GenoPeptide designs therapeutic peptides tailored to each individual's genome and biomarker profile. In-silico design in minutes, not years.
Peptide therapy is a $141B market growing to $295B by 2033. But every peptide on the market is one-size-fits-all. BPC-157, Thymosin Alpha-1, GHK-Cu — same molecule for every genome. Each person's receptor variants, enzyme polymorphisms, and metabolic pathways make generic peptides suboptimal or ineffective for many individuals.
GenoPeptide designs therapeutic peptides tailored to each individual's genome and biomarker profile. Using Evo 2 (Arc Institute, 40B param genomic model) + AlphaFold 3 (DeepMind) for in-silico design and structural validation. From variant to validated candidate in minutes, not years.
Upload a VCF file or connect a genomics provider. We extract clinically significant variants across pharmacogenomic, metabolic, and immune-related genes.
ClinVar annotates each variant for clinical significance. UniProt maps variants to protein targets, binding sites, and functional domains.
Evo 2 (40B parameters, trained on 9.3 trillion nucleotides) generates candidate peptide sequences optimized for each target protein, respecting the individual's genomic context.
ESMFold predicts 3D structure of each candidate. Binding affinity, stability, and solubility are scored computationally before any wet lab work.
Interactive report with 3D molecular viewers, scoring breakdown, and SPPS synthesis blueprints ready for compounding pharmacies (503A/503B pathway).
Tested on NA12878 — CEPH/Utah female from the 1000 Genomes Project and GIAB gold standard reference.
Targets include MTHFR, CYP2C19, CYP2D6, VKORC1, OPRM1, TCF7L2, IL6, IL10 — covering pharmacogenomics, metabolic, and immune pathways.
Each member's biomarker + genomic + outcome data improves AI models for everyone
More members = better predictions across genetic clusters
Members won't leave a platform that designs molecules for their DNA
In-silico design in minutes vs years for traditional pharma R&D
Novel peptide sequences = patentable compositions of matter
503A/503B compounding pathway — no FDA approval needed to launch
Full pipeline running on real human genome with ClinVar annotation, Evo 2 peptide generation, ESMFold 3D structure prediction, and interactive report.
Request a Demo