Proteome informatics articles from across Nature Portfolio

eSIG-Net is an interaction language model that predicts the effects of mutations on protein interaction.

Peptide-spectrum match (PSM) rescoring is crucial for accurate peptide identification in DDA proteomics. Here, the authors introduce DDA-BERT, an end-to-end transformer-based deep learning model that improves peptide identification across multiple species and HLA immunopeptidomics.

This study introduces ProteoCast, which uses protein sequence conservation to predict mutation effects proteome-wide. Validated on 400,000 Drosophila variants and genome editing experiments, it accurately identifies functionally critical sites in any organism.

DIAMOND DeepClust provides an ultra-fast clustering method for organizing the protein universe of life at low sequence identity, enabling large-scale dimensionality reduction and improving downstream structure prediction with AlphaFold2.

An integrated mass spectrometry platform enabling automated collision-, electron- and photon-based fragmentation techniques is presented. A deep learning-based model trained to predict spectra across all dissociation methods further enhances protein identification.

By analyzing the levels of thousands of proteins in blood samples from approximately 50,000 people, researchers estimate how biologically ‘old’ individual organs appear. When estimated organ age is older than a person’s actual age, risk of disease and earlier death increase, whereas a youthful brain and immune system signal a longer, healthier life.

The performance of omics prediction models can be significantly improved by combining limited patient proteomic data with widely available electronic health records.

Experimental and computational studies are paving way for a deeper understanding of the dynamic nature of protein–protein interactions.

Mass spectrometry-based proteomics provides broad and quantitative detection of the proteome, but its results are mostly presented as protein lists. Artificial intelligence approaches will exploit prior knowledge from literature and harmonize fragmented datasets to enable mechanistic and functional interpretation of proteomics experiments.

To handle increasingly large protein databases, a new ultrafast, highly sensitive method — Dense Homolog Retriever (DHR) — detects remote homologs using dense retrieval and protein language models. Its alignment-free nature makes it much faster than traditional approaches, and the newly found remote homologs benefit our understanding of protein evolution, structure and function.