tModBase v2.0: an expanded database of tRNA modifications with an AI-powered assistant.

tModBase v2.0 expands tRNA modification resources to 528 datasets across 23 species, 40 sequencing methods, and 42 modification types, enabling evidence-grounded exploration.

• Why it matters: tRNA chemical modifications affect stability, decoding fidelity, stress responses, regulation, and human disease, but growing sequencing data need integrated experimental context.
• What they did: The update compiles 68 third-generation and 210 modification-specific sequencing datasets, integrating sequencing-derived signals, condition-dependent changes, and annotations linking modifications to enzymes, variants, tsRNAs, diseases, and expression.
• The result: tModBot, an LLM with retrieval-augmented generation, supports natural-language queries grounded in retrieved tModBase evidence, improving investigation of modification diversity and dynamics.

AI summary of the abstract · may contain errors · not medical advice · read the paper.