# Prosit lac 2025 Fragment Ion Intensity (MS2) Training Data: Lactylated and Unmodified Peptide Spectra from PROSPECT and ProteomeTools

## Facts

- Provider: Wilhelmlab, Technical University of Munich
- License: CC-BY-4.0 (https://huggingface.co/datasets/Wilhelmlab/Prosit-2025-lac-ms2)
- Size: 65 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Wilhelmlab/Prosit-2025-lac-ms2
- Access: public, readable without an account
- Category: Proteomics
- Source: https://huggingface.co/datasets/Wilhelmlab/Prosit-2025-lac-ms2
- Page: https://scigantic.com/archive/5cf0c649-61ba-5108-ae5a-49298cbd8f0a
- Keywords: Proteomics, spectroscopy, Parquet, Hugging Face

## About

Annotated tandem mass spectra prepared by the Wilhelm lab (Technical University of Munich) to train Prosit-style deep learning models that predict fragment ion intensities from peptide sequences, extended to lysine lactylation (Klac). The data combine the PROSPECT dataset with spectra of a Klac peptide library, with upstream data from the ProteomeTools synthetic peptide datasets on PRIDE. Fragment annotation follows the expert-system rules in the PROSPECT paper, and the raw intensity vector of each spectrum is stored in the intensities_raw column. Train, validation and test splits are spread over 63 Parquet files (about 2.2 GB) and load with datasets.load_dataset('Wilhelmlab/Prosit-2025-lac-ms2'). Supports MS2 spectrum prediction for unmodified and lactylated peptides. Licensed CC BY 4.0 by Wilhelmlab, TUM; cite PROSPECT (Shouman et al., NeurIPS 2022) and ProteomeTools (Zolg et al., Nature Methods 2017).

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("Wilhelmlab/Prosit-2025-lac-ms2")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `data/test-00000-of-00007.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/5cf0c649-61ba-5108-ae5a-49298cbd8f0a
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/5cf0c649-61ba-5108-ae5a-49298cbd8f0a
- Over MCP: `get_schema_card` with id `5cf0c649-61ba-5108-ae5a-49298cbd8f0a` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Spectroscopy datasets](https://scigantic.com/topics/spectroscopy)
- [Proteomics datasets](https://scigantic.com/topics/proteomics)
