Overview
Mol2CCS supplies graph-neural-network training and prediction code used to study CCS generalization across datasets. Models operate on molecular structures, adducts and optional three-dimensional coordinates, with prepared datasets and evaluation notebooks.
What can this tool do?
Common uses
- Predict CCS from molecular structure for research and benchmarking.
Scientific fit
Primary function
Machine learning and prediction
Other functions
Metabolite annotation and identification
MS level
MS1
Acquisition
General or unspecified acquisition
Ion mobility
Ion mobility supported
Data compatibility
Inputs
Molecular structures
Outputs
Statistical results
Input formats
Prepared Parquet molecular/adduct tables with optional 3D coordinates
Output formats
CCS predictions, HDF5 model files and evaluation artifacts
Scope and considerations
Designed for
- A research codebase for graph-neural-network prediction of ion-mobility collision cross sections from molecular structure, developed to evaluate model generalizability across chemical domains.
Not designed for
- Calibrating experimental mobility measurements or guaranteeing accurate CCS for chemical domains absent from model training.
Important requirements
- The documented Python/Poetry environment, compatible molecular/adduct data and a trained model; reproducing the experiments additionally requires downloading and preparing the licensed METLIN-CCS/CCSBase datasets.
Considerations
- Dataset splits, adduct definitions and structural domain affect evaluation. The repository is a research framework; do not assume a maintained universal pretrained prediction service.
Access and documentation
MIT source distribution. External datasets, dependent packages and any hosted routes retain their own terms; software operation was not tested.
License: MIT
Evidence and review
Enveda intake rechecked against official repository documentation, package/license metadata and primary publication identity on 2026-10-27. Editorial source review only; no software execution or independent benchmark. See docs/enveda-expansion-review-2026-10-27-batch2.md.
Core publications
Resource papers, direct updates and tool-specific methods. Selection policy.
- Evaluating the generalizability of graph neural networks for predicting collision cross section. Journal of Cheminformatics (2024). doi:10.1186/s13321-024-00899-w
Related tools
Entry history
Changes to this catalogue entry, not the tool's software release history. Acceptance and editorial review are separate.
- 2026-10-27 · First committed entry
Add 24 reviewed Enveda tools in a separate second batch
View change · Entry at this revision