Mayo, S. | Search bioRxiv

bioRxiv2025

B-PPI: A Cross-Attention Model for Large-Scale Bacterial Protein-Protein Interaction Prediction

Protein-protein interactions (PPIs) are essential for the study of cellular function, yet computational prediction of bacterial PPIs remains limited. Most existing methods are trained on human data, reducing their applicability to bacterial systems. Here, we present B-PPI, a computational tool specifically designed for bacterial PPI prediction. B-PPI leverages embeddings from ProstT5, a structure-aware protein language model, and a cross-attention mechanism to capture residue-level inter-protein relationships. To facilitate training, we constructed B-PPI-DB, a large-scale bacterial PPI dataset derived from STRING, comprising 202,829 positive and negative interactions across 2,646 taxa with a 1:10 positive-to-negative ratio. We benchmarked B-PPI against TT3D, a state-of-the-art model trained on human PPI, which was previously evaluated on bacterial PPIs. B-PPI achieved substantially higher performance on bacterial data (AUPRC 0.926{+/-}0.006 vs. 0.230{+/-}0.005 and F1 0.866{+/-}0.007 vs. 0.299{+/-}0.005) with faster runtime. We further demonstrate that the model adapts to unseen bacterial interactions with minimal fine-tuning. Together, B-PPI and B-PPI-DB address a critical gap in computational microbiology, offering a framework for bacterial PPI prediction and a data resource for benchmarking and developing new tools in the field.

bioinformatics↗

bioRxiv2023

CDRxAbs: Antibody Small-Molecule Conjugates with Computationally Designed Target-Binding Synergy

Antibody-drug conjugates (ADCs) combine the advantages and offset the disadvantages of their constituent parts to achieve a refined spectrum of action. We combine the concept of ADCs with the full atomic simulation capability of computational protein design to define a new class of molecular recognition agents: CDR-extended antibodies, abbreviated as CDRxAbs. A CDRxAb incorporates a covalently attached small molecule into an antibody/target binding interface using computational protein design to create an antibody small-molecule conjugate that binds tighter to the target of the small molecule than the small molecule would alone. CDRxAbs are also expected to increase the target binding specificity of their associated small molecules. In a proof-of-concept study using monomeric streptavidin/biotin pairs at either a nanomolar or micromolar-level initial affinity, we designed nanobody-biotin conjugates that exhibited >20-fold affinity improvement against their protein targets with step-wise optimization of binding kinetics and overall protein stability. The workflow explored through this process promises a novel approach to optimize small-molecule based therapeutics and to explore new chemical and target space for molecular-recognition agents in general. SignificanceWe defined a general method for optimizing molecular recognition reagents that involve small molecules and demonstrated an application of this method using a model system. Instead of using traditional approaches for modifying a small molecule to improve its binding properties, we use computational protein design to build an antibody/small molecule conjugate that allows the target-binding strength (and specificity) of the small molecule to be tuned through changes in the amino acid sequence of the antibody scaffold. This method introduces a novel approach for optimizing the binding properties of small molecules and expands the potential application scenarios for antibody-drug conjugates.

bioengineering↗