Discovering context-specific relationships from biological literature by using multi-level context terms

Sejoon Lee, Jaejoon Choi, Kyunghyun Park, Min Song, Doheon Lee

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Background: The Swanson's ABC model is powerful to infer hidden relationships buried in biological literature. However, the model is inadequate to infer relations with context information. In addition, the model generates a very large amount of candidates from biological text, and it is a semi-automatic, labor-intensive technique requiring human expert's manual input. To tackle these problems, we incorporate context terms to infer relations between AB interactions and BC interactions. Methods. We propose 3 steps to discover meaningful hidden relationships between drugs and diseases: 1) multi-level (gene, drug, disease, symptom) entity recognition, 2) interaction extraction (drug-gene, gene-disease) from literature, 3) context vector based similarity score calculation. Subsequently, we evaluate our hypothesis with the datasets of the "Alzheimer's disease" related 77,711 PubMed abstracts. As golden standards, PharmGKB and CTD databases are used. Evaluation is conducted in 2 ways: first, comparing precision of the proposed method and the previous method and second, analysing top 10 ranked results to examine whether highly ranked interactions are truly meaningful or not. Results: The results indicate that context-based relation inference achieved better precision than the previous ABC model approach. The literature analysis also shows that interactions inferred by the context-based approach are more meaningful than interactions by the previous ABC model. Conclusions: We propose a novel interaction inference technique that incorporates context term vectors into the ABC model to discover meaningful hidden relationships. By utilizing multi-level context terms, our model shows better performance than the previous ABC model.

Original languageEnglish
Article numberS1
JournalBMC Medical Informatics and Decision Making
Volume12
Issue numberSUPPL. 1
DOIs
Publication statusPublished - 2012 May 7

Fingerprint

Genes
Drug Interactions
PubMed
Pharmaceutical Preparations
Alzheimer Disease
Databases
Datasets

All Science Journal Classification (ASJC) codes

  • Health Policy
  • Health Informatics

Cite this

@article{cd1df22bb4d04904b3c3aff9bf3ce642,
title = "Discovering context-specific relationships from biological literature by using multi-level context terms",
abstract = "Background: The Swanson's ABC model is powerful to infer hidden relationships buried in biological literature. However, the model is inadequate to infer relations with context information. In addition, the model generates a very large amount of candidates from biological text, and it is a semi-automatic, labor-intensive technique requiring human expert's manual input. To tackle these problems, we incorporate context terms to infer relations between AB interactions and BC interactions. Methods. We propose 3 steps to discover meaningful hidden relationships between drugs and diseases: 1) multi-level (gene, drug, disease, symptom) entity recognition, 2) interaction extraction (drug-gene, gene-disease) from literature, 3) context vector based similarity score calculation. Subsequently, we evaluate our hypothesis with the datasets of the {"}Alzheimer's disease{"} related 77,711 PubMed abstracts. As golden standards, PharmGKB and CTD databases are used. Evaluation is conducted in 2 ways: first, comparing precision of the proposed method and the previous method and second, analysing top 10 ranked results to examine whether highly ranked interactions are truly meaningful or not. Results: The results indicate that context-based relation inference achieved better precision than the previous ABC model approach. The literature analysis also shows that interactions inferred by the context-based approach are more meaningful than interactions by the previous ABC model. Conclusions: We propose a novel interaction inference technique that incorporates context term vectors into the ABC model to discover meaningful hidden relationships. By utilizing multi-level context terms, our model shows better performance than the previous ABC model.",
author = "Sejoon Lee and Jaejoon Choi and Kyunghyun Park and Min Song and Doheon Lee",
year = "2012",
month = "5",
day = "7",
doi = "10.1186/1472-6947-12-S1-S1",
language = "English",
volume = "12",
journal = "BMC Medical Informatics and Decision Making",
issn = "1472-6947",
publisher = "BioMed Central",
number = "SUPPL. 1",

}

Discovering context-specific relationships from biological literature by using multi-level context terms. / Lee, Sejoon; Choi, Jaejoon; Park, Kyunghyun; Song, Min; Lee, Doheon.

In: BMC Medical Informatics and Decision Making, Vol. 12, No. SUPPL. 1, S1, 07.05.2012.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Discovering context-specific relationships from biological literature by using multi-level context terms

AU - Lee, Sejoon

AU - Choi, Jaejoon

AU - Park, Kyunghyun

AU - Song, Min

AU - Lee, Doheon

PY - 2012/5/7

Y1 - 2012/5/7

N2 - Background: The Swanson's ABC model is powerful to infer hidden relationships buried in biological literature. However, the model is inadequate to infer relations with context information. In addition, the model generates a very large amount of candidates from biological text, and it is a semi-automatic, labor-intensive technique requiring human expert's manual input. To tackle these problems, we incorporate context terms to infer relations between AB interactions and BC interactions. Methods. We propose 3 steps to discover meaningful hidden relationships between drugs and diseases: 1) multi-level (gene, drug, disease, symptom) entity recognition, 2) interaction extraction (drug-gene, gene-disease) from literature, 3) context vector based similarity score calculation. Subsequently, we evaluate our hypothesis with the datasets of the "Alzheimer's disease" related 77,711 PubMed abstracts. As golden standards, PharmGKB and CTD databases are used. Evaluation is conducted in 2 ways: first, comparing precision of the proposed method and the previous method and second, analysing top 10 ranked results to examine whether highly ranked interactions are truly meaningful or not. Results: The results indicate that context-based relation inference achieved better precision than the previous ABC model approach. The literature analysis also shows that interactions inferred by the context-based approach are more meaningful than interactions by the previous ABC model. Conclusions: We propose a novel interaction inference technique that incorporates context term vectors into the ABC model to discover meaningful hidden relationships. By utilizing multi-level context terms, our model shows better performance than the previous ABC model.

AB - Background: The Swanson's ABC model is powerful to infer hidden relationships buried in biological literature. However, the model is inadequate to infer relations with context information. In addition, the model generates a very large amount of candidates from biological text, and it is a semi-automatic, labor-intensive technique requiring human expert's manual input. To tackle these problems, we incorporate context terms to infer relations between AB interactions and BC interactions. Methods. We propose 3 steps to discover meaningful hidden relationships between drugs and diseases: 1) multi-level (gene, drug, disease, symptom) entity recognition, 2) interaction extraction (drug-gene, gene-disease) from literature, 3) context vector based similarity score calculation. Subsequently, we evaluate our hypothesis with the datasets of the "Alzheimer's disease" related 77,711 PubMed abstracts. As golden standards, PharmGKB and CTD databases are used. Evaluation is conducted in 2 ways: first, comparing precision of the proposed method and the previous method and second, analysing top 10 ranked results to examine whether highly ranked interactions are truly meaningful or not. Results: The results indicate that context-based relation inference achieved better precision than the previous ABC model approach. The literature analysis also shows that interactions inferred by the context-based approach are more meaningful than interactions by the previous ABC model. Conclusions: We propose a novel interaction inference technique that incorporates context term vectors into the ABC model to discover meaningful hidden relationships. By utilizing multi-level context terms, our model shows better performance than the previous ABC model.

UR - http://www.scopus.com/inward/record.url?scp=84860430319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860430319&partnerID=8YFLogxK

U2 - 10.1186/1472-6947-12-S1-S1

DO - 10.1186/1472-6947-12-S1-S1

M3 - Article

C2 - 22595086

AN - SCOPUS:84860430319

VL - 12

JO - BMC Medical Informatics and Decision Making

JF - BMC Medical Informatics and Decision Making

SN - 1472-6947

IS - SUPPL. 1

M1 - S1

ER -