Mateus Bernabe Fiamenghi

Minerva extracts RNA base-pairing, protein-contact and repeat maps from a genome language model without alignments, flagging unannotated structured loci and a phage ncRNA array reverse-transcribed into short cDNAs.

The paper identifies novel enzybiotic candidates targeting human bacterial pathogens by analyzing viral-host associations in a large viral genome database.

VPF-Class 2.0 improves viral classification by using a supervised model on marker-composition features, achieving high accuracy and interpretability.

The paper presents a global virome atlas, UHGV, and identifies genomic factors influencing phage-host interactions.

Assembling the Sequence Read Archive (27.3 million accessions) with over 100-fold compression enables minutes-scale k-mer search and mining of enzymes, plasmids, proteins and viral reactivation.