A knowledge framework for computational molecular-disease relationships in cancer

Area of Science:

• Computational Biology
• Bioinformatics
• Genomics and Proteomics

Background:

• Biomedical knowledge is expanding rapidly, necessitating efficient information integration.
• Existing gene-disease knowledge bases (KBs) and databases (DBs) often lack a computable, integrated format.
• Cancer research requires robust understanding of molecular-disease relationships.

Purpose of the Study:

• To develop a framework and evaluation criteria for a computable knowledge model.
• To assess existing KBs and DBs for their ability to represent molecular-disease relationships.
• To propose a schema for integrating diverse biomedical knowledge resources.

Main Methods:

• Evaluation of articles in major biomedical journals for molecular-disease relationships in cancer.
• Development of a knowledge model framework and evaluation criteria.
• Assessment of major KBs, DBs, and terminologies using developed criteria.

Main Results:

• Many databases map gene-disease and protein-disease relationships but not in an integrated, coded form.
• Existing resources often fail to combine high-level and specific molecular-disease information computationally.
• A gap exists in the unified, computable representation of cancer-related molecular-disease data.

Conclusions:

• A model-mediated schema is proposed to facilitate the integration of disparate biomedical knowledge resources.
• Improved integration can enhance retrieval and enable decision support tools for biomedical research.
• Standardized, computable formats are crucial for leveraging the growing volume of biomedical data.