PeptidomicsDB: a unified resource of endogenous peptidomes for peptide discovery and functional exploration across species and biological contexts
peer-reviewed · Nucleic Acids Research · 2026
| Date | 2026-10-23 |
| Type | peer-reviewed |
| Venue | Nucleic Acids Research |
| Publisher | Oxford University Press (OUP) |
| Contribution | downstream-application |
| DOI | 10.1093/nar/gkag1032 |
| Venue 2-year citedness | 16.42 |
Abstract
Endogenous peptides represent a rich source of bioactive molecules that regulate diverse physiological processes and offer opportunities for functional discovery. Mass spectrometry-based peptidomics enables their systematic characterization, yet public datasets remain fragmented and are often processed using heterogeneous tools and parameters, limiting peptide-level retrieval, cross-study exploration, and data reuse. We developed PeptidomicsDB (freely accessible at http://www.inbirg.com/ppd/), a unified resource based on the standardized reanalysis of 151 public peptidomics projects comprising 3992 biological samples from major branches, including global peptidomics, neuropeptidomics, and immunopeptidomics, across species and biological contexts. Using the DeepNovo Peptidome workflow implemented in PEAKS Studio 13.0, which integrates database searching, homology searching, and de novo sequencing, we identified >33 million peptide records representing over 4 million unique peptide sequences, of which >400 000 showed sequence similarity to bioactive peptides curated in public repositories. Analysis of 249 curated pairwise comparison datasets identified 78 910 differentially abundant peptide records, while 14 context-specific datasets yielded 148 175 peptide records enriched in individual tissues, diseases, or cell types. PeptidomicsDB links peptide, precursor-protein, sample, and dataset levels with experimental provenance, quantitative evidence, mass spectrometric data, and functional annotations. Together, these capabilities support peptide discovery and functional exploration.
Methods and tools
- PeptidomicsDB: A database of endogenous peptides built by reanalysing 151 public peptidomics projects (3992 samples; global, neuro- and immunopeptidomics, DDA and DIA) with PEAKS Studio 13’s DeepNovo Peptidome workflow, which combines database, DeepNovo de novo and SPIDER homology search: over 4 million unique peptides, linked to precursor proteins, samples, differential abundance and bioactive-peptide similarity.
Methods it uses
- DeepNovo Peptidome: Immunopeptidomics pipeline
- PEAKS: Commercial DP-based de novo
- SPIDER: Protein identification from sequence tags while accounting for de novo sequencing error.