October 16, 2025
A new web-based platform called MOlecular traNSporT inhibitoR and substrate predictOr Utility Server (MONSTROUS) is bringing artificial intelligence (AI) into early-stage countermeasure discovery and drug safety research. This system helps scientists predict harmful transporter interactions long before laboratory testing begins, saving time and resources in the development of safe and effective therapeutics.
MONSTROUS uses machine-learning models trained on thousands of public data points to identify whether a compound is likely to block or be transported by key human transporter proteins. Discovering these interactions too late in development can make promising drugs ineffective or even toxic.
“Transporters control how drugs and nutrients move in and out of cells,” explains Dr. Mohamed Diwan Mohideen AbdulHameed, the HJF scientist who led the project in collaboration with Dr. Anders Wallqvist at Biotechnology High-Performance Computing Software Applications Institute (BHSAI). “By predicting those interactions computationally, MONSTROUS allows researchers to focus experimental testing on the most promising and safest molecules.”
Dr. AbdulHameed’s team modeled 12 essential transporter proteins involved in drug absorption and elimination. For transporters with abundant data, they used graph convolutional neural networks (GCNNs), a modern AI approach that learns directly from molecular structures. For data-sparse cases, they applied cheminformatics modeling to maintain broad coverage.
Designed for accessibility, the MONSTROUS website requires no programming background. Users can upload compound structures and receive model predictions within minutes, enabling faster and safer decision-making in both DoD and academic research programs.
“Our goal was to make cutting-edge AI modeling usable by anyone involved in drug discovery,” Dr. AbdulHameed adds. “It’s freely available online, and we hope it will shorten the path from idea to impact.”
MONSTROUS is the latest of a series of publicly available AI tools developed at BHSAI to support DoD countermeasure development and threat assessment efforts. “Our AI tools are already being applied in real-world drug discovery and DoD threat assessment pipelines” Dr. AbdulHameed notes.