AI for Drug Discovery
Serhat Tetikol, PhD
Associate Director, AI & Bioinformatics · Bristol Myers Squibb
I build and lead the AI/ML and bioinformatics teams that discover better medicines — applying generative models, structure prediction, and large-scale machine learning to antibody and biologics discovery.
Boston, Massachusetts, United States
Focus
For most of the last decade I've worked at the front of AI-driven drug discovery — leading multidisciplinary teams that turn large-scale biological data into therapeutic decisions. My work centers on generative models for biologics, protein structure prediction, and multi-parameter optimization of candidates for affinity, developability, and immunogenicity, built on datasets like antibody repertoire sequencing and deep mutational scanning.
I care about the whole path from model to molecule: rigorous ML, scalable pipelines, and tight collaboration between computation and the wet lab. The goal is always the same — find the right candidate faster, and know why it's right.
Experience
-
Associate Director, AI & Bioinformatics
Bristol Myers Squibb · Cambridge, MA
Lead a multidisciplinary AI/ML and bioinformatics team advancing biologics drug discovery. We integrate generative models, structure prediction, and multi-parameter optimization with large-scale NGS data to accelerate candidate selection and predict drug-like properties, embedding these methods across the discovery pipeline with wet-lab and research-IT partners.
-
Director, Bioinformatics
Absci · Boston, MA
Led bioinformaticians and data scientists building AI-driven computational solutions for de novo antibody discovery, optimizing candidates for binding affinity, developability, and immunogenicity.
-
Mentor
Nucleate · Cambridge, MA
Advising early-stage biotech founders on translating AI and computational science into products that reach real patients.
-
Product Director, Graph Genomics (GRAF)
Seven Bridges · Boston, MA
Grew from R&D to Product Director, building the ML and computational-genomics foundation for precision medicine and population-scale sequencing. Led the international team behind GRAF, a variation-aware NGS platform that set state-of-the-art accuracy for machine-learning-based variant analysis.
Recognition & Selected Work
-
Winner — precisionFDA Truth Challenge V2 (2020)
-
Top 10 Innovations of 2020 — The Scientist
-
Patent — Systems and methods for generating graph references
Writing
Contact
Open to collaboration, advising, and speaking on AI for drug discovery.