AI therapeutics company, insitro, has agreed to acquire CombinAbleAI in a move to expand its artificial intelligence platform across multiple drug modalities, including biologics, small molecules and oligonucleotides.

The acquisition, expected to close in late January 2026, will see CombinAbleAI integrated into insitro’s research and development organisation, with the Israeli team becoming a new R&D centre focused on large molecule design. Financial terms were not disclosed.

The deal is intended to complete insitro’s end-to-end, modality-agnostic approach to drug discovery, enabling the company to apply AI-driven design across the full development workflow, from target discovery through to developability assessment.

According to insitro, combining CombinAbleAI’s physics-informed modelling capabilities with its existing causal biology platform will allow potency and manufacturability to be optimised in parallel, rather than sequentially, during early-stage development.

Philip Tagari, chief scientific officer at insitro, said: “Drug discovery has traditionally optimised molecules for potency before assessing developability – often discovering that highly potent candidates face manufacturing constraints. By integrating CombinAbleAI’s physics-informed, AI-driven design for complex biologic therapeutics with our causal biology platform, we treat potency and manufacturability as interdependent design criteria from the outset.”

CombinAbleAI specialises in AI-driven optimisation of complex biologics, including multi-specific antibodies and T-cell engagers. Its platform uses physics-informed models trained on large-scale molecular dynamics data to predict protein structure and flexibility, with the aim of improving stability and manufacturability alongside biological activity.

insitro said the combined platform will support multiple therapeutic modalities within a single AI framework. For small molecules, the company applies its proprietary quantitative adaptive libraries to explore chemical space and generate high-resolution training data. For oligonucleotides, AI models and automation are used to optimise siRNA design and target knockdown across a range of disease areas.

The platform also incorporates machine learning models for predicting ADMET properties, drawing on both internal datasets and data generated through industry collaborations, with the goal of reducing late-stage attrition by identifying drug-like liabilities earlier in development.

Daphne Koller, founder and chief executive officer at insitro, said: “We are delighted to welcome the CombinAbleAI team in Israel as fellow insitrocytes, alongside our colleagues in the US, Poland and Malaysia. Their addition helps launch our integrated platform, which is a critical part of our end-to-end AI-enabled system built for repeatable and scalable drug discovery.”

CombinAbleAI was created within AION Labs, an Israeli biotech venture studio focused on building AI-native drug discovery companies in collaboration with pharmaceutical and technology partners. Following completion of the acquisition, the CombinAbleAI team will continue operating in Israel while working closely with insitro’s global research organisation.

Noam Katz, co-founder of CombinAbleAI, said: “Effective therapeutic design requires optimisation across affinity, selectivity, stability and manufacturability. We’re very excited to join insitro and integrate our physics-informed AI modelling into an end-to-end discovery platform to translate high-value targets into molecules that perform as intended.”

insitro said the expanded platform will be used to advance both internal programmes and partnered drug discovery efforts, as the company continues to scale its AI-driven approach to therapeutic design.