About
Dr. Gregory W. Kyro is a Machine Learning Scientist at Lila Sciences, where he designs and executes machine learning projects toward the realization of Scientific Superintelligence. He earned his PhD in Computational Biophysical Chemistry from Yale University in three years, conducting research pertaining to the development of deep learning methods for modeling protein-small molecule interactions. His work contributed significantly to early-stage drug discovery and involved collaborations with leading companies such as Pfizer, NVIDIA, Moderna, Novartis, Merck Group, and others.
Some noteworthy developments that Gregory created during his PhD are:
- HAC-Net — a deep learning model that was the state of the art for predicting protein-ligand binding affinity — which was used to identify a potential inhibitor of a G protein-coupled receptor whose overexpression leads to cancer, diabetes, and multiple sclerosis, as well as a potential antivirulence drug for drug-resistant staphylococcal infections
- CardioGenAI — a generative deep learning framework for re-engineering drugs for reduced hERG-related cardiotoxicity while preserving their primary pharmacology — which was applied to specific programs within Pfizer R&D that were dealing with hERG liabilities
- T-ALPHA — the current state-of-the-art deep learning model for predicting protein-ligand binding affinity — which is being used in collaboration with Yale School of Medicine to identify potent small-molecule binders to the R1R2 interface of talin, thereby mitigating thoracic aortic aneurysm, a vascular disease driven by impaired cellular stiffness sensing
- ChemSpaceAL — the first active learning methodology for fine-tuning a molecular generative model toward a specified protein target — which is being utilized in collaboration with experimental biochemists at Brown University to design small-molecule binders to the HNH domain of CRISPR-Cas9 to enhance its specificity for target DNA sequences
In addition, he also developed a method for describing intraprotein information transfer as the propagation of electrostatic couplings throughout a secondary structure element-based network, which has led to valuable insights into the allosteric mechanisms of multiple important biological systems such as CRISPR-Cas9, imidazole glycerol phosphate synthase, and D-dopachrome tautomerase.
Aside from his academic achievements, Gregory contributed to the fine-tuning of OpenAI’s large language models, developed software for PROTACs screening at OpenEye Scientific, aided in the development of quantum computing-based methods for studying small molecules in collaboration with NVIDIA and Moderna, and co-developed an AI-based rapid synthesis framework that was acquired by Merck Group and is currently being integrated across multiple teams within the company.
Gregory founded and served as President of the Yale University Chapter of the Biophysical Society, published numerous papers in top-tier academic journals, won first place at highly selective and competitive competitions, presented his work at several conferences, created multiple Python packages, and established various collaborations with labs around the world. For these reasons, he has received numerous highly prestigious awards, and has been featured across multiple prominent media platforms including Yale News, Yale Alumni Magazine, the Biophysical Society, Merck Group, and others.