Identification of CXCR4 inhibitory activity in natural compounds using cheminformatics-guided machine learning algorithms

Abstract Neurodegenerative disorders are characterised by progressive damage to neurons that leads to cognitive impairment and motor dysfunction. Current treatment options focus only on symptom management and palliative care, without addressing their root cause. In our previous study, we reported the upregulation of the CXC motif chemokine receptor 4 (CXCR4), in Alzheimer’s disease (ad) and Parkinson’s disease (PD). We reached this conclusion by analysing gene expression patterns of ad and PD patients, compared to healthy individuals of similar age. We used RNA sequencing data from Gene Expression Omnibus to carry out this analysis. Herein, we aim to identify natural compounds that have potential inhibitory activity against CXCR4 through cheminformatics-guided machine learning, to aid drug discovery for neurodegenerative disorders, especially ad and PD. Natural compounds are gaining prominence in the treatment of neurodegenerative disorders due to their biocompatibility and potential neuroprotective properties, including their ability to modulate CXCR4 expression. Recent advances in artificial intelligence (AI) and machine learning (ML) algorithms have opened new avenues for drug discovery research across various therapeutic areas, including neurodegenerative disorders. We aim to produce an ML model using cheminformatics-guided machine learning algorithms using data of compounds with known CXCR4 activity, retrieved from the Binding Database, to analyse various physicochemical attributes of natural compounds obtained from the COCONUT Database and predict their inhibitory activity against CXCR4. Insight Box This work extends our previous study published in Integrative Biology (DOI: 10.1093/intbio/zyad012). We aim to demonstrate the effectiveness of AI and ML in identifying potential treatment options for Alzheimer’s and Parkinson’s diseases. By analysing vast amounts of data and identifying patterns that may not be apparent to human researchers, AI-powered systems can provide valuable insight into potential treatment options that may have been overlooked through traditional research methods. Our study underscores the significance of interdisciplinary collaboration between computational and experimental scientists in drug discovery and in developing a robust pipeline to identify potential leads for drug development.