perturbations represent one of;

experimentally-observed perturbations and theoretical perturbations;

(iii) with respect to at least one particular target node, to traverse causal relationships in the database in a forward direction, from a cause to an effect of each causal relationship, to identify a set of one or more paths from each of the source nodes forward to the particular target node, the set of paths comprised of casual relationships of which at least one such causal relationship is not known or suspected to be relevant to the particular biological context of interest;

(iv) based on results obtained from the traversing step, to define, as the hypothesized biological relationship, the particular target node, together with the one or more casual relationships that comprise the set of paths that link the particular target node to the set of source nodes; and

(v) to compute one or more scores associated with the hypothesized biological relationship based on the set of perturbations and the set of paths.

2. The apparatus as described in claim 1 wherein in the reverse logical simulation the set of paths that link the particular source node to the set of target nodes includes zero or more additional nodes that are other than the particular source node and the set of target nodes and that lie along a path of the set of paths.

3. The apparatus as described in claim 1 wherein in the forward logical simulation the set of paths that link the particular target node to the set of source nodes includes zero or more additional nodes that are other than the particular target node and the set of source nodes and that lie along a path of the set of paths.

4. The apparatus as described in claim 1 wherein the reverse or forward logical simulation is initiated from a subset of the database.

13. The apparatus as described in claim 1 further including program code executed by the processor to apply a probability algorithm to the hypothesized biological relationship to determine if the hypothesized biological relationship has a given probability of representing real biology.

14. The apparatus as described in claim 1 wherein at least one of the set of perturbations is selected to simulate a perturbation suspected to be present in a mammalian biological system when in a state of disease, toxicity, environmental exposure, abnormality, morbidity, aging, or response to other stimulus.

15. The apparatus as described in claim 1, wherein said relationship descriptors comprise descriptors of the condition, location, source, amount, or substructure of a molecule, biological structure, physiological condition, trait, phenotype, biological process, clinical data, medical data, or disease data and chemistry.

16. The apparatus as described in claim 1, wherein one or more relationship descriptors correspond to an epistemological relationship between a pair of nodes.

17. The apparatus as described in claim 1, wherein the database comprises at least 5,000 nodes.

18. The apparatus as described in claim 1, wherein the database comprises at least 10,000 nodes.

19. The apparatus as described in claim 1, wherein the database comprises at least 50,000 nodes.

20. The apparatus as described in claim 1, wherein the comprises at least 100,000 nodes.

21. The apparatus as described in claim 1 wherein the scores associated with the hypothesized biological relationship are based at least in part on the extent to which the nodes within the set of paths affect a specified perturbation in the set of perturbations.

22. The apparatus as described in claim 1 wherein the database further comprises representations of reactions, covalent modification, multi-molecular interactions, transport, catalysis, and inhibition.

23. The apparatus as described in claim 1 further including a display for displaying information associated with the hypothesized biological relationship.

5. An article comprising a tangible machine readable medium that stores a program, the program being executable by a machine to perform a method for hypothesizing a biological relationship from biological information in a database, the database comprising a multiplicity of nodes representative of biological elements, and relationship descriptors describing causal relationships between nodes, the causal relationships defining how an increase or decrease in activity or abundance of one node has been shown, or is suspected, to cause an increase or decrease in the activity or abundance of another node, the database being non-specific in that at least some relationship descriptors define casual relationships that are not known or suspected to be relevant to any particular biological context of interest, the method comprising:

hypothesizing the biological relationship from the collection of non-specific relationship descriptors by executing one of;

a reverse logical simulation, and a forward logical simulation, wherein;

(a) for a reverse logical simulation, the method is operative;

(i) to select a set of target nodes;

(ii) to associate to the set of target nodes a set of perturbations, wherein each target node of the set of target nodes has an associated perturbation, and wherein the set of perturbations represent one of;

experimentally-observed perturbations and theoretical perturbations;

(iii) with respect to at least one particular source node that is hypothetically responsible for the set of perturbations, to traverse causal relationships in the database in a reverse direction, from an effect to a cause of each causal relationship, to identify a set of one or more paths from each of the target nodes back to the particular source node, the set of paths comprised of casual relationships of which at least one such causal relationship is not known or suspected to be relevant to the particular biological context of interest;

(iv) based on results obtained from the traversing step, to define, as a hypothesized biological relationship, the particular source node, together with the one or more casual relationships that comprise the set of paths that link the particular source node to the set of target nodes; and

(v) to compute one or more scores associated with the hypothesized biological relationship based on the set of perturbations and the set of paths;

(b) for a forward logical simulation, the method is operative;

(i) to select a set of source nodes;

(ii) to associate to the set of source nodes a set of perturbations, wherein each source node of the set of source nodes has an associated perturbation, and wherein the set of perturbations represent one of;

experimentally-observed perturbations and theoretical perturbations;

(iii) with respect to at least one particular target node, to traverse causal relationships in the database in a forward direction, from a cause to an effect of each causal relationship, to identify a set of one or more paths from each of the source nodes forward to the particular target node, the set of paths comprised of casual relationships of which at least one such causal relationship is not known or suspected to be relevant to the particular biological context of interest;

(iv) based on results obtained from the traversing step, to define, as the hypothesized biological relationship, the particular target node, together with the one or more casual relationships that comprise the set of paths that link the particular target node to the set of source nodes; and

(v) to compute one or more scores associated with the hypothesized biological relationship based on the set of perturbations and the set of paths.

6. The article as described in claim 5 wherein in the reverse logical simulation the set of paths that link the particular source node to the set of target nodes includes zero or more additional nodes that are other than the particular source node and the set of target nodes and that lie along a path of the set of paths.

7. The article as described in claim 5 wherein in the forward logical simulation the set of paths that link the particular target node to the set of source nodes includes zero or more additional nodes that are other than the particular target node and the set of source nodes and that lie along a path of the set of paths.

8. The article as described in claim 5 wherein the reverse or forward logical simulation is initiated from a subset of the database.

9. A method for hypothesizing a biological relationship, comprising:

receiving information from an electronic data store holding a database, the database comprising a multiplicity of nodes representative of biological elements, and relationship descriptors describing causal relationships between nodes, the causal relationships defining how an increase or decrease in activity or abundance of one node has been shown, or is suspected, to cause an increase or decrease in the activity or abundance of another node, the database being non-specific in that at least some relationship descriptors define casual relationships that are not known or suspected to be relevant to any particular biological context of interest; and

executing computer program instructions by a machine having a hardware element to hypothesize the biological relationship from the collection of non-specific relationship descriptors by executing one of;

a reverse logical simulation, and a forward logical simulation, the computer program instructions executable;

(a) for a reverse logical simulation;

(i) to select a set of target nodes;

(ii) to associate to the set of target nodes a set of perturbations, wherein each target node of the set of target nodes has an associated perturbation, and wherein the set of perturbations represent one of;

experimentally-observed perturbations and theoretical perturbations;

(iii) with respect to at least one particular source node that is hypothetically responsible for the set of perturbations, to traverse causal relationships in the database in a reverse direction, from an effect to a cause of each causal relationship, to identify a set of one or more paths from each of the target nodes back to the particular source node, the set of paths comprised of casual relationships of which at least one such causal relationship is not known or suspected to be relevant to the particular biological context of interest;

(iv) based on results obtained from the traversing step, to define, as a hypothesized biological relationship, the particular source node, together with the one or more casual relationships that comprise the set of paths that link the particular source node to the set of target nodes; and

(v) to compute one or more scores associated with the hypothesized biological relationship based on the set of perturbations and the set of paths;

(b) for a forward logical simulation;

(i) to select a set of source nodes;

(ii) to associate to the set of source nodes a set of perturbations, wherein each source node of the set of source nodes has an associated perturbation, and wherein the set of perturbations represent one of;

experimentally-observed perturbations and theoretical perturbations;

(iii) with respect to at least one particular target node, to traverse causal relationships in the database in a forward direction, from a cause to an effect of each causal relationship, to identify a set of one or more paths from each of the source nodes forward to the particular target node, the set of paths comprised of casual relationships of which at least one such causal relationship is not known or suspected to be relevant to the particular biological context of interest;

(iv) based on results obtained from the traversing step, to define, as the hypothesized biological relationship, the particular target node, together with the one or more casual relationships that comprise the set of paths that link the particular target node to the set of source nodes; and

(v) to compute one or more scores associated with the hypothesized biological relationship based on the set of perturbations and the set of paths.

10. The method as described in claim 9 wherein in the reverse logical simulation the set of paths that link the particular source node to the set of target nodes includes zero or more additional nodes that are other than the particular source node and the set of target nodes and that lie along a path of the set of paths.

11. The method as described in claim 10 wherein in the forward logical simulation the set of paths that link the particular target node to the set of source nodes includes zero or more additional nodes that are other than the particular target node and the set of source nodes and that lie along a path of the set of paths.

12. The method as described in claim 9 wherein the reverse or forward logical simulation is initiated from a subset of the database.