Simulation-Supervised Foundation Models for Retention Time Prediction in High-Performance Liquid Chromatography beyond Experimental Data Coverage
Abstract
Accurate prediction of high-performance liquid chromatography (HPLC) retention times (RTs) across diverse molecules and chromatographic methods remains challenging because experimental training data cover only a limited region of chemical and method spaces. Here, we develop FUSE-RT (Foundation model Unifying Simulation and Experimental supervision for Retention Time), a multitask foundation model that integrates RT data from 179 chromatographic methods and adapts to unseen molecules and methods using limited target-domain data. To extend transferability beyond experimental coverage, we introduce simulation-to-real (Sim2Real) transfer learning, in which molecular representations learned from large-scale computational data are transferred to experimental RT prediction. Specifically, we use PolyOmics, comprising 39 properties for approximately 21,400 molecules generated by molecular dynamics and density-functional theory calculations, as auxiliary supervision. We evaluate generalization under molecular, method, and joint molecular--method distribution shifts. Simulation-derived supervision substantially improves transfer beyond the experimental molecular domain, particularly under pronounced coverage gaps and few-shot adaptation. Moreover, RT-prediction error decreases systematically with increasing simulation-data size, following a significant power-law relationship. These results establish Sim2Real transfer as a scalable strategy for extending RT prediction beyond the finite coverage of experimental chromatographic data.
Related
- Ultra-High-Performance Liquid Chromatography-High Resolution Mass Spectrometry and Gas Chromatography-Mass Spectrometry Screening of Classical Psychoactive Substances, Novel Psychoactive Substances, and Metabolites in Postmortem Urine Samples
- ［2+4］型手性多孔有机笼Pim高效液相色谱固定相的制备及其手性拆分性能评价 Preparation and chiral separation performance evaluation of a high performance liquid chromatography stationary phase based on a ［2+4］-type chiral porous organic cage Pim
- Modern Optimization Strategies in High‐Performance Liquid Chromatography Analysis
- Separation of polar and neutral lipids from mammalian cell lines by high-performance thin-layer chromatography
- Precision Profiling of α‐1,4‐ Versus α‐1,6‐Glycosidase Activity Through Synthetic Linear and Branched Glycogen Tetramers Using Cyclic Ion Mobility Mass Spectrometry and High‐Performance Liquid Chromatography‐Mass Spectrometry
BibTeX
@misc{wu2026simulationsupervised,
title = {{Simulation-Supervised Foundation Models for Retention Time Prediction in High-Performance Liquid Chromatography beyond Experimental Data Coverage}},
author = {Stephen Wu and Yufeng Han and Yasuhiro Mito and Yoshiyuki Watabe and Yoshihiro Hayashi and Hikaru Takaya and Takuya Kubo and Ryo Yoshida},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.07531},
eprint = {2609.07531},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.07531},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - Simulation-Supervised Foundation Models for Retention Time Prediction in High-Performance Liquid Chromatography beyond Experimental Data Coverage
AU - Stephen Wu
AU - Yufeng Han
AU - Yasuhiro Mito
AU - Yoshiyuki Watabe
AU - Yoshihiro Hayashi
AU - Hikaru Takaya
AU - Takuya Kubo
AU - Ryo Yoshida
PY - 2026
DA - 2026/09/07
PB - arXiv
DO - 10.48550/arxiv.2609.07531
UR - https://arxiv.org/abs/2609.07531
AB - Accurate prediction of high-performance liquid chromatography (HPLC) retention times (RTs) across diverse molecules and chromatographic methods remains challenging because experimental training data cover only a limited region of chemical and method spaces. Here, we develop FUSE-RT (Foundation model Unifying Simulation and Experimental supervision for Retention Time), a multitask foundation model that integrates RT data from 179 chromatographic methods and adapts to unseen molecules and methods using limited target-domain data. To extend transferability beyond experimental coverage, we introduce simulation-to-real (Sim2Real) transfer learning, in which molecular representations learned from large-scale computational data are transferred to experimental RT prediction. Specifically, we use PolyOmics, comprising 39 properties for approximately 21,400 molecules generated by molecular dynamics and density-functional theory calculations, as auxiliary supervision. We evaluate generalization under molecular, method, and joint molecular--method distribution shifts. Simulation-derived supervision substantially improves transfer beyond the experimental molecular domain, particularly under pronounced coverage gaps and few-shot adaptation. Moreover, RT-prediction error decreases systematically with increasing simulation-data size, following a significant power-law relationship. These results establish Sim2Real transfer as a scalable strategy for extending RT prediction beyond the finite coverage of experimental chromatographic data.
ER -