Gregory W. Kyro, PhD

Research

Selected Publications18
Open-Source Software6

Machine Learning Frameworks

  • T-ALPHA
    A hierarchical transformer-based deep neural network for protein-ligand binding affinity prediction with uncertainty-aware self-learning for protein-specific alignment. Currently the state-of-the-art model for predicting protein-ligand binding affinity.

  • CardioGenAI
    A generative deep learning framework for re-engineering drugs to reduce hERG-related cardiotoxicity while preserving primary pharmacology. Successfully applied in Pfizer R&D programs.

  • ChemSpaceAL
    The first active learning methodology for fine-tuning molecular generative models toward specified protein targets. Currently being used in collaboration with Brown University for designing small-molecule binders to CRISPR-Cas9.

  • HAC-Net
    A hybrid attention-based convolutional neural network for highly accurate protein-ligand binding affinity prediction. Used to identify potential inhibitors for G protein-coupled receptors and antivirulence drugs.

Computational Biology Tools

  • Molecular Dynamics Analyses
    A comprehensive toolkit for analyzing molecular dynamics simulations, with specialized focus on protein allostery.

  • Eigenvector Centrality
    A novel method for analyzing information transfer in proteins using eigenvector centrality in protein structure networks. Provides insights into allosteric mechanisms of biological systems including CRISPR-Cas9, imidazole glycerol phosphate synthase, and D-dopachrome tautomerase.

Innovations Acquired by Pharmaceutical Companies1
  • SAGE Platform
    An innovative platform that incorporates AI, cutting-edge reactions, and automated synthesis to reduce attrition in early-stage drug discovery.
    Acquired by: Merck Group (August 2024)
Selected Presentations & Lectures33
  • “Industry Trends in AI for Scientific Discovery”. Invited Interview at PhD Pathways Speaker Series, Yale University (2025).
  • “Artificial Intelligence for Preclinical Drug Safety”. Invited Round-Table Discussion at Novartis Institutes for Biomedical Research (2025).
  • “Modeling Protein-Ligand Interactions and Generative AI for Lead Optimization”. Invited Round-Table Discussion at Novartis Institutes for Biomedical Research (2025).
  • “Quantum-Classical Machine Learning Methods for Optimizing Drug Toxicity”. Invited Research Talk at QuantumCT Industry Collaboration Forum, Yale Ventures (2025).
  • “Exploring How Quantum Computing Will Impact Pharmaceutical Research”. Invited Technology Panel Discussion at QuantumCT Industry Collaboration Forum, Yale Ventures (2025).
  • “Current State-of-the-Art Deep Learning Models for Protein Design”. Invited Research Talk at Biophysical Chemistry Seminar, Yale University (2025).
  • “Introduction to Deep Learning for Biochemistry”. Invited Guest Lecture at CHEM 584: Machine Learning and Quantum Computing, Yale University (2025).
  • “Transformers for Modeling Protein-Ligand Interactions”. Poster Presentation at Chemical Research Symposium, Yale University (2025).
  • “A Hybrid Quantum-Classical Transformer Architecture with a Quantized Self-Attention Mechanism Applied to Molecular Generation”. Poster Presentation at Chemical Research Symposium, Yale University (2025).
  • “Applications of Deep Learning in Cardiovascular and Safety Domains”. Invited Round-Table Discussion at Novartis Institutes for Biomedical Research (2025).
  • “Machine Learning for Modeling Cardiac Ion Channels”. Invited Research Talk at Scientific Seminar, Novartis Institutes for Biomedical Research (2025).
  • “T-ALPHA: A Hierarchical Transformer-Based Deep Neural Network for Protein–Ligand Binding Affinity Prediction with Uncertainty-Aware Self-Learning for Protein-Specific Alignment”. Invited Research Talk at 10th Annual Biophysics and Structural Biology Research Symposium, Yale University (2025).
  • “Assessment of Different Machine Learning Architectures for hERG Activity Prediction”. Poster Presentation at Computational Medicinal Chemistry School, Novartis Institutes for Biomedical Research (2024).
  • “Some of the Problems with Public Datasets for Protein-Ligand Binding Affinity Prediction”. Invited Research Talk at Computational Biophysics and Drug Design Meeting, Bernal Institute at the University of Limerick (2024).
  • “Generative AI Methods for Lead Optimization in Drug Discovery”. Poster Presentation at Chemical Research Symposium, Yale University (2024).
  • “CardioGenAI: A Machine Learning Framework for Re-engineering Drugs for Reduced hERG Liability”. Invited Research Talk at Innovation Cup Alumni Symposium, Merck Group (2024).
  • “A Machine Learning Framework for Re-engineering Drugs for Reduced hERG Liability”. Invited Research Talk at Global Discovery Investigative Toxicology and Translational Sciences—Computational Safety Sciences Town Hall, Groton Laboratories at Pfizer Research & Development (2024).
  • “CardioGenAI: A Machine Learning-based Framework for Re-engineering Drugs for Reduced hERG Liability”. Poster Presentation at Summer Intern Poster Session, Groton Laboratories at Pfizer Research & Development (2024).
  • “A Generative AI-based Framework for Toxicity Applications in Early-Stage Drug Development”. Poster Presentation at 9th Annual Biophysics and Structural Biology Research Symposium, Yale University (2024).
  • “A Hybrid Quantum-Classical Machine Learning Framework for Drug Toxicity Applications”. Invited Research Talk at QuantumCT Industry Collaboration Forum, Yale West Campus (2024).
  • “Generative Machine Learning and Active Learning Methods for Hit Identification in Drug Discovery”. Poster Presentation at Sterling Chemistry Laboratory 101st Anniversary Symposium, Yale University (2024).
  • “CardioGenAI: A Machine Learning-based Framework for Re-engineering Drugs for Reduced hERG Liability”. Poster Presentation at 19th Annual Drug Discovery Chemistry Conference (2024).
  • “ChemSpaceAL: An Efficient Active Learning Methodology Applied to Protein-Specific Molecular Generation”. Invited Research Talk at Annual Biophysical Society Meeting (2024).
  • “Machine Learning and Statistical Methods for Modulating Protein Function with Small Molecule Inhibitors”. Invited Research Talk at National Institutes of Health Biophysics Seminar, Yale University (2023).
  • “HAC-Net: A Hybrid Attention-based Convolutional Neural Network for Highly Accurate Protein-Ligand Binding Affinity Prediction”. Poster Presentation at Annual Biophysical Society Meeting (2023).
  • “Introduction to Deep Learning for Chemistry”. Invited Guest Lecture at CHEM 584: Machine Learning and Quantum Computing, Yale University (2022).
  • “Photophysics of Binuclear Rhenium(I) Tricarbonyl Complexes and Their Employment as Anion Sensors Through Charge-Mediated Hydrogen Bonding”. Poster Presentation at 261st American Chemical Society National Meeting & Exposition (2021).
  • “Variable Anion Recognition Sites in Phosphorescent Rhenium(I) Polypyridyl-based Sensors”. Poster Presentation at 259th American Chemical Society National Meeting & Exposition (2020).
  • “Photophysics of Polypyridyl-based Rhenium(I) Complexes and Their Employment as Anion Sensors”. Poster Presentation at 3rd SUNY Binghamton Conference in Chemistry Research (2020).
  • “Highly Sensitive Rhenium(I) Sensors for Anions Through Amide Hydrogen Bonding”. Poster Presentation at Undergraduate Research Conference, SUNY Binghamton (2020).
  • “Amide Protons as Binding Groups in a Polypyridyl-based Rhenium(I) Anion Sensor”. Poster Presentation at 257th American Chemical Society National Meeting & Exposition (2019).
  • “Excited-State Properties of Rhenium(I)-based Anion Sensors”. Poster Presentation at 2nd SUNY Binghamton Conference in Chemistry Research (2019).
  • “Organometallic Complexes as Anion Sensors: A Highly Sensitive Rhenium(I) Complex for Cyanide and Halide Anions”. Poster Presentation at 1st SUNY Binghamton Conference in Chemistry Research (2018).
Selected Apps1

  • BibleVerse
    A modern Bible interface with latent-space verse and passage retrieval, scripture-augmented LLMs designed for breadth- and depth-first knowledge expansion, and a reader with chapter-conditioned LLM chat and annotation.

Scientific Peer Review Contributions21

Reviewing Editor

  • Springer Nature

Trusted Reviewer

  • Nature Communications (Impact Factor: 15.7) | Nature Portfolio
  • npj Digital Medicine (Impact Factor: 15.1) | Nature Portfolio
  • Environmental Science & Technology (Impact Factor: 11.4) | American Chemical Society
  • Journal of Cheminformatics (Impact Factor: 8.6) | BioMed Central
  • Engineering Applications of Artificial Intelligence (Impact Factor: 7.5) | Elsevier
  • EPJ Quantum Technology (Impact Factor: 5.8) | Springer
  • Journal of Chemical Theory and Computation (Impact Factor: 5.7) | American Chemical Society
  • Journal of Chemical Information and Modeling (Impact Factor: 5.6) | American Chemical Society
  • Computational and Structural Biotechnology Journal (Impact Factor: 4.5) | Elsevier
  • Biotechnology (Impact Factor: 4.4) | Oxford University Press
  • ACS Synthetic Biology (Impact Factor: 4.3) | American Chemical Society
  • BMC Bioinformatics (Impact Factor: 3.9) | BioMed Central
  • Frontiers in Bioinformatics (Impact Factor: 3.9) | Frontiers
  • Scientific Reports (Impact Factor: 3.8) | Nature Portfolio
  • Chemical Research in Toxicology (Impact Factor: 3.8) | American Chemical Society
  • European Journal of Nuclear Medicine & Molecular Imaging Research (Impact Factor: 3.1) | Springer
  • Journal of Computer-Aided Molecular Design (Impact Factor: 3.0) | Springer
  • Journal of Molecular Modeling (Impact Factor: 2.1) | Springer
  • npj Drug Discovery (Impact Factor: Pending) | Nature Portfolio
  • npj Artificial Intelligence (Impact Factor: Pending) | Nature Portfolio