Proceedings · Session S-775 · filed November 19, 2026
AI & Emerging Tech in R&DSession paper
Chugai becomes first pharma partner in Ten63's simulation push on undruggable targets
Chugai joins Ten63's BEYOND-Alpha program to receive molecular designs for undruggable targets, as the company reports 11-75% hit rates against MYC and HPV proteins.
By Priya Raman4 min read791 words
Summary
- Chugai Pharmaceutical is the first participant in Ten63's BEYOND-Alpha early-adopter program, with rights to negotiate development of platform-designed molecules.
- Ten63 reports prospective hit rates of 11% to 75%, SPR-confirmed, using fewer than 20 synthesized molecules per program.
- BEYOND ran 260.3 molecular dynamics steps per second on an Nvidia A100, versus 0.0188 for Meta's UMA-M, per a company benchmark.
- Chugai Venture Fund and the Gates Foundation invested in Ten63's February financing; Mike Dial joined the board.
- A Gates Foundation cervical cancer grant funds Ten63's program against HPV proteins including E6.

Chugai Pharmaceutical, a member of the Roche Group, has joined Ten63 Therapeutics's BEYOND-Alpha early-adopter program as its first participant, gaining the right to nominate difficult drug targets and receive ranked molecular designs from the Durham, N.C.-based biotech's simulation platform. Chugai can negotiate terms for further research, development and commercialization of molecules containing those designs under a separate definitive agreement. Chugai Venture Fund had already invested in Ten63's February financing alongside the Gates Foundation, and its managing director, Mike Dial, joined the company's board.
The partnership tests a specific thesis: that physics-based simulation can supply the binding data which "undruggable" targets, lacking conventional pockets, have never generated. "By definition there's no positive data for undruggables, and it's very difficult to train an AI model on a relationship where there are no positive results," said Ten63 co-founder and CEO Marcel Frenkel, Ph.D.
What does the platform actually measure?
Ten63, founded in 2018 as a spinout from Duke University and the Toyota Technological Institute at Chicago, claims its BEYOND platform searches a chemical space of more than 100 trillion synthetically accessible molecules. A benchmark on the company's website reports 260.3 molecular dynamics simulation steps per second on a 2020-vintage Nvidia A100 GPU, against 0.0188 for UMA-M, a machine-learned model from Meta. The company names itself after 1063, the estimated number of drug-like molecules that follow Lipinski's rules.
Frenkel was candid about the gap between simulation and bench work: BEYOND's simulations have "several iterations to go" before matching experimental accuracy. "Once that's possible, BEYOND will allow you to create unlimited synthetic data at experimental accuracy, and that changes the game," he said.
Prospective results, which the company reports itself, show hit rates of 11% to 75% in recent studies, confirmed by surface plasmon resonance, with fewer than 20 molecules synthesized and tested per program. Those figures cover targets where co-folding models struggle, but sample sizes remain small and Ten63 disclosed no external validation.
Why MYC is the proving ground
MYC, a cancer-implicated transcription factor that Frenkel said has been undruggable for more than 40 years, anchors the company's internal pipeline. "This is a protein that's moving all the time and is highly dynamic, and really doesn't have a natural pocket," he said. "It's the canonical undruggable target that we want to solve."
The most advanced clinical effort against MYC belongs to Barcelona-based Peptomyc, whose OMO-103, a 91-amino-acid miniprotein delivered by weekly intravenous infusion, showed target engagement and preliminary clinical activity in a phase 1 trial published in Nature Medicine in 2024. Ten63 is pursuing an oral small molecule instead.
When Ten63 tested published and patented anti-MYC compounds from academic, pharma and biotech groups side by side with its own molecules in several assays, Frenkel said "the grand majority" failed to hit MYC. "We're seeing some positive data around assays that other groups have found challenging," he said. "Still early days, still a lot to develop." He noted that more than 20,000 papers link MYC to cancer and that the protein affects "12,000 genes"; a 2012 Cell study found MYC at nearly every active gene in immune and embryonic stem cells, amplifying their output, though some later studies challenged that model.
How the technology spread across targets
The approach grew out of the Duke lab of co-founder Bruce Donald, whose work applied provable search algorithms to protein and molecule shapes and, in the 2010s, predicted drug resistance and overcame it computationally. Ten63 now runs more than a dozen programs. A Gates Foundation grant for a global cervical cancer strategy funds work on HPV proteins, including E6, whose biology Frenkel called well understood but biophysically "very tricky." Because they are viral, "there's nothing in the human that's going to care that you hit these targets," he said.
Frenkel framed BEYOND-Alpha as a way to diversify biology risk while the platform matures. He compared its trajectory to OpenAI's GPT-4, which METR estimated succeeded about half the time on software tasks taking human experts roughly four minutes; for GPT-5.4, released in March 2026, that success rate extended to tasks of about 5.7 hours.
On Demis Hassabis's goal of curing all disease, Frenkel called it a "lofty vision," arguing that chemistry which unlocks more targets would let researchers test more biology in humans. The Chugai agreement gives Ten63 its first external read on whether large-scale simulation can deliver what he called "the unlocking step," the kind of breakthrough Kevan Shokat's UCSF lab achieved for KRAS after roughly four decades of failure.
via R&D World (Source)
Filed under
- computational-drug-discovery
- undruggable-targets
- molecular-dynamics
- ai-drug-discovery
- chugai-pharmaceutical
More from Priya Raman
Show full bio
Staff writer covering money, funding and market shifts around research and development at Hypothesis Wire.
283 articles
References
- Isomorphic Labs predicts AI-designed drugs as $5bn raise talks surface
- Nature asks if AI can restructure translational research pipeline
- AI accountability, FDA shifts and a $1.15 billion deal
- HHS Bets on AI-Driven Seamless Trials With New ARPA-H Program
- Isomorphic Labs banks $5 billion in biotech's largest venture round