Pedro Beltrão

The paper identifies a nucleolar module involving Fap7 that ensures proper uS11 modification and incorporation into pre-ribosomes, safeguarding ribosome biogenesis.

The paper presents a fast BINN implementation trained on UK Biobank data to predict disease, revealing limitations in interpretability and showing that model multiplicity can reveal disease heterogeneity.

The paper identifies phosphorylation sites that regulate protein interactions using pan-cancer data and AlphaFold predictions, revealing 6,160 associations between phosphosites and interacting protein abundance.

This paper evaluates AlphaFold3's performance across diverse structural biology tasks, finding uneven accuracy and strong dependence on training data.

The paper reports the structure of the mammalian ciliary transition zone microtubule doublet using cryo-electron tomography, revealing unique proteins and their roles in microtubule stabilization and transport regulation.

The paper develops predictors to functionally characterize human phosphosites using integrated data, revealing novel functions and connections to DNA repair and cell survival.

This paper develops a deep-learning method to unify genetic evidence from common and rare diseases by integrating diverse data sources and identifying shared genetic mechanisms.

The paper predicts protein-protein interactions in Mycoplasma genitalium using a pooled-AlphaFold3 approach, improving accuracy and efficiency over traditional methods.

This study systematically maps metal-binding sites in E. coli using Metal-LiP, identifies 424 candidate metal-binding proteins, and suggests functional roles for metal binding.

The paper analyzes how phosphorylation affects protein structure and dynamics using structural data from the PDB, finding that it often induces stabilizing conformational changes and modulates residue fluctuations.

The paper maps the functional importance of site-specific ubiquitination across the human proteome by analyzing conservation, perturbation, and machine learning scores.

This paper introduces UbyW, a platform for site-specific ubiquitylation using UBE2W, enabling efficient production of ubiquitylated proteins and identifying USP15 as a deubiquitylase for Ran K71 monoubiquitylation.

The paper uses AlphaFold to predict structures of host-pathogen PPIs, revealing molecular mimicry modes and convergent evolution of interaction interfaces.

This study uses network propagation to improve the clustering of primary ciliopathies and identify candidate genes, including a novel variant in CEP43.

This study introduces PulsPhos, a method to quantify site-specific phosphorylation lifetimes in living cells and applies it to pharmacological perturbations.

The paper predicts the mechanistic impacts of human protein missense variants on stability, protein-protein interactions, and small-molecule binding using AlphaFold2 structures.

The paper constructs a tissue-specific protein interaction atlas using proteomic data from 11 human tissues, identifying key drivers of inter-tissue variation and prioritizing disease genes in brain disorders.

This study uses machine learning to predict yeast phenotypes from genotypes, finding gradient boosting machines perform best and accessory genome contributes significantly.

This study systematically identifies structure-specific protein-protein interactions using LiP-MS, demonstrating its application to alpha-synuclein in Parkinson's disease.

The paper constructs a PPI network of 100 high-confidence ASD genes, identifies novel interactions, and uses AlphaFold-Multimer to prioritize variants for functional validation in organoids.

A Foldseek-based linear-time structural clustering groups 52 million AlphaFold DB representatives into 2.27 million non-singleton clusters; 31% lack known structural or domain annotation.

This paper introduces Sequence UNET, a deep learning model for variant effect prediction that is highly scalable and computationally efficient, achieving comparable performance to existing methods on ClinVar data.

The paper uses network-based expansion of GWAS results to identify pleiotropic gene modules linked to multiple traits and demonstrates their application in studying inflammatory bowel disease.

The paper uses AlphaFold2 to predict structures for 65,484 human protein interactions, identifying high confidence models and disease-related mutations.

The paper systematically analyzes proteomic and genomic data of SARS-CoV-2 variants to identify how mutations affect viral gene expression, host interactions, and immune evasion.

The paper defines a mutational landscape of amino acid functions using deep mutational scanning data and identifies 100 functional amino acid subtypes.

The paper generates and provides computational predictions of the effects of all possible SARS-CoV-2 missense variants to support variant analysis and experimental validation.

Meta-analysing UK Biobank family-history GWAS with diagnosed-AD GWAS yields 37 loci, four novel (NCK2, TSPAN14, SPRED2, CCDC6); fine-mapping, colocalisation and network scoring prioritize causal genes.

COSMOS integrates multi-omics data with prior knowledge to generate mechanistic hypotheses about biological processes in renal cell carcinoma.

This study identifies an FGF2-mediated incoherent feedforward loop that inhibits Erk and promotes naive pluripotency in ES cells and embryos.