---
pretty_name: UltraMS SpectraVerse dataset (molecule-level 8:1:1 split)
license: cc-by-4.0
language:
  - en
size_categories:
  - 100K<n<1M
task_categories:
  - other
configs:
  - config_name: spectraverse_split
    data_files: splits/spectraverse_compound_split.csv
  - config_name: spectraverse_spectra
    data_files: Spectraverse.csv
---

# UltraMS SpectraVerse dataset

A frozen, molecule-disjoint conversion of the SpectraVerse tandem mass spectral library,
released by UltraMS. **This is a data resource, not a standalone benchmark.** The manuscript
uses it as the spectrum source of the negative-ion structure-family evaluation (Figure 2c;
661 molecule-spectrum pairs after negative-ion selection and MoNA exclusion) and as one of
the three structure-annotated sources (with MassSpecGym and MSnLib) for the contrastively
enhanced UltraMS training, chemical-class evaluation and UltraAtlas reference spectra.

## Contents

| file | description |
| --- | --- |
| `Spectraverse.csv` | 464,346 spectra; columns `mzs, intensities, smiles, precursor_mz, adduct, collision_energy, fold`; 707,447,509 B, SHA-256 `06e084501e91f92cc642c3d60de2863266ff3d0db958354a011e32851bef3976` |
| `splits/spectraverse_compound_split.csv` | 42,308 exact SMILES → `fold` (train / validation / test) with `n_spectra` |
| `README.md`, `NOTICE.md`, `VERIFY.md`, `stats.json`, `SHA256SUMS` | documentation, attribution, verification and checksums |

## Split

| fold | spectra | unique exact SMILES |
| --- | ---: | ---: |
| train | 370,362 | 33,846 |
| validation | 47,825 | 4,231 |
| test | 46,159 | 4,231 |

The split unit is the exact SMILES string: every spectrum of a SMILES string lies in one
fold and no string crosses folds (molecule/SMILES-disjoint 8:1:1). It is **not**
formula-disjoint: structural isomers share a formula and formulas recur across folds.

## Usage

```python
import pandas as pd
base = "https://huggingface.co/datasets/dsadd4/UltraMS-Spectraverse-Dataset/resolve/v1.0.0/"
spectra = pd.read_csv(base + "Spectraverse.csv")
split = pd.read_csv(base + "splits/spectraverse_compound_split.csv")
test = spectra[spectra.fold == "test"]
```

## Provenance and licence

The spectra are the public SpectraVerse library (SpectraVerse v1.0.1, Zenodo
doi:10.5281/zenodo.17870921, CC BY 4.0; Gupta, Qiang, Chung, Herbst and Skinnider).
UltraMS contributed the CSV representation, the frozen molecule-level split and the release.
Released under CC BY 4.0; attribution to the SpectraVerse authors is required.