You can also check out my Google Scholar profile.
LIGNOVA: An automated pipeline for large-scale generation of high-quality protein–ligand complexes
Computational Medicinal Chemistry School · Cambridge, MA · October 2026
LIGNOVA docks bioactive PubChem compounds into their known protein targets with GNINA, then filters for physically plausible poses to build large-scale receptor-bound data for training next-generation ML models — 125M+ poses across 175,200 compounds and 4,180 protein structures so far, with an NSF ACCESS allocation to scale it nationally. My role: lead developer.
Benchmarking consensus rescoring against explicit water in protein–ligand pose prediction
Manuscript in preparation · 2026
Benchmarked molecular docking methods under three hydration conditions to pinpoint when consensus-based rescoring can replace expensive explicit-water modeling — preserving the throughput needed for virtual screening. My role: lead author.
Putting the pieces together in silico to build drugs from fragments
MCDB Noon Seminar Series · Durrant Lab, University of Pittsburgh · Jan 2026
Department seminar on assembling fragment-based drug design components into an end-to-end in silico pipeline.
Hypermutable hotspot enables the rapid evolution of self/non-self recognition genes in Dictyostelium
PNAS · 2025 · 122(51)
Investigated how the social amoeba Dictyostelium discoideum maintains the extreme genetic diversity its self/non-self recognition system requires. We found that the recognition genes tgrB1 and tgrC1 sit in a hypermutable genomic hotspot that generates new alleles faster than selection can fix any one of them — resolving Crozier’s paradox, a long-standing puzzle in the evolution of kin recognition. My role: co-first author where I led the computational genomics, genome assembly, QC, and targeted annotation across 10 chromosome-length genomes.
DeepFrag meets LIGNOVA: Transforming drug design with fragment-based lead optimization
MCDB Noon Seminar Series · Durrant Lab, University of Pittsburgh · Nov 2024
Department seminar connecting the data generated from the LIGNOVA pipeline with DeepFrag for fragment-based lead optimization.
Beyond the Blueprint: A Novel Database for Innovative Drug Design and Discovery
Gordon Research Conference on Computational Chemistry · Portland, ME · 2024
Presented the database concept behind LIGNOVA — docking PubChem compounds into PDB targets with ComBind rescoring — validated on human aldose reductase and anaplastic lymphoma kinase.
Beyond the Blueprint: A Novel Database for Innovative Drug Design and Discovery
MCDB Noon Seminar Series · Durrant Lab, University of Pittsburgh · Jan 2024
Department seminar presenting the database concept behind LIGNOVA.
MolModa: accessible and secure molecular docking in a web browser
Nucleic Acids Research · 2024 · 52(W1), W498–W506
MolModa is a web-based molecular docking tool that makes protein–ligand docking more accessible, secure, and efficient — running docking workflows directly in the browser, with no installation or advanced technical skills required, while preserving data privacy and reproducibility.
From byte to bench to bedside: molecular dynamics simulations and drug discovery
BMC Biology · 2023 · 21(1), 299
A review of how advances in molecular dynamics (MD) simulations — and their integration with machine learning and quantum approaches — can bridge the gap between computational modeling and experimental drug discovery.
Augmenting Protein–Ligand Complex Databases with PubChem for Enhanced Drug Discovery
Gordon Research Conference & Seminar on Computer-Aided Drug Design · West Dover, VT · 2023
Presented a workflow assembling ~350,000 modeled protein–ligand complexes from PubChem to expand ML training data, demonstrated on the oncogenic kinase CLK2.
Testing the impact of a newly built receptor–ligand dataset on DeepFrag accuracy
MCDB Noon Seminar Series · Durrant Lab, University of Pittsburgh · Sep 2022
Department seminar evaluating how a newly generated receptor–ligand dataset affects DeepFrag’s predictive accuracy.