Application of automatic mutation-gene pair extraction to diseases.

Area of Science:

• Bioinformatics
• Genetics
• Computational Biology

Background:

• Understanding disease mechanisms requires knowledge of mutations and their associated genes.
• Current methods for retrieving mutation-gene information from databases and literature are often incomplete, outdated, or time-consuming.
• Manual curation of mutation data is labor-intensive and prone to delays.

Purpose of the Study:

• To develop an automated system, MuGeX (Mutation Gene eXtractor), for extracting mutation-gene pairs from Medline abstracts.
• To address the limitations of manual data retrieval for disease-related genetic information.
• To facilitate faster and more accurate access to mutation-gene data for research.

Main Methods:

• Developed MuGeX, an automated system employing natural language processing techniques.
• Tested the system on a corpus of 231 Medline abstracts.
• Evaluated performance using recall and precision metrics for mutation detection and mutation-gene pair extraction, specifically for Alzheimer's disease.

Main Results:

• MuGeX achieved 85.9% recall and 95.9% precision for mutation detection.
• For Alzheimer's disease, MuGeX demonstrated 91.3% recall and 88.9% precision in identifying mutation-gene pairs.
• The system effectively automates the extraction process, overcoming manual retrieval challenges.

Conclusions:

• MuGeX significantly enhances the efficiency and accuracy of retrieving mutation-gene information from biomedical literature.
• The automated approach reduces the time and effort required for data acquisition.
• This tool supports a better understanding of disease mechanisms by providing timely and precise genetic mutation data.