Biomedical Text Mining and Ontologies

The paper describes a literature mining component of the Open Targets platform that identifies and ranks target-disease associations from PubMed and Open Access articles.

MetaSRA normalizes SRA sample metadata using a schema inspired by ENCODE, enabling better analysis of biological relationships across diverse samples.

Gene ORGANizer links human genes to anatomical parts using a curated database, revealing trends like X chromosome enrichment in facial feature genes.

OncoScore is a text-mining tool that ranks genes by their association with cancer, achieving high accuracy in prioritizing cancer-related genes.

The paper implements a BeCalm API for efficient and stable named entity recognition in biomedical text, supporting multiple document sources and handling large volumes of abstracts quickly.

The paper describes an updated text-mining tool, EXTRACT 2.0, designed to assist in the interactive annotation of biomedical named entities and ontology terms, improving usability and curation efficiency.

WikiGenomes provides a web platform for community curation of gene annotations in Wikidata, addressing the lack of resources for comprehensive genomic data curation.

The paper assesses the sustainability of expert curation in UniProtKB/Swiss-Prot by analyzing literature triage and curatorial workload.

This study evaluates how varying levels of annotation granularity affect semantic similarity metrics in phenotype profiles, finding that coarse annotations may suffice for detecting significant similarity.

This paper presents a binary relation extraction system using dependency tree features and SVMs, showing improved performance over co-occurrence and rule-based methods.

The paper introduces goldi, an R package for identifying ontological terms in text, using word matching and an acceptance function.