Insilico Medicine has released three open-source resources in Cell: a benchmark for evaluating AI reasoning across aging biology domains, compact language models trained on clinical and multi-omics aging data, and an agentic platform for identifying and prioritizing therapeutic targets. The work follows the company's Phase IIa report that rentosertib, an AI-discovered candidate for idiopathic pulmonary fibrosis, lowered biological age across six proteomic aging clocks. Among the genes nominated by the new toolkit is KDM1A, cited as independently validated for lifespan extension in C. elegans.
Key Points
- Open benchmark, language models and agentic platform released together
- Rentosertib reduced biological age across six proteomic aging clocks
- KDM1A nominated as target; lifespan extension shown in C. elegans
Longevity Analysis
Target discovery for aging has been constrained less by data than by the capacity to interpret it — multi-omics signals accumulate faster than they can be read against meaningful endpoints. Open benchmarks and agentic tooling shift part of that interpretive work into systems that can be audited and replicated by independent groups, which matters when the claimed output is a shift in biological age rather than a single disease marker. The rentosertib precedent is notable in that a compound developed for fibrotic lung tissue registered measurable change in proteomic aging measures, a reminder that interventions aimed at repair in one tissue can register across regenerative and immune signaling more broadly.
Original published by LT Wire.