Uncovering protein interaction in abstracts and text using a novel linear model and word proximity networks

Abi-Haidar, Alaa; Kaur, Jasleen; Maguitman, Ana; Radivojac, Predrag; Rechtsteiner, Andreas; Verspoor, Karin; Wang, Zhiping; Rocha, Luis M.

doi:10.1186/gb-2008-9-s2-s11

Uncovering protein interaction in abstracts and text using a novel linear model and word proximity networks

Journal Article · Mon Sep 01 00:00:00 EDT 2008 · GenomeBiology.com

DOI:https://doi.org/10.1186/gb-2008-9-s2-s11· OSTI ID:1626732

Abi-Haidar, Alaa ^[1]; Kaur, Jasleen ^[2]; Maguitman, Ana ^[3]; Radivojac, Predrag ^[2]; Rechtsteiner, Andreas ^[2]; Verspoor, Karin ^[4]; Wang, Zhiping ^[2]; Rocha, Luis M. ^[5]

Indiana University, Bloomington, IN (United States); Instituto Gulbenkian de Ciência, Oeiras (Portugal); DOE/OSTI
Indiana University, Bloomington, IN (United States)
Universidad Nacional del Sur, Buenos Aires (Argentina)
Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
Indiana University, Bloomington, IN (United States); Instituto Gulbenkian de Ciência, Oeiras (Portugal)

Second BioCreative Challenge: classification of abstracts relevant for protein-protein interaction (interaction article subtask [IAS]), discovery of protein pairs (interaction pair subtask [IPS]), and identification of text passages characterizing protein interaction (interaction sentences subtask [ISS]) in full-text documents. We approached the abstract classification task with a novel, lightweight linear model inspired by spam detection techniques, as well as an uncertainty-based integration scheme. We also used a support vector machine and singular value decomposition on the same features for comparison purposes. Our approach to the full-text subtasks (protein pair and passage identification) includes a feature expansion method based on word proximity networks. Our approach to the abstract classification task (IAS) was among the top submissions for this task in terms of measures of performance used in the challenge evaluation (accuracy, F-score, and area under the receiver operating characteristic curve). We also report on a web tool that we produced using our approach: the Protein Interaction Abstract Relevance Evaluator (PIARE). Our approach to the full-text tasks resulted in one of the highest recall rates as well as mean reciprocal rank of correct passages. Our approach to abstract classification shows that a simple linear model, using relatively few features, can generalize and uncover the conceptual nature of protein-protein interactions from the bibliome. Because the novel approach is based on a rather lightweight linear model, it can easily be ported and applied to similar problems. In full-text problems, the expansion of word features with word proximity networks is shown to be useful, although the need for some improvements is discussed.

Research Organization:: Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)

Sponsoring Organization:: USDOE Office of Science (SC), Biological and Environmental Research (BER)

Grant/Contract Number:: AC52-06NA25396

OSTI ID:: 1626732

Journal Information:: GenomeBiology.com, Journal Name: GenomeBiology.com Journal Issue: Suppl 2 Vol. 9; ISSN 1465-6906

Publisher:: BioMed CentralCopyright Statement

Country of Publication:: United States

Language:: English

References (12)

Support Vector Machines Joachims, Thorsten Learning to Classify Text Using Support Vector Machines https://doi.org/10.1007/978-1-4615-0907-3_3	book	January 2002
Literature mining for the biologist: from information retrieval to biological discovery Jensen, Lars Juhl; Saric, Jasmin; Bork, Peer Nature Reviews Genetics, Vol. 7, Issue 2 https://doi.org/10.1038/nrg1768	journal	February 2006
Stable Association of 70-kDa Heat Shock Protein Induces Latent Multisite Specificity of a Unisite-specific Endonuclease in Yeast Mitochondria Mizumura, Hikaru; Shibata, Takehiko; Morishima, Nobuhiro Journal of Biological Chemistry, Vol. 274, Issue 36 https://doi.org/10.1074/jbc.274.36.25682	journal	September 1999
Mining the Biomedical Literature in the Genomic Era: An Overview Shatkay, Hagit; Feldman, Ronen Journal of Computational Biology, Vol. 10, Issue 6 https://doi.org/10.1089/106652703322756104	journal	December 2003
ABNER: an open source tool for automatically tagging genes, proteins and other entity names in text Settles, B. Bioinformatics, Vol. 21, Issue 14 https://doi.org/10.1093/bioinformatics/bti475	journal	April 2005
MIPS: analysis and annotation of proteins from whole genomes Mewes, H. W. Nucleic Acids Research, Vol. 32, Issue 90001 https://doi.org/10.1093/nar/gkh092	journal	January 2004
The Universal Protein Resource (UniProt) Bairoch, Amos; Bougueleret, Lydie; Altairac, Severine Nucleic Acids Research, Vol. 35, Issue Database, p. D193-D197 https://doi.org/10.1093/nar/gkl929	journal	January 2007
MINT: the Molecular INTeraction database Chatr-aryamontri, A.; Ceol, A.; Palazzi, L. M. Nucleic Acids Research, Vol. 35, Issue Database https://doi.org/10.1093/nar/gkl950	journal	January 2007
IntAct--open source resource for molecular interaction data Kerrien, S.; Alam-Faruque, Y.; Aranda, B. Nucleic Acids Research, Vol. 35, Issue Database https://doi.org/10.1093/nar/gkl958	journal	January 2007
Large-Scale Testing of Bibliome Informatics Using pfam Protein Families Maguitman, Ana G.; Rechtsteiner, Andreas; Verspoor, Karin Biocomputing 2006 https://doi.org/10.1142/9789812701626_0008	conference	December 2005
Overview of BioCreAtIvE: critical assessment of information extraction for biology Hirschman, Lynette; Yeh, Alexander; Blaschke, Christian BMC Bioinformatics, Vol. 6, Issue Suppl 1 https://doi.org/10.1186/1471-2105-6-s1-s1	journal	January 2005
Protein annotation as term categorization in the gene ontology using word proximity networks Verspoor, Karin; Cohn, Judith; Joslyn, Cliff BMC Bioinformatics, Vol. 6, Issue Suppl 1 https://doi.org/10.1186/1471-2105-6-s1-s20	journal	January 2005

Cited By (8)

Construction of antimicrobial peptide-drug combination networks from scientific literature based on a semi-automated curation workflow Jorge, Paula; Pérez-Pérez, Martín; Pérez Rodríguez, Gael Database, Vol. 2016 https://doi.org/10.1093/database/baw143	journal	January 2016
The Protein-Protein Interaction tasks of BioCreative III: classification/ranking of articles and linking bio-ontology concepts to full text Krallinger, Martin; Vazquez, Miguel; Leitner, Florian BMC Bioinformatics, Vol. 12, Issue S8 https://doi.org/10.1186/1471-2105-12-s8-s3	journal	October 2011
Extraction of Pharmacokinetic Evidence of Drug–Drug Interactions from the Literature Kolchinsky, Artemy; Lourenço, Anália; Wu, Heng-Yi PLOS ONE, Vol. 10, Issue 5 https://doi.org/10.1371/journal.pone.0122199	journal	May 2015
Prioritization of Therapeutic Targets of Inflammation Using Proteomics, Bioinformatics, and In Silico Cell-Cell Interactomics Haqqani, Arsalan S.; Stanimirovic, Danica B. Methods in Molecular Biology https://doi.org/10.1007/978-1-4939-9597-4_20	book	January 2019
Prioritization of Therapeutic Targets of Inflammation Using Proteomics, Bioinformatics, and In Silico Cell–Cell Interactomics Haqqani, Arsalan S.; Stanimirovic, Danica B. Methods in Molecular Biology https://doi.org/10.1007/978-1-62703-589-7_21	book	January 2013
Distance closures on complex networks Simas, Tiago; Rocha, Luis M. Network Science, Vol. 3, Issue 2 https://doi.org/10.1017/nws.2015.11	journal	March 2015
Quorum sensing inhibition in Pseudomonas aeruginosa biofilms: new insights through network mining Pérez-Pérez, Martín; Jorge, Paula; Pérez Rodríguez, Gael Biofouling, Vol. 33, Issue 2 https://doi.org/10.1080/08927014.2016.1272104	journal	January 2017
The Protein-Protein Interaction tasks of BioCreative III: classification/ranking of articles and linking bio-ontology concepts to full text Krallinger, M.; Vazquez, M.; Leitner, F. BioMed Central https://doi.org/10.5167/uzh-58309	text	January 2011

Similar Records

Protein annotation as term categorization in the gene ontology using word proximity networks

Journal Article · Tue May 24 00:00:00 EDT 2005 · BMC Bioinformatics · OSTI ID:1626313

An Introduction to Word Embeddings and Language Models

Technical Report · Thu Apr 01 00:00:00 EDT 2021 · OSTI ID:1773690

PNNL: A Supervised Maximum Entropy Approach to Word Sense Disambiguation

Conference · Sat Jun 23 00:00:00 EDT 2007 · OSTI ID:924370

Related Subjects

59 BASIC BIOLOGICAL SCIENCES
proximity network
singular value decomposition
singular value decomposition method
support vector machine
word pair

Uncovering protein interaction in abstracts and text using a novel linear model and word proximity networks

Citation Formats

References (12)

Cited By (8)

Similar Records

Related Subjects