Title: Machine learning approaches to optimize small-molecule inhibitors for RNA targeting.
Description: We report the use of data-driven algorithms that consolidate bioisosteric rules for the preferential modifications on small molecules with a common molecular scaffold obtained by NMR-fragment screening. Our manuscript presents complementary data-driven algorithms to minimize the search in chemical space for phenylthiazole-containing molecules that bind the RNA hairpin within the ribosomal peptidyl transferase center (PTC) of Mycobacterium tuberculosis. The RNA hairpin we chose as a target for inhibition is an important hub for protein synthesis and is hence the target of many antibiotics in clinical use. we have applied machine learning to smartly design, at an unprecedented pace, new small-molecule inhibitors that target RNA in the heart of the ribosome.
Technologies: we used Lasso regression, decision tree, and convolutional neural networks that use pictorial and 3D structures as input.
Copyright (c) 2021 Barak Akabayov Licensed under the MIT License.