Insuk Lee

LoGoPPI infers protein-protein interactions from sequence using a combination of global and local representations, achieving faster inference than state-of-the-art models.

The paper introduces MycobiomeDB, a fungal genome catalog, and site-specific MycoProfilers that enhance mycobiome analysis by improving taxonomic recall and distinguishing disease-associated mycobiomes from healthy controls.

The paper presents an expanded gut microbiome genome catalogue with improved quality and geographic diversity, enabling accurate metabolic modeling of the human gut microbiota.

The paper presents a high-quality genomic catalog of the human oral microbiome, revealing novel species and highlighting its clinical relevance to systemic diseases.

The paper develops scNET to infer cell-type-specific co-expression networks from single-cell data, enhancing the human interactome for improved disease gene prediction.

The paper uses single-cell multi-omics to identify immunosuppressive factors in ALK-positive lung cancer that contribute to poor immunotherapy responses.

The paper introduces MRGM, a comprehensive catalog of mouse gut microbial genomes that expands taxonomic and functional diversity, improves gene annotation, and reveals diet-driven differences between human and mouse gut microbiota.

HCNetlas links disease genes to cell types by analyzing cell type-specific gene networks in healthy tissues and comparing them to disease-affected samples.

The paper introduces GMWI2, a gut microbiome health index that improves disease prediction accuracy compared to previous models and traditional diversity indices.

This study uses single-cell sequencing to identify cellular and molecular differences in the kidneys and pancreases of insectivorous and frugivorous bats, linking them to diabetes-related mechanisms.

scHumanNet identifies cell-type-specific gene functions by analyzing single-cell transcriptomes, revealing disease-related gene effects in specific cell types.

NGSEA improves gene set enrichment analysis by incorporating functional networks, outperforming GSEA in pathway identification and drug repositioning.

The paper constructs WheatNet, a genome-scale functional network for hexaploid bread wheat, integrating 20 genomics datasets and 156,000 co-expression links.

The paper shows that network topologies in real-world systems are shaped by interaction modes, revealing differences between pairwise and group interactions.