MassNet
A Large-scale Al-ready Proteomics Copous
MassNet is an AI-ready proteomics corpus comprising 1.54 billion spectra from ~28,000 DDA-MS files across 648 projects, covering 35 species and 10 Mass spectrometers types and providing a scalable foundation for proteomics and AI research.
Raw Files27,643
Projects648
Species35
Instruments11
PSMs548M
Peptides4.61M
Proteins2.54M
Overview
Instruments
Q Exactive HF46.2%
Orbitrap Fusion Lumos18.7%
Q Exactive HF-X10.9%
Orbitrap Exploris 4807.5%
Others16.8%
All in-house DDA runs were grouped as a single project.
Species Distribution
35Species
Homo sapiens56.6%
Mus musculus24.7%
Saccharomyces cerevisiae6.6%
Bacillus subtilis2.1%
Arabidopsis thaliana1.5%
Others8.5%
Project File Count (Top 10)
In-house experiments3,379
PXD0309832,170
IPX0001400000983
PXD019643742
IPX0005714000552
PXD028028472
PXD038906355
PXD028136347
PXD023118344
PXD022625331
Top 10 projects ranked by raw file count.
PSM Distribution Across Species (Top 10)
Homo sapiens424,207,738
Mus musculus287,649,421
Saccharomyces cerevisiae36,947,858
Arabidopsis thaliana11,467,419
Escherichia coli9,541,701
Bacillus subtilis6,675,807
Rattus norvegicus6,100,756
Caenorhabditis elegans2,510,124
Drosophila melanogaster1,482,536
Staphylococcus aureus1,358,308
Run-level 1% FDR PSM counts summed from FragPipe.