Discovery of a structural class of antibiotics with explainable deep learning.

The discovery of novel structural classes of antibiotics is urgently needed to address the ongoing antibiotic resistance crisis1-9. Deep learning approaches have aided in exploring chemical spaces1,10-15; these typically use black box models and do not provide chemical insights. Here we reasoned that the chemical substructures associated with antibiotic activity learned by neural network models can be identified and used to predict structural classes of antibiotics. We tested this hypothesis by developing an explainable, substructure-based approach for the efficient, deep learning-guided exploration of chemical spaces. We determined the antibiotic activities and human cell cytotoxicity profiles of 39,312 compounds and applied ensembles of graph neural networks to predict antibiotic activity and cytotoxicity for 12,076,365 compounds. Using explainable graph algorithms, we identified substructure-based rationales for compounds with high predicted antibiotic activity and low predicted cytotoxicity. We empirically tested 283 compounds and found that compounds exhibiting antibiotic activity against Staphylococcus aureus were enriched in putative structural classes arising from rationales. Of these structural classes of compounds, one is selective against methicillin-resistant S. aureus (MRSA) and vancomycin-resistant enterococci, evades substantial resistance, and reduces bacterial titres in mouse models of MRSA skin and systemic thigh infection. Our approach enables the deep learning-guided discovery of structural classes of antibiotics and demonstrates that machine learning models in drug discovery can be explainable, providing insights into the chemical substructures that underlie selective antibiotic activity.

Optimization of a series of heterocycles as survival motor neuron gene transcription enhancers.

Spinal muscular atrophy (SMA) is a neurodegenerative disorder that results from mutations in the SMN1 gene, leading to survival motor neuron (SMN) protein deficiency. One therapeutic strategy for SMA is to identify compounds that enhance the expression of the SMN2 gene, which normally only is a minor contributor to functional SMN protein production, but which is unaffected in SMA. A recent high-throughput screening campaign identified a 3,4-dihydro-4-phenyl-2(1H)-quinolinone derivative (2) that increases the expression of SMN2 by 2-fold with an EC50 = 8.3 µM. A structure-activity relationship (SAR) study revealed that the array of tolerated substituents, on either the benzo portion of the quinolinone or the 4-phenyl, was very narrow. However, the lactam ring of the quinolinone was more amenable to modifications. For example, the quinazolinone (9a) and the benzoxazepin-2(3H)-one (19) demonstrated improved potency and efficacy for increase in SMN2 expression as compared to 2.