Finding Complex Biological Relationships in Recent PubMed Articles Using Bio-LDA
PLoS ONEPublished 23 March 2011Open access
Huijun Wang, Ying Ding, Jie Tang, Xiao Dong, Bing He, Judy Qiu
Citations96
SJR quartileQ1
SJR score0.80
SNIP1.07
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
An algorithm called Bio-LDA is described that uses extracted biological terminology to automatically identify latent topics, and provides a variety of measures to uncover putative relations among topics and bio-terms, and offers new opportunities for knowledge discovery in many areas of biology.
Abstract
The overwhelming amount of available scholarly literature in the life sciences poses significant challenges to scientists wishing to keep up with important developments related to their research, but also provides a useful resource for the discovery of recent information concerning genes, diseases, compounds and the interactions between them. In this paper, we describe an algorithm called Bio-LDA that uses extracted biological terminology to automatically identify latent topics, and provides a variety of measures to uncover putative relations among topics and bio-terms. Relationships identified using those approaches are combined with existing data in life science datasets to provide additional insight. Three case studies demonstrate the utility of the Bio-LDA model, including association predication, association search and connectivity map generation. This combined approach offers new opportunities for knowledge discovery in many areas of biology including target identification, lead hopping and drug repurposing.
Keywords
Computer ScienceBiochemistry, Genetics and Molecular Biology
Journal of Machine Learning ResearchLatent dirichlet allocation
27,049 Citations2003David M. Blei, Andrew Y. Ng +1 more
Probabilistic latent semantic indexing
3,916 Citations1999Thomas Hofmann
Probabilistic Latent Semantic Indexing is a novel approach to automated document indexing which is based on a statistical latent class model for factor analysis of count data.
ArnetMiner
2,084 Citations2008Jie Tang, Jing Zhang +4 more
The architecture and main features of the ArnetMiner system, which aims at extracting and mining academic social networks, are described and a unified modeling approach to simultaneously model topical aspects of papers, authors, and publication venues is proposed.
arXiv (Cornell University)Supervised Topic Models
1,315 Citations2010David M. Blei, Jon McAuliffe
The supervised latent Dirichlet allocation (sLDA) model, a statistical model of labelled documents, is introduced, which derives a maximum-likelihood procedure for parameter estimation, which relies on variational approximations to handle intractable posterior expectations.
Uncertainty in Artificial IntelligenceThe author-topic model for authors and documents
1,313 Citations2004Michal Rosen‐Zvi, Thomas Griffiths +2 more
Topics over time
1,132 Citations2006Xuerui Wang, Andrew McCallum
An LDA-style topic model is presented that captures not only the low-dimensional structure of data, but also how the structure changes over time, showing improved topics, better timestamp prediction, and interpretable trends.
Journal of Biomedical InformaticsBio2RDF: Towards a mashup to build bioinformatics knowledge systems
798 Citations2008François Belleau, Marc-Alexandre Nolin +3 more
The present article details this new approach to build mashups of bioinformatics data and illustrates the building of a mashup used to explore the implication of four transcription factor genes in Parkinson's disease.
Probabilistic author-topic models for information discovery
581 Citations2004Mark Steyvers, Padhraic Smyth +2 more
The methodology is applied to a large corpus of 160,000 abstracts and 85,000 authors from the well-known CiteSeer digital library, and a model with 300 topics is learned using a Markov chain Monte Carlo algorithm.
D2R Server - Publishing Relational Databases on the Semantic Web
309 Citations2004Christian Bizer, Richard Cyganiak +1 more
D2R Server is a tool for publishing the content of relational databases on the Semantic Web that allows Web agents to retrieve RDF and XHTML representations of resources and to query non-RDF databases using the SParQL query language over the SPARQL protocol.
BMC BioinformaticsChem2Bio2RDF: a semantic framework for linking and data mining chemogenomic and systems chemical biology data
230 Citations2010Bin Chen, Xiao Dong +5 more
The utility of Chem2Bio2RDF is demonstrated in investigating polypharmacology, identification of potential multiple pathway inhibitors, and the association of pathways with adverse drug reactions, as well as extending SPARQL with cheminformatics and bioinformatics functionality.
Distributed Inference for Latent Dirichlet Allocation
220 Citations2007David Newman, Padhraic Smyth +2 more
Using five real-world text corpora, it is shown that distributed learning works very well for LDA models, i.e., perplexity and precision-recall scores for distributed learning are indistinguishable from those obtained with single-processor learning.
Scholarworks (University of Massachusetts Amherst)Topics over Time: A NonMarkov ContinuousTime Model of Topical Trends
217 Citations2006Xuerui Wang, Andrew McCallum
Lecture notes in computer sciencePLDA: Parallel Latent Dirichlet Allocation for Large-Scale Applications
190 Citations2009Yi Wang, Hongjie Bai +3 more
PLDA smooths out storage and computation bottlenecks and provides fault recovery for lengthy distributed computations and can be applied to large, real-world applications and achieves good scalability.
PLoS Computational BiologyLiterature Mining for the Discovery of Hidden Connections between Drugs, Genes and Diseases
163 Citations2010Raoul Frijters, Marianne van Vugt +4 more
The results show that CoPub Discovery is able to identify novel associations between genes, drugs, pathways and diseases that have a high probability of being biologically valid, making it a useful tool to unravel the mechanisms behind disease, to find novel drug targets, or to finding novel applications for existing drugs.
Computational biologyNatural Language Processing and Systems Biology
96 Citations2007Kevin Bretonnel Cohen, Lawrence Hunter
This chapter outlines the basic families of applications of natural language processing techniques to questions of interest to systems biologists and describes publicly available resources for such applications.
Tag refinement by regularized LDA
92 Citations2009Hao Xu, Jingdong Wang +2 more
A novel graphical model, regularized Latent Dirichlet Allocation (rLDA), is presented, which facilitates the topic modeling by exploiting both the statistics of tags and visual affinities of images in the corpus.
BMC BioinformaticsCoPub Mapper: mining MEDLINE based on search term co-publication
75 Citations2005Blaise Alako, Antoine Veldhoven +6 more
The CoPub Mapper program allows for quick and versatile querying of co-published genes and keywords and can be successfully used to cluster predefined groups of genes and microarray data.
BMC BioinformaticsIdentifying biological concepts from a protein-related corpus with a probabilistic topic model
70 Citations2006Bin Zheng, David C. McLean +1 more
The major and recurring biological concepts within a collection of protein-related MEDLINE documents can be extracted by the LDA model and provide parsimonious and semantically-enriched representation of the texts in a semantic space with reduced dimensionality and can be used to index text.
NeuroImageTractography segmentation using a hierarchical Dirichlet processes mixture model
68 Citations2010Xiaogang Wang, W. Eric L. Grimson +1 more
A new nonparametric Bayesian framework to cluster white matter fiber tracts into bundles using a hierarchical Dirichlet processes mixture (HDPM) model that does not require computing pairwise distances between fibers and can cluster a huge set of fibers across multiple subjects.
ACM SIGMOD RecordThe ρ operator
65 Citations2002Kemafor Anyanwu, Amit Sheth
An approach that supports querying for Semantic Associations on the Semantic Web is introduced and a formal basis to the notion of Semantic Associations is given and a brief discussion on the overall approach for discovering and ranking them is given.
Linking open drug data
59 Citations2009Anja Jentzsch, Jun Zhao +4 more
The different datasets published by the Linking Open Drug Data project are described, which are strongly interlinked with other Linked Data sources and contain 8.4 million RDF triples.
Anticipating annotations and emerging trends in biomedical literature
47 Citations2008Fabian Mörchen, Mathäus Dejori +4 more
It is described how probabilistic topic models can be used to annotate recent articles with the most likely MeSH terms, providing users with a competitive advantage because, when searching for MeSH Terms, articles are found long before they are manually annotated.
BMC BioinformaticsStatistical modeling of biomedical corpora: mining the Caenorhabditis Genetic Center Bibliography for genes related to life span
40 Citations2006DM Blei, Katie Franks +2 more
Like other graphical models for genetic, genomic and other types of biological data, LDA provides a method for extracting unanticipated insights and generating predictions amenable to subsequent experimental validation.
BIOSILICOMining the biomedical literature using semantic analysis and natural language processing techniques
32 Citations2003Ronen Feldman, Yizhar Regev +2 more
This work focuses on text mining as applied to the biomedical literature, in particular on finding relationships among genes, proteins, drugs and diseases, to facilitate an understanding and prediction of complex biological processes.
Expert Opinion on Drug DiscoveryMining large heterogeneous data sets in drug discovery
19 Citations2009David Wild
The future of mining large data sets for drug discovery requires intelligent, semantic aggregation of information from all of the data sources described in this review, along with the application of advanced methods such as intelligent agents and inference engines in client applications.
Lecture notes in computer scienceTractography Segmentation Using a Hierarchical Dirichlet Processes Mixture Model
8 Citations2009Xiaogang Wang, W. Eric L. Grimson +1 more
A new nonparametric Bayesian framework to cluster white matter fiber tracts into bundles using a hierarchical Dirichlet processes mixture (HDPM) model that does not require computing pairwise distances between fibers and can cluster a huge set of fibers across multiple subjects without subsampling.
PubMedPubMed Interact: an interactive search application for MEDLINE/PubMed.
6 Citations2006Michael Muin, Paul Fontelo +1 more
Using PHP, Document Object Model (DOM) manipulation and Asynchronous JavaScript and XML (Ajax), PubMed Interact allows greater functionality so users can refine search parameters with ease and interact with the search results to retrieve and display relevant information and related articles.