We engineer high-affinity protein binders and rank the most clinically viable therapeutic candidates, providing transparency into the metrics driving each ranking.
Dr. Shen Yan has joined Interlit as co-founder and Chief Scientific Officer, with a PhD in Molecular and Translational Biomedicine, an MBA, and 15+ years of wet lab experience.
01
A single antibody sequence has more possible variants than atoms in the observable universe.
A 100-residue protein domain has 20^100 (about 1.3x10^130) possible sequences; the observable universe holds an estimated 10^80 atoms.
02
It is prohibitively expensive to test every candidate, and most fail late for reasons that have nothing to do with binding: developability, aggregation, immunogenicity.
03
We open the ranking, so you see why each candidate scores as it does.
The EMA states an a priori preference for transparent, interpretable models. (EMA reflection paper on AI in the medicinal product lifecycle, 2024.)
We start from your starting sequence(s), design new therapeutic protein binders for the desired target, and rank the most clinically viable candidates, narrowing down to a few you can test.
Each ranking shows its developability and immunogenicity scores, so you decide which risks matter most for your pipeline.
Fewer designs to synthesize means less wet-lab spend and a shorter research cycle.
The lab results
In the 2026 EGFR/cetuximab binder optimization competition, three of our ten designs surpassed the previous world best, and the candidate we ranked first proved to be the tightest binder measured, at 43.9 times tighter than the cetuximab starting antibody and 1.6 times tighter than the best prior design, all done in under one day.
Affinities independently measured by Adaptyv, an independent wet-lab partner (SPR), as measured mid-July 2026. "World best" means the best publicly recorded result we are aware of in Adaptyv's competition data.
10 out of 10 candidates actually binding is already a great achievement, but having 3 even higher affinity than the previous best is amazing!
Our story
A two-person team I led set a world best in the 2026 EGFR/cetuximab binder optimization competition. The secret? While most teams ranked their candidates using AlphaFold confidence scores like ipTM or ipSAE, we focused on selecting the tools and metrics that correlate with the physical reality of protein binding. Today I'm building Interlit, an AI protein design company, to help biotech and pharmaceutical teams push protein design's frontier.
Early access
Tell us what you're working on. We'll reach out to you as soon as we can.
