![Author ORCID: We display the ORCID iD icon alongside authors names on our website to acknowledge that the ORCiD has been authenticated when entered by the user. To view the users ORCiD record click the icon. [opens in a new tab]](https://chemrxiv.org/engage/assets/public/chemrxiv/images/logos/orcid.png)
Abstract
Controlling the formation and growth of ice is essential to successfully cryopreserve cells, tissues and biologics for research or clinical use. Current programmes to identify materials capable of modulating ice growth are guided solely by iterative changes and human intuition, with a major focus on macromolecules (proteins or polymers). This process is fundamentally constrained by a poor understanding of the mechanisms and underlying structure-activity relationships. Here, we overcome this barrier by constructing machine learning models capable of pre- dicting the ice growth inhibition activity of small molecules. Due to current limitations in experimental throughput, we leverage ensemble models which combine state-of-the-art descriptors with domain-specific features derived from molecular simulations. When applied to virtually screen a commercial compound library, these models successfully predicted novel ice recrystallisation inhibitors that are experimentally verified to function at low millimolar concentrations. This data-driven approach will enable the discovery of new cryoprotectants to address the rapidly growing clinical and biotechnological cold-chain demands.
Supplementary materials
Title
Supplementary Information
Description
Supplementary information includes: Additional experimental and computational details and methods including: nanoliter osmometry, classification metrics, molecular descriptor parameters, neural network hyperparameters, repeated LOO CV results, maximum expected performance estimation, Tanimoto similarities and dimensionality reduction. Supplementary figures including: Tanimoto coefficient distributions, dimensionality reduction biplots, hydra- tion number and index distributions, % MGS prediction distribution, chemical structures for the prediction set, additional IRI measurements in NaCl and PBS, hydration descriptors schematic, and cryomicrographs from nanolitre osmometry.
Actions
