Adam Bess
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Adam Bess has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Chris Alvin
Furman University
4 shared publications - Supratik Mukhopadhyay
Louisiana State University
3 shared publications - Magesh Rajasekaran
Department of Environmental Sciences and the Center for Computation and Technology, Louisiana State University , Baton Rouge, LA 70803
2 shared publications - Md Saiful Islam Sajol
Department of Computer Science, Louisiana State University , Baton Rouge, LA 70803
2 shared publications - Aashutosh Ghimire
Department of Biological Sciences Louisiana State University Baton Rouge Louisiana USA
1 shared publication - Amit Ranjan
Louisiana State University, Baton Rouge, United States of America
1 shared publication
Latest Publications
- Open Access
- 29 October 2026
- Published by Wiley in Molecular Informatics
- Vol. 45 (11)
Interpretable Drug–Target Prediction via Geometric Alignment: A Linear Alternative to Deep Architectural Complexity
Designing selective kinase inhibitors is a multiobjective optimization problem constrained by the structural conservation of the human kinome, along with the lack of high‐quality interaction data. Traditional deep learning methods often train complex neural networks directly on small datasets and fail to generalize to new chemical scaffolds. To this end, we present Geometric Manifold Alignment , a novel linear approach that shifts away from architectural complexity. We introduce Multi‐View Linear Manifold ( MVLM ), which aligns the latent spaces of ChemBERTa (small molecules) and ESM‐2 (protein sequences) into a shared geometric interaction space. We pretrain MVLM on the Universal Foundation Set of 473 923 high‐confidence interactions, where binding affinity is represented as geometric proximity between drug and target embeddings. Achieving an AUROC of 0.936 on KIBA and 0.928 on Davis, MVLM establishes the new state‐of‐the‐art on industry‐standard benchmarks, outperforming existing methods ( p < 0.001). In this work, interpretability is geometric where interaction scores are cosine similarities in a shared manifold, and the learned structure recovers kinase phylogenetic relationships without supervision. We also demonstrate the model’s utility as a transparent discovery engine by retrieving known clinical targets of FDA‐approved drugs. Our findings suggest that aligning pretrained chemical and biological representations provides a scalable and interpretable approach to advancing computational drug discovery.
- 29 October 2026
- Published by Wiley in Molecular Informatics
- Vol. 45 (11)
Open AccessInterpretable Drug–Target Prediction via Geometric Alignment: A Linear Alternative to Deep Architectural Complexity
Designing selective kinase inhibitors is a multiobjective optimization problem constrained by the structural conservation of the human kinome, along with the lack of high‐quality interaction data. Traditional deep learning methods often train complex neural networks directly on small datasets and fail to generalize to new chemical scaffolds. To this end, we present Geometric Manifold Alignment , a novel linear approach that shifts away from architectural complexity. We introduce Multi‐View Linear Manifold ( MVLM ), which aligns the latent spaces of ChemBERTa (small molecules) and ESM‐2 (protein sequences) into a shared geometric interaction space. We pretrain MVLM on the Universal Foundation Set of 473 923 high‐confidence interactions, where binding affinity is represented as geometric proximity between drug and target embeddings. Achieving an AUROC of 0.936 on KIBA and 0.928 on Davis, MVLM establishes the new state‐of‐the‐art on industry‐standard benchmarks, outperforming existing methods ( p < 0.001). In this work, interpretability is geometric where interaction scores are cosine similarities in a shared manifold, and the learned structure recovers kinase phylogenetic relationships without supervision. We also demonstrate the model’s utility as a transparent discovery engine by retrieving known clinical targets of FDA‐approved drugs. Our findings suggest that aligning pretrained chemical and biological representations provides a scalable and interpretable approach to advancing computational drug discovery.
- Open Access
- 13 October 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (20) , 13023-13031
e MolFrag 2 and e Synth 2: Sound Fragment-Based Molecular Decomposition and Reconstruction
We present eMolFrag2 and eSynth2, enhanced successors to eMolFrag and eSynth for fragment-based molecular decomposition and reconstruction. eMolFrag2 is a complete reimplementation of eMolFrag in Python, using RDKit and graph-based molecular representations to preserve connectivity, identify free atoms, combine adjacent linkers, and avoid aromaticity ambiguities that impair downstream reconstruction. eSynth2 generalizes fragment recombination by removing hard-coded molecular-property filters, supporting multiple bonds to a single brick atom, and adding a targeted reconstruction mode for soundness validation. Across 36,534 nonredundant DUD-E active compounds, the eMolFrag2 → eSynth2 workflow reconstructed every input molecule exactly (Tanimoto coefficient, TC = 1.0), whereas the original eMolFrag→ eSynth workflow reconstructed 51 of the 100 molecules in a benchmark subset. The upgraded workflow also produced broader recombinant chemical spaces, including higher entropy in a four-molecule diversity analysis (2.78 versus 2.70). Both tools are open-source, platform-agnostic, and available for academic and commercial use.
- 13 October 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (20) , 13023-13031
Open Accesse MolFrag 2 and e Synth 2: Sound Fragment-Based Molecular Decomposition and Reconstruction
We present eMolFrag2 and eSynth2, enhanced successors to eMolFrag and eSynth for fragment-based molecular decomposition and reconstruction. eMolFrag2 is a complete reimplementation of eMolFrag in Python, using RDKit and graph-based molecular representations to preserve connectivity, identify free atoms, combine adjacent linkers, and avoid aromaticity ambiguities that impair downstream reconstruction. eSynth2 generalizes fragment recombination by removing hard-coded molecular-property filters, supporting multiple bonds to a single brick atom, and adding a targeted reconstruction mode for soundness validation. Across 36,534 nonredundant DUD-E active compounds, the eMolFrag2 → eSynth2 workflow reconstructed every input molecule exactly (Tanimoto coefficient, TC = 1.0), whereas the original eMolFrag→ eSynth workflow reconstructed 51 of the 100 molecules in a benchmark subset. The upgraded workflow also produced broader recombinant chemical spaces, including higher entropy in a four-molecule diversity analysis (2.78 versus 2.70). Both tools are open-source, platform-agnostic, and available for academic and commercial use.
- Open Access
- 27 March 2026
- Published by Oxford University Press (OUP) in NAR Genomics and Bioinformatics
- Vol. 8 (2)
A multimodal human protein embeddings database: DeepDrug Protein Embeddings Bank (DPEB)
Computationally predicting protein–protein interactions (PPIs) is challenging due to the lack of integrated, multimodal protein representations. DeepDrug Protein Embeddings Bank (DPEB) is a curated collection of 22 043 human proteins that integrates four embedding types: structural (AlphaFold2), transformer-based sequence (BioEmbeddings), contextual amino acid patterns (ESM-2: Evolutionary Scale Modeling), and sequence-based n-gram statistics (ProtVec). AlphaFold2 protein structures are available through public databases (e.g. AlphaFold2 Protein Structure Database), but the internal neural network embeddings are not. DPEB addresses this gap by providing AlphaFold2-derived embeddings for computational modeling. Our benchmark evaluations show GraphSAGE with BioEmbedding achieved the highest PPI prediction performance (87.37% AUROC, 79.16% accuracy). The framework also achieved 77.42% accuracy for enzyme classification and 86.04% accuracy for protein family classification. DPEB supports multiple graph neural network methods for PPI prediction, enabling applications in systems biology, drug target identification, pathway analysis, and disease mechanism studies.
- 27 March 2026
- Published by Oxford University Press (OUP) in NAR Genomics and Bioinformatics
- Vol. 8 (2)
Open AccessA multimodal human protein embeddings database: DeepDrug Protein Embeddings Bank (DPEB)
Computationally predicting protein–protein interactions (PPIs) is challenging due to the lack of integrated, multimodal protein representations. DeepDrug Protein Embeddings Bank (DPEB) is a curated collection of 22 043 human proteins that integrates four embedding types: structural (AlphaFold2), transformer-based sequence (BioEmbeddings), contextual amino acid patterns (ESM-2: Evolutionary Scale Modeling), and sequence-based n-gram statistics (ProtVec). AlphaFold2 protein structures are available through public databases (e.g. AlphaFold2 Protein Structure Database), but the internal neural network embeddings are not. DPEB addresses this gap by providing AlphaFold2-derived embeddings for computational modeling. Our benchmark evaluations show GraphSAGE with BioEmbedding achieved the highest PPI prediction performance (87.37% AUROC, 79.16% accuracy). The framework also achieved 77.42% accuracy for enzyme classification and 86.04% accuracy for protein family classification. DPEB supports multiple graph neural network methods for PPI prediction, enabling applications in systems biology, drug target identification, pathway analysis, and disease mechanism studies.
- Open Access
- 8 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-9
KG-DTA: A knowledge graph-based meta-path learning framework to predict drug-target binding affinity
- 8 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-9
Open AccessKG-DTA: A knowledge graph-based meta-path learning framework to predict drug-target binding affinity