Marcelo D. T. Torres

OmegAMP generates antimicrobial peptides with controlled physicochemical properties and safety profiles, showing activity against drug-resistant bacteria in vitro and in vivo.

ApexFold predicts peptide secondary structure changes across environments using sequence and physicochemical descriptors, outperforming static predictors.

The paper identifies antimicrobial peptides encoded in ancient human mitochondrial genomes and validates their activity against bacteria.

This paper introduces APEXGO, a generative AI framework for optimizing antimicrobial peptides, achieving high success rates in enhancing antimicrobial activity and outperforming existing methods.

This study introduces Termini, an AI framework for designing antimicrobial peptides with terminal modifications, achieving high hit rates and therapeutic potential in vitro and in vivo.

This paper introduces the EDEN family of foundation models for AI-programmable therapeutic design, demonstrating their ability to generate functional gene insertions, antibiotic peptides, and synthetic microbiomes.

This paper investigates the anti-biofilm mechanism of bac7 (1-35) against Klebsiella pneumoniae, showing it disrupts bacterial membranes and ribosomal function, leading to biofilm collapse and reduced dissemination.

This study evaluates four peptide backbone modifications to enhance stability while preserving function, finding that azapeptides maintain activity and offer improved stability.

This paper introduces a deep reinforcement learning framework for designing new antibiotics, producing potent peptides with broad-spectrum activity against bacteria.

This study introduces AMP-Diffusion, a generative model for designing antimicrobial peptides, which produces candidates with broad-spectrum antibacterial activity and low toxicity.

This study uses deep learning to analyze archaeal proteomes, identifying 12,623 potential antimicrobial molecules, with 93% showing in vitro activity and one candidate effective in vivo against bacterial infections.

This study identifies 53 potent antimicrobial venom-derived peptides using computational methods and experimental validation, showing promise for combating antibiotic-resistant infections.

The paper uses deep learning to identify novel antimicrobial peptides from proteasomal cleavage products, finding 21 sequence-diverse peptides with potential therapeutic applications against drug-resistant pathogens.

The paper improves functional protein generation by optimizing likelihood in both amino acid sequence and latent space, outperforming existing models in AMP and MDH generation.

The paper introduces KCM, an optimization framework for protein and peptide design that outperforms generative models by enabling precise backbone geometry control with minimal computational resources.

This paper introduces ApexDuo, a deep learning model that generates peptides with both cell-penetrating and antimicrobial properties, and demonstrates its effectiveness against intracellular pathogens.

This study tests peptides from non-immune proteins for antimicrobial and immunomodulatory effects, finding potent activity against bacteria and immunomodulation in vitro and in mouse models.

This study identifies two quorum-sensing systems in E. faecalis that regulate purine biosynthesis, enhancing its pathogenicity in macrophages and murine infections.

A uniformly processed set of 2,440,377 bacterial and archaeal genomes with search tools and annotations, plus an AI-mined peptide set yielding ATB20, active in a mouse skin infection model.

This paper develops a method to engineer phage tail fibers to alter host range and suppress bacterial resistance, showing effectiveness in vitro and in vivo.