Limits...
Exploiting Literature-derived Knowledge and Semantics to Identify Potential Prostate Cancer Drugs.

Zhang R, Cairelli MJ, Fiszman M, Kilicoglu H, Rindflesch TC, Pakhomov SV, Melton GB - Cancer Inform (2014)

Bottom Line: We used the semantic relationships in SemMedDB, a database of structured knowledge extracted from all MEDLINE citations using SemRep, to extract potential relationships using knowledge of cancer drugs pathways.Two cancer drugs pathway schemas were constructed using these relationships extracted from SemMedDB.Through both pathway schemas, we found drugs already used for prostate cancer therapy and drugs not currently listed as the prostate cancer medications.

View Article: PubMed Central - PubMed

Affiliation: Institute for Health Informatics, University of Minnesota, Minneapolis, MN, USA. ; Department of Surgery, University of Minnesota, Minneapolis, MN, USA.

ABSTRACT
In this study, we report on the performance of an automated approach to discovery of potential prostate cancer drugs from the biomedical literature. We used the semantic relationships in SemMedDB, a database of structured knowledge extracted from all MEDLINE citations using SemRep, to extract potential relationships using knowledge of cancer drugs pathways. Two cancer drugs pathway schemas were constructed using these relationships extracted from SemMedDB. Through both pathway schemas, we found drugs already used for prostate cancer therapy and drugs not currently listed as the prostate cancer medications. Our study demonstrates that the appropriate linking of relevant structured semantic relationships stored in SemMedDB can support the discovery of potential prostate cancer drugs.

No MeSH data available.


Related in: MedlinePlus

Prostate cancer concepts are found from the UMLS using MetaMap. SemRep extracts semantic predications from the MEDLINE database and stores them in SemMedDB. Predications from SemMedDB are found containing the prostate cancer concepts as objects and genes as subjects and more predications are found that contain drugs as subjects and genes as objects. Additional predications are selected that contain genes as both subject and object. These predications are lined up in either the Drug→Gene→Cancer pathway schema or the Drug→ Gene1→ Gene2→Cancer pathway schema to produce a list of potential drugs and their mechanism of action in treating prostate cancer. A physician selects the best candidates based on the source citations and other relevant knowledge.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4216049&req=5

f1-cin-suppl.1-2014-103: Prostate cancer concepts are found from the UMLS using MetaMap. SemRep extracts semantic predications from the MEDLINE database and stores them in SemMedDB. Predications from SemMedDB are found containing the prostate cancer concepts as objects and genes as subjects and more predications are found that contain drugs as subjects and genes as objects. Additional predications are selected that contain genes as both subject and object. These predications are lined up in either the Drug→Gene→Cancer pathway schema or the Drug→ Gene1→ Gene2→Cancer pathway schema to produce a list of potential drugs and their mechanism of action in treating prostate cancer. A physician selects the best candidates based on the source citations and other relevant knowledge.

Mentions: Our approach (Fig. 1) included four basic components: (1) identifying possible UMLS concepts (with MetaMap) related to prostate cancer, (2) extracting all semantic predications relevant to prostate cancer concepts as well as the genes and drugs that are in a relationship with those concepts from SemMedDB, (3) discovering all possible cancer drugs based on combinations of semantic predications according to pathway schemas, and (4) providing potential unknown prostate cancer drugs after human review and exclusion of known drugs. These components are achieved through a series of steps detailed below.


Exploiting Literature-derived Knowledge and Semantics to Identify Potential Prostate Cancer Drugs.

Zhang R, Cairelli MJ, Fiszman M, Kilicoglu H, Rindflesch TC, Pakhomov SV, Melton GB - Cancer Inform (2014)

Prostate cancer concepts are found from the UMLS using MetaMap. SemRep extracts semantic predications from the MEDLINE database and stores them in SemMedDB. Predications from SemMedDB are found containing the prostate cancer concepts as objects and genes as subjects and more predications are found that contain drugs as subjects and genes as objects. Additional predications are selected that contain genes as both subject and object. These predications are lined up in either the Drug→Gene→Cancer pathway schema or the Drug→ Gene1→ Gene2→Cancer pathway schema to produce a list of potential drugs and their mechanism of action in treating prostate cancer. A physician selects the best candidates based on the source citations and other relevant knowledge.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4216049&req=5

f1-cin-suppl.1-2014-103: Prostate cancer concepts are found from the UMLS using MetaMap. SemRep extracts semantic predications from the MEDLINE database and stores them in SemMedDB. Predications from SemMedDB are found containing the prostate cancer concepts as objects and genes as subjects and more predications are found that contain drugs as subjects and genes as objects. Additional predications are selected that contain genes as both subject and object. These predications are lined up in either the Drug→Gene→Cancer pathway schema or the Drug→ Gene1→ Gene2→Cancer pathway schema to produce a list of potential drugs and their mechanism of action in treating prostate cancer. A physician selects the best candidates based on the source citations and other relevant knowledge.
Mentions: Our approach (Fig. 1) included four basic components: (1) identifying possible UMLS concepts (with MetaMap) related to prostate cancer, (2) extracting all semantic predications relevant to prostate cancer concepts as well as the genes and drugs that are in a relationship with those concepts from SemMedDB, (3) discovering all possible cancer drugs based on combinations of semantic predications according to pathway schemas, and (4) providing potential unknown prostate cancer drugs after human review and exclusion of known drugs. These components are achieved through a series of steps detailed below.

Bottom Line: We used the semantic relationships in SemMedDB, a database of structured knowledge extracted from all MEDLINE citations using SemRep, to extract potential relationships using knowledge of cancer drugs pathways.Two cancer drugs pathway schemas were constructed using these relationships extracted from SemMedDB.Through both pathway schemas, we found drugs already used for prostate cancer therapy and drugs not currently listed as the prostate cancer medications.

View Article: PubMed Central - PubMed

Affiliation: Institute for Health Informatics, University of Minnesota, Minneapolis, MN, USA. ; Department of Surgery, University of Minnesota, Minneapolis, MN, USA.

ABSTRACT
In this study, we report on the performance of an automated approach to discovery of potential prostate cancer drugs from the biomedical literature. We used the semantic relationships in SemMedDB, a database of structured knowledge extracted from all MEDLINE citations using SemRep, to extract potential relationships using knowledge of cancer drugs pathways. Two cancer drugs pathway schemas were constructed using these relationships extracted from SemMedDB. Through both pathway schemas, we found drugs already used for prostate cancer therapy and drugs not currently listed as the prostate cancer medications. Our study demonstrates that the appropriate linking of relevant structured semantic relationships stored in SemMedDB can support the discovery of potential prostate cancer drugs.

No MeSH data available.


Related in: MedlinePlus