Liviu Copoiu

AbBFN2 is a generative antibody foundation model that unifies diverse data sources and enables flexible conditional generation for multiple antibody design tasks.

This paper introduces ProtBFN, a Bayesian Flow Network for protein sequence generation, which outperforms existing models and is fine-tuned for antibody generation.

This study evaluates genomic language models on protein tasks, showing they are competitive with proteomic models and benefit from joint genomic-proteomic training and optimized tokenization.

The paper introduces BioCLIP, a contrastive learning framework for pre-training Protein Structure Models using Protein Language Models, achieving performance comparable to specialized methods on various tasks.

LightMHC predicts full-atom pMHC structures from sequences using a lightweight model with attention mechanisms, achieving comparable performance to AlphaFold 2 and ESMFold with faster speed.

The paper introduces a graph neural network for peptide-MHC structure prediction, achieving accuracy comparable to AlphaFold 2 with significantly fewer parameters.

The paper identifies potential structural mimicry of coronavirus spike receptor-binding motifs by human and non-human proteins using structural bioinformatics and docking simulations.