Article Automated synthesis and fragment descriptor-based machine learning for retention time prediction in supercritical fluid chromatography

Sitanan Sartyoungkul ; Balasubramaniyan Sakthivel ; Pavel Sidorov ORCID ; Yuuya Nagata ORCID

D5DD00437C.pdf
Download file (1.93 MB)
Download as zip (1.74 MB)
Collection

Citation
Sitanan Sartyoungkul, Balasubramaniyan Sakthivel, Pavel Sidorov, Yuuya Nagata. Automated synthesis and fragment descriptor-based machine learning for retention time prediction in supercritical fluid chromatography. Digital Discovery. 2025, (), . https://doi.org/10.1039/d5dd00437c

Description:

(abstract)

This study integrates automated synthesis with machine learning to enable data-driven chromatographic method development. A diverse amide library was synthesized and analyzed on the DCpak PBT column, generating a curated retention-time dataset for supercritical fluid chromatography. Fragment count descriptors such as ChyLine and CircuS outperformed traditional fingerprints, offering higher accuracy and interpretable structure–retention relationships. Visualization methods provided mechanistic insight into key molecular features. Overall, the workflow demonstrates a scalable strategy for producing high-quality data and predictive models, highlighting how automation and chemoinformatics can accelerate column characterization and support autonomous analytical workflows.

Rights:

Keyword: Automated synthesis, Fragment descriptor, Machine learning, Retention time prediction, Supercritical fluid chromatography

Date published: 2025-11-26

Publisher: Royal Society of Chemistry (RSC)

Journal:

Funding:

  • Exploratory Research for Advanced Technology JPMJER1903
  • Ministry of Education, Culture, Sports, Science and Technology
  • Japan Society for the Promotion of Science JP23H03810
  • Japan Society for the Promotion of Science JP23H03807
  • Japan Society for the Promotion of Science JP23H03806

Manuscript type: Publisher's version (Version of record)

MDR DOI:

First published URL: https://doi.org/10.1039/d5dd00437c

Related item:

Other identifier(s):

Contact agent:

Updated at: 2025-12-24 15:05:55 +0900

Published on MDR: 2025-12-24 16:20:29 +0900

Filename Size
Filename D5DD00437C.pdf (Thumbnail)
application/pdf 		Size 1.93 MB Detail