Rethinking Reaction Discovery: From Serendipity to Data-Driven Chemistry

23 December 2025, Version 1
This content is an early or alternative research output and has not been peer-reviewed by Cambridge University Press at the time of posting.

Abstract

The discovery of novel chemical reactions attracts chemists both in academia and industry. Discovery can arise from either targeted design or unforeseen accident. However, in most cases an impulse towards the detailed investigation of the particular field is detection of an unexpected product in the chemical reaction. Herein, we report the possibility of searching for new chemical reactions by analyzing reaction mixtures using an artificial intelligence platform – OdanChem. This approach has high potential to accelerate the discovery of new reactions, and it shows advanced capabilities for (semi-)autonomous data analysis. Our results indicate that new reactions can be found even with simple chemical compounds, demonstrating the versatility and efficiency of this approach.

Keywords

OdanChem
NMR
AI
Products prediction
Reactions discovery

Supplementary materials

Title
Description
Actions
Title
ESI
Description
Supporting Information
Actions

Supplementary weblinks

Comments

Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting and Discussion Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.