Method and system for creating and using knowledge patterns

• US 6,768,982 B1
• Filed: 09/06/2000
• Issued: 07/27/2004
• Est. Priority Date: 09/06/2000
• Status: Active Grant

• Current Assignee: Cellomics Incorporated (Thermo Fisher Scientific Incorporated)
• Original Assignee: Cellomics Incorporated (Thermo Fisher Scientific Incorporated)
• Inventors: Shaw, John R., Collins, Mark A. D., Colasanti, Ricardo L.
• Primary Examiner(s): Davis, George B.
• Application Number:
  US09/656,400
• Time in Patent Office: 1,420 Days
• Field of Search: 706/45, 706/46
• US Class Current: 706/45
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G16C 20/30 Prediction of properties of chemical compounds, compositions or mixtures
• CPC Class Codes: G16C 20/70 Machine learning, data mining or chemometrics
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: G16H 70/40 relating to drugs, e.g. their side effects or intended usage

Abstract

A method and system for creating and using knowledge patterns. One or more patterns derived from one or more pharmaceutical data sources are acquired. A knowledge map is created using the one or more patterns. The knowledge map, such as a self-organizing knowledge map, includes a selected representation of a pattern space. A set of selected regions from the knowledge map are annotated. Others patterns are classified with annotated regions of the knowledge map, thereby adding additional knowledge to the knowledge map. The knowledge map is used for recognizing previously unseen or unknown patterns from large amounts of pharmaceutical data obtained by automated screening systems. The knowledge map includes knowledge for a new real or virtual drug compounds or drug therapies. The knowledge patterns may be used to select new known pharmaceutical compounds, or model new virtual pharmaceutical compounds. The method and system described herein may be used to locate and describe previously unrecognized patterns of biological or chemical activity concealed with in a selected range of drug discovery data. This may improve the identification, selection, validation and screening of new real or virtual pharmaceutical compounds. The method and system may also be used to provide new bioinformatic techniques for storing and manipulating pharmaceutical knowledge.

Claims

1. A method for creating and using knowledge maps, comprising:

acquiring a plurality of patterns derived from one or more pharmaceutical data sources;

creating a knowledge map using the plurality of acquired patterns, wherein the knowledge map includes a selected representation of a pattern space;

annotating a set of selected regions from the knowledge map; and

classifying other patterns with annotated regions of the knowledge map, thereby adding additional knowledge to the knowledge map.

2. A computer readable medium having stored therein instructions for causing a processor to execute the method of claim 1.

3. The method of claim 1 wherein the step of acquiring a plurality of patterns derived from one or more pharmaceutical data sources includes deriving a plurality of patterns with co-evolutionary techniques.

4. The method of claim 1 wherein the step of acquiring a plurality of patterns includes acquiring cluster patterns from biological or chemical data.

5. The method of claim 1 wherein the plurality of patterns include toxicity, absorption, distribution, metabolism or excretion patterns.

6. The method of claim 1 wherein the step of creating a knowledge map using the plurality of acquired patterns includes creating a self-organizing knowledge map.

7. The method of claim 1 wherein the step of annotating a set of selected regions from the knowledge map includes automatically annotating a set of selected regions of a knowledge map with text, chemical structures or biological profiles.

8. The method of claim 1 wherein the step of classifying other patterns with annotated regions of the knowledge map includes classifying other known patterns with the annotated regions of the knowledge map.

9. The method of claim 1 wherein the step of classifying other patterns with annotated regions of the knowledge map includes classifying other unknown patterns with annotated regions of the knowledge map.

10. The method of claim 1 wherein the step of step of classifying other patterns with annotated regions of the knowledge map includes classifying virtual assays or virtual features for unknown or virtual chemical compounds.

11. The method of claim 1 wherein the additional knowledge includes fused knowledge.

12. The method of claim 1 wherein the step of classifying other patterns with annotated regions of the knowledge map includes:

creating a plurality of groups of similar patterns; and

adding an unknown pattern to a group to which the unknown pattern it closest to in the pattern space.

13. The method of claim 1 further comprising:

recognizing an unknown pattern as a familiar pattern by comparing the unknown pattern with known annotated patterns on the knowledge map; and

annotating the unknown pattern as a familiar pattern.

14. The method of claim 1 wherein the knowledge map includes knowledge for new real or virtual drug compounds or drug therapies.

15. The method of claim 1 wherein the knowledge map includes knowledge from real drug compounds or for virtual drug compounds.

16. A method for creating and using knowledge maps to determine pharmaceutical compounds, comprising:

acquiring a plurality of pharmaceutical patterns from a set of pharmaceutical compound assay or feature screening data;

creating a plurality of pharmaceutical knowledge maps from the plurality of pharmaceutical patterns;

annotating the plurality of pharmaceutical knowledge maps with other information to define a plurality of pharmaceutical pattern groupings; and

recognizing a plurality of previously unseen or unknown pharmaceutical patterns from the plurality of pharmaceutical pattern groupings, thereby creating new pharmaceutical pattern grouping knowledge useful for selecting a pharmaceutical compound.

17. A computer readable medium having stored therein instructions for causing a processor to execute the method of claim 16.

18. The method of claim 16 wherein the step of creating a plurality of knowledge maps from the plurality of pharmaceutical patterns includes creating a plurality of self-organizing maps.

19. The method of claim 16 wherein the step of annotating the plurality of pharmaceutical knowledge maps with other information to define a plurality of pharmaceutical pattern groupings includes automatically annotating a plurality of pharmaceutical knowledge maps with toxicity, absorption, distribution, metabolism, excretion or image processing pattern information.

20. The method of claim 16 wherein the step of recognizing a plurality of previously unseen or unknown pharmaceutical patterns from the plurality of pharmaceutical pattern groupings useful for selecting a pharmaceutical compound includes adding an unknown pattern to a pharmaceutical pattern to which the unknown pattern it closest to in a pattern space used to define the pharmaceutical knowledge map.

21. The method of claim 16 wherein the plurality of pharmaceutical knowledge maps include pharmaceutical knowledge for new real or virtual drug compounds or drug therapies.

22. The method of claim 16 wherein the plurality of pharmaceutical knowledge maps include knowledge from real drug compounds or for virtual drug compounds.

23. A pattern creation and analysis system, comprising in combination:

a pattern acquirer for acquiring a plurality of patterns derived from one or more pharmaceutical data sources;

a pattern creator for creating a knowledge map using the plurality of acquired patterns, wherein the knowledge map includes a selected representation of a pattern space;

a pattern annotator for automatically annotating a set of selected regions from the knowledge map; and

a pattern classifier for classifying other patterns with annotated regions of the knowledge map, thereby adding additional knowledge to the knowledge map.

24. The system of claim 23 wherein pattern creator includes creating a knowledge map as a self-organizing map.

25. The system of claim 23 wherein the knowledge map includes knowledge for new or real virtual drug compounds or drug therapies.

26. The system of claim 23 wherein the knowledge map includes knowledge for real drug compounds or virtual drug compounds.