A comprehensive database of manually curated, evidence-based associations between human microbiota and diseases/traits. The database is being compiled using a blend of 'auto-curation' through Natural Language Processing (NLP) and in house custom algorithms, followed by a stage of rigorous manual validation.
The auto-curation sieves through published evidences, extracting a combination of disease and microbial associations, along with possible mechanisms of disease pathogenesis and/or traits.
While auto-curation significantly reduces the time and efforts, the manual validation ensures establishing evidence-based associations between microbiota (composition and abundance) and diseases / traits.
The database MicroByte as a whole, a comprehensive collection of such evidence-based associations, enabling us to utilize this data at disposal, and in combination with MetaRich and NutriBite.