insitro Acquires CombinAbleAI, Launches Full-Stack AI Drug Design Platform

SOUTH SAN FRANCISCO, CA – January 12, 2026 – In a significant move to consolidate its position in the AI-driven drug discovery arms race, insitro announced today its acquisition of Israeli startup CombinAbleAI. The deal, expected to close this month, coincides with the launch of insitro’s ambitious TherML™ (Therapeutic Machine Learning) platform, creating what the company describes as a complete, modality-agnostic system for designing novel therapeutics.

The acquisition integrates CombinAbleAI’s specialized physics-informed AI for complex biologics into insitro's broader platform, which already spans small molecules and oligonucleotides. This strategic enhancement aims to industrialize drug discovery by enabling scientists to design molecules that are not only potent but also manufacturable from the very beginning, a paradigm shift intended to drastically reduce costly late-stage failures.

“We are delighted to welcome the CombinAbleAI team in Israel as fellow insitrocytes,” said Daphne Koller, Ph.D., founder and CEO of insitro. “Their addition helps launch our integrated TherML platform, which is a critical part of our end-to-end industrialized AI-enabled system built for repeatable, scalable, predictable drug discovery.”

A Strategic Play in the AI Arms Race

The acquisition arrives as the AI drug discovery landscape, valued at over $5 billion, becomes increasingly competitive. Companies like Recursion Pharmaceuticals and Insilico Medicine are also racing to build end-to-end platforms that leverage massive datasets and machine learning to map biology and generate novel drug candidates. In this environment, building a comprehensive, integrated system is seen as a critical strategic advantage.

insitro's move to create a “full-stack, modality-agnostic” platform is central to this strategy. By unifying design capabilities across all major drug types—from small molecules to complex multi-specific antibodies—the TherML platform allows the underlying biology of a disease to guide the choice of therapeutic, rather than being constrained by a company's internal expertise in a specific modality. This flexibility is crucial for tackling a wider range of diseases with the most appropriate tool.

Backed by approximately $800 million in capital from world-class investors, insitro has been methodically building its capabilities. The company leverages its causal biology platform to identify disease drivers from vast human datasets and is now positioning TherML as the engine to translate those biological insights into clinic-ready medicines. The integration of CombinAbleAI is the final piece in constructing this end-to-end vision.

Redefining 'Clinic-Ready' with Integrated Design

At the heart of insitro’s newly enhanced platform is a fundamental shift away from the industry's traditional, sequential drug design process. For decades, the primary focus of early discovery has been on maximizing a molecule's potency, or its ability to hit a biological target. Critical properties related to developability—such as stability, solubility, and ease of manufacturing—were often addressed much later, leading to a high attrition rate where promising candidates failed because they couldn't be practically developed into a safe and stable drug.

“Drug discovery has traditionally optimized molecules for potency before assessing developability – often discovering that highly potent candidates face manufacturing constraints,” explained Philip Tagari, insitro's Chief Scientific Officer. “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.”

This simultaneous optimization is powered by sophisticated technology. For complex biologics, TherML now incorporates CombinAbleAI's engine, which was pre-trained on over 100,000 molecular dynamics surrogates. In essence, it uses AI models informed by the fundamental physics of how proteins fold and interact to predict not only a drug's efficacy but also its structural behavior and stability. This allows for the design of complex molecules like multi-specific antibodies and T-cell engagers with greater precision and a higher probability of success.

“Effective therapeutic design requires optimization across affinity, selectivity, stability, and manufacturability,” said Noam Katz, Ph.D., co-founder of CombinAbleAI. “We’re very excited to join insitro and integrate CombinAbleAI’s physics-informed AI modeling into insitro’s end-to-end discovery platform to reliably and rapidly translate high-value targets into molecules that perform as intended in the real world.”

This integrated approach is directly connected to insitro’s automated labs, creating a closed-loop system. The platform’s AI models generate molecular designs, which are then rapidly synthesized and tested experimentally. The resulting data is fed back into the models, continuously improving their predictive power in an iterative cycle of design, build, test, and learn.

A Triumph for Israel's Biotech Innovation Engine

The acquisition is not only a strategic win for insitro but also a major validation for AION Labs, the unique Israeli venture studio where CombinAbleAI was created and incubated in 2023. AION Labs was founded by a powerful coalition of pharmaceutical giants—including Pfizer, AstraZeneca, Merck KGaA, and Teva—along with Amazon Web Services (AWS) and leading venture funds. Its mission is to build AI-native companies that solve critical challenges in drug discovery.

The successful exit of CombinAbleAI showcases the power of this model, which brings together corporate R&D needs, entrepreneurial talent, and venture funding to fast-track innovation. It also shines a spotlight on Israel's burgeoning status as a global hub for AI and biotechnology. The nation has cultivated a vibrant ecosystem supported by government initiatives like the Israel Innovation Authority, a deep talent pool, and a robust flow of venture capital into its life sciences sector.

The CombinAbleAI team will continue to operate from its base in Rehovot, Israel, forming a new insitro R&D center dedicated to advancing large molecule design. This move anchors a key part of insitro’s global R&D footprint in a thriving innovation hub, ensuring continued access to top-tier talent and scientific collaboration.

By integrating CombinAbleAI’s deep expertise into its platform, insitro is betting that a holistic, predictive, and data-driven approach is the key to finally industrializing the art of drug discovery. The goal is to make the process of creating new medicines more scalable and predictable, ultimately shortening the long and uncertain path from biological insight to patient impact.