New Inhibitors of the P38 Mitogen-Activated Protein Kinase: Repurposing Existing Drugs with Deep Learning

Abstract

The p38-alpha (MAPK14) is a protein kinase that is implicated in the pathological mechanisms of BAG3P209L myofibrillar myopathy, cancers, and inflammatory diseases like Alzheimer’s and rheumatoid arthritis. Inhibition of p38 has shown promise as a treatment for these diseases. Traditional drug discovery methods could not create effective and safe small molecule inhibitors, so we used machine learning to elucidate potential p38 blockers from existing FDA-approved drugs. Using PubChem bioactivity data, we determined the best existing p38 inhibitors and applied fingerprint clustering to isolate the compounds with similar structures. Descriptors were calculated for these clustered compounds, and the most important of these descriptors were determined through a machine learning-based feature selection algorithm. This data served as the training set for a deep neural network that was fine-tuned to a 92% validation accuracy. The neural network model was applied to a database of FDA-approved drugs, revealing 149 potential p38 inhibitors, whose efficacy was confirmed by docking simulations to be statistically significantly higher than random FDA drugs and slightly higher than known inhibitors. Our study not only reveals potential medications for p38-mediated diseases that we recommend for physical trials but also demonstrates the ability of our novel deep learning-based computational pipeline to predict new functions of existing pharmaceuticals.