OnTheFly^2.0: a text-mining web application for automated biomedical entity recognition, document annotation, network...

Area of Science:

• Bioinformatics
• Computational Biology
• Biomedical Informatics

Background:

• Vast amounts of experimental results are stored in local files, necessitating automated methods for extracting and analyzing biomedical information.
• Efficiently processing diverse document types (text, office, PDF, images) is crucial for accessing stored scientific findings.

Purpose of the Study:

• To introduce OnTheFly^2.0, a web application designed for automated extraction of biomedical entities from individual files.
• To enable users to generate informative summaries, link identified terms to databases, and perform downstream analyses like network generation.

Main Methods:

• Utilizes the EXTRACT tagging service for Named Entity Recognition (NER) of various biomedical terms (genes, chemicals, diseases, etc.).
• Supports multiple file formats including plain text, office documents, PDFs, and images.
• Integrates STRING and STITCH services for generating protein-protein and protein-chemical interaction networks.

Main Results:

• OnTheFly^2.0 successfully extracts and analyzes biomedical entities, providing linked knowledge summaries.
• Demonstrated utility in COVID-19 research by identifying inflammatory and senescence pathways involved in disease pathogenesis.
• Supports 197 species and offers functional enrichment analysis and association with diseases and PubMed entries.

Conclusions:

• OnTheFly^2.0 is a valuable tool for automated biomedical knowledge discovery from local files.
• The application enhances research by facilitating entity recognition, data integration, and network analysis.
• Its application in COVID-19 research highlights its potential for uncovering disease mechanisms and biomarkers.