Santiago Segarra

GRNITE improves gene regulatory network inference by combining text embeddings and graph neural networks, enhancing AUROC and recall across methods.

This paper introduces rhea, a reference-free method for detecting structural variants in microbiomes using long-read coassembly graphs, showing improved performance over existing methods in simulated and real metagenomic data.

This paper introduces ETNA, a joint embedding method for cross-species functional alignment using biological networks and sequence orthologs, showing improved preservation of gene functional relationships.

SeqScreen uses ensemble learning to accurately label pathogenic sequences with functional annotations, achieving high recall and precision for microbial pathogenesis.

KOMB identifies repetitive and homologous genomic regions in metagenomes using K-core graph decomposition, validated on simulated and real datasets.