Machine Learning to Unravel Anti-Ageing Compounds
Lead Research Organisation:
University of Kent
Department Name: Sch of Computing
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Organisations
Publications
Ribeiro Caio
(2023)
Predicting lifespan-extending chemical compounds for C. elegans with machine learning and biologically interpretable features
in AGING-US
| Description | We have used machine learning (a type of Artificial Intelligence method) for generating computational models that predict whether or not a given drug (chemical compound) will extend the lifespan of a nematode worm (C. elegans), based on data about the effects of many drugs on a worm's lifespan (data extracted from the DrugAge database, freely available on the web). We also analysed the generated predictive models from the perspective of the biology of ageing, in order to identify the main properties of drugs (chemical compounds) that are relevant for predicting a worm's lifespan extension, and one noteworthy property was a compound's role in cellular redox homeostasis and detoxification. In addition, we used the generated models to predict the most promising novel compounds for extending a worm's lifespan from a list of compounds with no previously known lifespan-increase effect. These identified novel compounds include e.g. nitroprusside, which is used as an antihypertensive medication. We also used machine learning to predict whether or not a given chemical compound will extend the lifespan of mice (M. musculus). When analysing these predictive models, we noted that predictive features related to G-protein coupled receptors, especially receptors for neurotransmitters, metabolic hormones and sex hormones, were identified as strong predictors of lifespan extension in mice. We also used the top-performed predictive models to predict the most promising novel compounds for extending mice lifespan, among compounds in the DrugBank database with no previously known lifespan-increase effect. The most promising compounds identified by this procedure appear to target IGF1 and insulin receptors, beta adrenergic receptors, carbonic anhydrases, dopamine and serotonin receptors, voltage-gated potassium and calcium channels, sodium-dependent dopamine, serotonin and noradrenalin transporters, muscarinic acetylcholine receptors and adenosine receptors. |
| Exploitation Route | Broadly speaking, this work can lead to future work in employing machine learning to predict novel life-extending compounds. |
| Sectors | Pharmaceuticals and Medical Biotechnology |