Deep3DCCS: Geometry-Driven Collision Cross Section Prediction from Multi-View Molecular Projection Tensors
Issued Date
2026-09-28
Resource Type
ISSN
15499596
eISSN
1549960X
Scopus ID
2-s2.0-105051764640
Journal Title
Journal of Chemical Information and Modeling
Volume
66
Issue
18
Start Page
11619
End Page
11630
Rights Holder(s)
SCOPUS
Bibliographic Citation
Journal of Chemical Information and Modeling Vol.66 No.18 (2026) , 11619-11630
Suggested Citation
Wisanpitayakorn P., Mathema V.B., Wanichthanarak K., Keawhawong T., Jariyasopit N., Saiviroonporn P., Tongdee T., Sirivatanauksorn Y., Khoomrung S. Deep3DCCS: Geometry-Driven Collision Cross Section Prediction from Multi-View Molecular Projection Tensors. Journal of Chemical Information and Modeling Vol.66 No.18 (2026) , 11619-11630. 11630. doi:10.1021/acs.jcim.6c00371 Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/119385
Title
Deep3DCCS: Geometry-Driven Collision Cross Section Prediction from Multi-View Molecular Projection Tensors
Author's Affiliation
Corresponding Author(s)
Other Contributor(s)
Abstract
Collision cross section (CCS) measurements from ion mobility–mass spectrometry improve confidence in molecular identification in metabolomics and lipidomics and reduce false-positive annotations in untargeted analyses. However, experimental CCS coverage remains incomplete, and accurate computational prediction remains needed. Here, we present Deep3DCCS, a geometry-driven deep learning framework that predicts CCS from stacked multi-angle molecular projection tensors using a three-dimensional convolutional neural network operating across the projection-view and two spatial image dimensions. Three-dimensional molecular structures generated from SMILES strings were rotated around the x, y, and z-axes, converted into binary two-dimensional projections, and assembled into standardized multi-view tensors for CCS regression. Deep3DCCS was developed and evaluated using curated AllCCS and METLIN lipid CCS datasets containing 3948 molecule–adduct entries across [M – H]<sup>−</sup>, [M + H]<sup>+</sup>, and [M + Na]<sup>+</sup> ions, and input optimization identified 32 × 32 pixel projections with five rotations per axis as a practical balance between prediction accuracy, model stability, and computational feasibility. In independent true-validation experiments, AllCCS-trained models achieved MRPE values of 3.27–3.79% on AllCCS validation data and 2.93–3.72% on METLIN validation data, whereas METLIN-trained models transferred less effectively to the broader AllCCS chemical space. Overall, these results support multi-view molecular projection tensors as a viable geometry-driven representation for CCS prediction and demonstrate that substantial CCS-relevant information can be learned directly from three-dimensional molecular morphology using a 3DCNN architecture, with low single-digit prediction errors across independent and cross-dataset validation experiments.