Weakly Supervised Learning of Biomedical Information Extraction from Curated Data

January 2016 in “ BMC Bioinformatics ”

Suvir Jain, R.L. Kashyap, Tsung-Ting Kuo, Shitij Bhargava, Gordon Lin, Chun‐Nan Hsu

biomedical information extraction curated biomedical databases Genome-Wide Association Studies GWAS phenotypes ethnicity backgrounds disease/trait extraction stage-ethnicity extraction cost-sensitive learning biomedical text mining

Studysummary This study found that using curated biomedical databases as training examples for information extraction tasks in Genome-Wide Association Studies can outperform cost-insensitive methods, demonstrating their potential use without expert annotation.

Our plain-language summary. Not medical advice or a treatment recommendation. Consult a qualified healthcare professional before changing treatment. Full disclaimer ·

View this study on bmcbioinformatics.biomedcentral.com →

Discuss this study in the Community →

Similar Research

5 / 1000+ results

research MendelVar: gene prioritization at GWAS loci using phenotypic enrichment of Mendelian disease genes

2 citations , April 2020 in “bioRxiv (Cold Spring Harbor Laboratory)”

This article discusses the MendelVar webserver, which integrates Mendelian disease data with GWAS findings to prioritize candidate genes for complex traits and reports no new experimental results.

research Analysis of the human diseasome using phenotype similarity between common, genetic and infectious diseases

85 citations , June 2015 in “Scientific Reports”

This study applied semantic text-mining to identify phenotypes linked to over 6,000 diseases, demonstrating that these phenotypes can accurately identify known disease-associated genes, creating a human disease network based on phenotypic similarity.

research Evaluating HWE and Association in Genome Wide Association Studies: A Unified Procedure

June 2026 in “arXiv (Cornell University)”

This study proposes a new test for genome-wide association studies that incorporates Hardy-Weinberg equilibrium into SNP analysis, demonstrating improved power and interpretability over traditional methods, as evidenced by simulations and an alopecia study dataset.

research LabeledIn: Cataloging labeled indications for human drugs

37 citations , December 2014 in “Journal of Biomedical Informatics”

This study introduced LabeledIn, a semi-automatic, human-reviewed catalog of drug-disease treatment relationships, highlighting the important role of human input in creating accurate and detailed drug indication resources.

research Abstracts

2 citations , January 2019 in “Medizinische Genetik”

Among families with pediatric brain disease, this study identified over 200 novel genetic causes, revealing potential treatment points using drug repurposing or nutritional supplementation.