Recursion's CEO outlines a two-pronged pipeline strategy. The first prong uses phenomics to uncover novel biological insights for new targets, like their FAP program. The second uses their AI-driven small molecule design platform to improve the therapeutic index for known but historically 'hard-to-drug' targets, like CDK7. This balanced portfolio approach de-risks development by leveraging different strengths of their end-to-end platform.

The company focuses on disease-specific 3D protein conformations, which exposes new binding sites (epitopes) not present on the same protein in healthy cells. This allows for highly selective drugs that avoid the toxicity common with targets defined by genetic sequence alone.

Instead of the traditional 'disease-target-drug' approach, Soleil finds compounds that create a desired cellular change first. Only after identifying a promising, well-tolerated molecule with a known cellular mechanism do they use bioinformatics to determine which disease and patient population it's best suited for.

Recludix succeeded in drugging SH2 domains, a target class abandoned in the 90s, by integrating five modern technologies. This platform includes proprietary DNA-encoded libraries, machine learning, a selectivity tool, novel crystallography methods, and a pro-drug approach to ensure cell permeability, demonstrating the complex approach needed for modern drug discovery breakthroughs.

The company's foundational insight is that cellular stress is a central mechanism in vastly different diseases. In cancer, they increase stress to kill cells; in degenerative conditions like Parkinson's or hair loss, they aim to decrease stress to restore function. This unifying principle allows their single platform to tackle a diverse therapeutic portfolio.

To target MYC, Dewpoint uses phenotypic screens that monitor the entire MYC condensate. This approach is mechanism-agnostic, capable of identifying compounds that work via previously attempted methods (e.g., disrupting binding) as well as novel ones like dissolving the condensate itself.

Beyond accelerating timelines, AI's real value lies in its ability to design molecules for targets previously considered 'hard-to-drug.' These models operate on different principles than traditional lab methods and are indifferent to historical challenges, opening up entirely new therapeutic possibilities.

The current, tangible breakthrough for AI in drug discovery is not identifying completely novel biological targets. Instead, it's rapidly designing effective molecules for known targets that have historically been considered "undruggable," compressing years of screening work into a month.

Soleil moves beyond the single-target model by mapping the entire flow of information a drug creates within a cell. They argue that even approved drugs have 30-40 other effects. By understanding the global cellular response from day one, they aim to better predict both efficacy and toxicity, addressing a key failure point in traditional discovery.