CoPriNet: Deep learning compound price prediction for use in de novo molecule generation and prioritization.

03 March 2022, Version 1
This content is a preprint and has not undergone peer review at the time of posting.


Compound availability is a critical property for design prioritization across the drug discovery pipeline. Historically, and despite its multiple limitations, compound-oriented synthetic accessibility scores have been used as proxies for this problem. However, the size of the catalogues of commercially available molecules has dramatically increased over the last decade, redefining the problem of compound accessibility as a matter of budget. In this paper we show that if compound prices are an alternative proxy for compound availability, then synthetic accessibility scores are not effective strategies for assessing availability. Instead, we learn how to predict prices directly from the catalogues. Our approached, CopriNet, is a retrosynthesis-free deep learning model trained on pairs of compound/prices extracted from the Mcule catalogue. CoPriNet is able to provide price predictions that exhibit far better correlation with actual compound prices than any synthetic accessibility measurement. Moreover, unlike standard retrosynthesis methods, CoPriNet is rapid, comparable in execution time to popular synthetic accessibility metrics and thus is suitable for high-throughput experiments including virtual screening and de novo compound generation.


Deep learning
synthetic accesibility
price prediction

Supplementary materials

Supplementary Material
Supplementary material sections 1-8


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 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.