Alberto Santos

The paper uses graph neural networks to interpret microbial association patterns in IBD, showing that coordinated associations, not just abundance shifts, are predictive.

This paper introduces InstaNovo-FM, a self-supervised foundation model for proteomics that learns from tandem mass spectra to enable transferable applications across datasets and instruments.

The paper presents a Xenopus oocyte-based platform to discover transporter substrates using untargeted metabolomics, identifying known and novel substrates for five human SLC transporters.

The paper uses single-cell metagenomics to link plasmid-derived antimicrobial resistance to host bacteria in pigs, identifying resistance-associated features in longitudinal gut microbiome data.

This study analyzes 1432 transcriptomics datasets from 132 Actinomycetota strains to identify patterns of BGC expression and regulatory mechanisms.

ABaCo integrates metagenomic data from multiple studies, corrects technical heterogeneity, and outperforms existing methods while preserving biological signals.

VueGen automates report generation from bioinformatics outputs, enabling non-experts to create reports with minimal coding experience.

PanKB is a microbial pangenome knowledgebase that integrates pangenomic data and literature to support research and biotechnological applications.

The paper quantifies the human liver proteome at cell-type resolution across multiple conditions, revealing proteome changes in disease states and providing a web-based resource for exploration.

The paper uses in vivo pulsed SILAC to measure protein incorporation in mice tissues, finding reduced turnover and molecular changes linked to aging.

This study uses proteomics to analyze dynamic phospho-signaling networks in human erythropoiesis, identifying key regulatory mechanisms and markers across differentiation stages.

The Clinical Knowledge Graph (CKG) links over 16 million nodes and 220 million relationships with statistical and machine learning workflows to accelerate proteomics data analysis and clinical biomarker interpretation.

The paper compares cardiac protein profiles across species to identify suitable models for human cardiac diseases.

The paper develops an orthology-based computational method to predict host-parasite protein-protein interaction networks for 12 human parasites, identifying common pathogenic mechanisms.

This study compares large-scale tissue expression datasets and finds they agree well, supporting tissue-specific and ubiquitous expression distinctions, and introduces a resource for integrated data access.