Predicting Collision Cross Sections with GRACE: Geometric Residual Adduct Conditioning via Early-fusion
Abstract
Collision cross section (CCS), derived from ion mobility mass spectrometry, is a common descriptor for molecular annotation. Prediction is challenging for machine learning models because it reflects the size, shape, and ionization state of a gas-phase molecular ion. Most predictors either ignore explicit 3D structure or treat adduct identity as a late categorical feature, which limits their ability to capture adduct-dependent geometric effects. We present GRACE (Geometric Residual Adduct Conditioning via Early-fusion), a 3D CCS predictor that adapts a pretrained molecular geometry encoder using geometric residual adduct conditioning via early fusion. GRACE combines two inductive biases: a residual objective relative to an adduct-aware physical descriptor baseline and adduct conditioning within the encoder via a learned adduct token and low-rank attention adapters. We evaluate the model on a curated set of over 9,000 experimental molecule-adduct CCS records with random, scaffold, and adduct-sensitive splits designed to separate interpolation, scaffold generalization, and adduct-driven generalization. GRACE achieves the best mean percentage difference among the evaluated learned models on all three splits: 1.67% on the random split, 2.11% on the scaffold split, and 2.36% on the adduct-sensitive split. Diagnostic analyses suggest that residual learning stabilizes training by removing the dominant mass-CCS trend, while early fusion improves adduct-sensitive prediction relative to late fusion. Across four independent external test sets, GRACE shows consistently lower error than the other evaluated models. On a held-out set, GRACE also attains the lowest mean percent difference when compared with four previously reported physics-based workflows. These results support residual learning and encoder-level adduct conditioning as practical inductive biases for fast, accurate CCS prediction.
Related
- Predicting Capsid Protein Binding Sites in Single-Stranded RNA Viruses Using Machine Learning from Local Geometric Features
- Improving adhesion and targeting of adherent areas for high-throughput cryo electron tomography of vitreous sections
- Visualizing Cell Type–Specific Glycan Profiles in Tissue Sections With Glycan Painting
- Molecular Architecture of the Human GCN1-ABCF3 Ribosome Collision Sensor
- MD-LLM-2: A Transferable Language Model of Molecular Dynamics with Physical Conditioning and Explicit Path Probabilities
BibTeX
@misc{suryanarayanan2026predicting,
title = {{Predicting Collision Cross Sections with GRACE: Geometric Residual Adduct Conditioning via Early-fusion}},
author = {Parthasarathy Suryanarayanan and Susanta Das and Shreyans Sethi and Kenneth M. Merz, Jr. and Joseph A. Morrone},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2609.12223},
eprint = {2609.12223},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.12223},
}
RIS
TY - UNPB
TI - Predicting Collision Cross Sections with GRACE: Geometric Residual Adduct Conditioning via Early-fusion
AU - Parthasarathy Suryanarayanan
AU - Susanta Das
AU - Shreyans Sethi
AU - Kenneth M. Merz, Jr.
AU - Joseph A. Morrone
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2609.12223
UR - https://arxiv.org/abs/2609.12223
AB - Collision cross section (CCS), derived from ion mobility mass spectrometry, is a common descriptor for molecular annotation. Prediction is challenging for machine learning models because it reflects the size, shape, and ionization state of a gas-phase molecular ion. Most predictors either ignore explicit 3D structure or treat adduct identity as a late categorical feature, which limits their ability to capture adduct-dependent geometric effects. We present GRACE (Geometric Residual Adduct Conditioning via Early-fusion), a 3D CCS predictor that adapts a pretrained molecular geometry encoder using geometric residual adduct conditioning via early fusion. GRACE combines two inductive biases: a residual objective relative to an adduct-aware physical descriptor baseline and adduct conditioning within the encoder via a learned adduct token and low-rank attention adapters. We evaluate the model on a curated set of over 9,000 experimental molecule-adduct CCS records with random, scaffold, and adduct-sensitive splits designed to separate interpolation, scaffold generalization, and adduct-driven generalization. GRACE achieves the best mean percentage difference among the evaluated learned models on all three splits: 1.67% on the random split, 2.11% on the scaffold split, and 2.36% on the adduct-sensitive split. Diagnostic analyses suggest that residual learning stabilizes training by removing the dominant mass-CCS trend, while early fusion improves adduct-sensitive prediction relative to late fusion. Across four independent external test sets, GRACE shows consistently lower error than the other evaluated models. On a held-out set, GRACE also attains the lowest mean percent difference when compared with four previously reported physics-based workflows. These results support residual learning and encoder-level adduct conditioning as practical inductive biases for fast, accurate CCS prediction.
ER -