Molecular level similarity search and computer aided drug discovery process
View Patent ↗Systems, methods and articles of manufacture are disclosed for searching chemical compounds referenced in chemical literature. References to chemical compounds in the documents may be transformed to corresponding references using a standard notation for representing chemical compounds. Criteria specifying desired molecular characteristics may be received. A regular expression may be generated based on the received criteria. The chemical documents may be searched using the regular expression. Based on the search, a chemical document may be determined that references a chemical compound satisfying the received criteria. Further, the regular expression may be generated based on the received criteria and a received mutation rule. Based on the generated regular expression and the chemical documents, one or more chemical compounds may be discovered that satisfy both the received criteria and the received mutation rule, but that is are not referenced in the chemical documents.
1. A computer-implemented method, comprising:
configuring one or more processors to perform an operation for discovering chemical compounds not referenced in a plurality of documents, comprising:
receiving the plurality of documents describing chemical compounds;
receiving criteria specifying desired characteristics;
receiving at least one mutation rule specifying desired mutation types;
generating at least one regular expression based on the received criteria and the at least one mutation rule;
evaluating, by operation of the one or more computer processors, the generated regular expression against the chemical compound to determine that the chemical compound satisfies the received criteria; and
in response to determining, based on the evaluation, that the chemical compound is not referenced in the plurality of documents, identifying the chemical as a discovered chemical.
2. The computer-implemented method of claim 1 , wherein the operation further comprises:
parsing each of the plurality of documents to identify references to chemical compounds; and
transforming each of the identified references into a corresponding reference using the standard notation.
3. The computer-implemented method of claim 1 , wherein the standard notation is the simplified molecular input line entry specification (SMILES) notation and wherein each of the plurality of documents is selected from at least a chemical patent, a chemical research paper, and a chemical lab experiment document.
4. The computer-implemented method of claim 1 , wherein the operation further comprises:
computing a score representing a commercial value of the discovered chemical compound.
5. The computer-implemented method of claim 4 , wherein computing the score includes determining a binding affinity of the discovered chemical compound to a desired receptor.
6. The computer-implemented method of claim 4 , wherein the operation further comprises:
upon determining that the computed score exceeds a predefined threshold score, simulating a molecular docking between the discovered chemical compound and a specified receptor.
7. A computer program product to perform an operation for discovering chemical compounds not referenced in a plurality of documents, comprising:
a non-transitory computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code comprising:
computer-readable program code configured to receive the plurality of documents describing chemical compounds;
computer-readable program code configured to receive criteria specifying desired characteristics;
computer-readable program code configured to receive at least one mutation rule specifying desired mutation types;
computer-readable program code configured to generate at least one regular expression based on the received criteria and the at least one mutation rule;
computer-readable program code configured to evaluate the generated regular expression against the chemical compound to determine that the chemical compound satisfies the received criteria; and
computer-readable program code configured to, in response to determining, based on the evaluation, that the chemical compound is not referenced in the plurality of documents, identify the chemical as a discovered chemical.
8. The computer program product of claim 7 , the computer-readable program code further comprising:
computer-readable program code configured to parse each of the plurality of documents to identify references to chemical compounds; and
computer-readable program code configured to transform each of the identified references into a corresponding reference using the standard notation.
9. The computer program product of claim 7 , wherein the standard notation is the simplified molecular input line entry specification (SMILES) notation and wherein each of the plurality of documents is selected from at least a chemical patent, a chemical research paper, and a chemical lab experiment document.
10. The computer program product of claim 7 , the computer-readable program code further comprising:
computer-readable program code configured to compute a score representing a commercial value of the discovered chemical compound.
11. The computer program product of claim 10 , wherein computing the score includes determining a binding affinity of the discovered chemical compound to a desired receptor.
12. The computer program product of claim 10 , the computer-readable program code further comprising:
computer-readable program code configured to, upon determining that the computed score exceeds a predefined threshold score, simulate a molecular docking between the discovered chemical compound and a specified receptor.
13. A system, comprising:
one or more computer processors; and
a memory containing a program, which when executed by the one or more computer processors, performs an operation for discovering chemical compounds not referenced in a plurality of documents, the operation comprising:
receiving the plurality of documents describing chemical compounds;
receiving criteria specifying desired characteristics;
receiving at least one mutation rule specifying desired mutation types;
generating at least one regular expression based on the received criteria and the at least one mutation rule;
evaluating the generated regular expression against the chemical compound to determine that the chemical compound satisfies the received criteria; and
in response to determining, based on the evaluation, that the chemical compound is not referenced in the plurality of documents, identifying the chemical as a discovered chemical.
14. The system of claim 13 , wherein the operation further comprises:
parsing each of the plurality of documents to identify references to chemical compounds; and
transforming each of the identified references into a corresponding reference using the standard notation.
15. The system of claim 13 , wherein the standard notation is the simplified molecular input line entry specification (SMILES) notation and wherein each of the plurality of documents is selected from at least a chemical patent, a chemical research paper, and a chemical lab experiment document.
16. The system of claim 13 , wherein the operation further comprises:
computing a score representing a commercial value of the discovered chemical compound.
17. The system of claim 16 , wherein computing the score includes determining a binding affinity of the discovered chemical compound to a desired receptor.
18. The system of claim 16 , wherein the operation further comprises:
upon determining that the computed score exceeds a predefined threshold score, simulating a molecular docking between the discovered chemical compound and a specified receptor.