Abstract
By targeting invasive organisms, antibiotics insert themselves into the ancient struggle of the host-pathogen evolutionary arms race. As pathogens evolve tactics for evading antibiotics, therapies decline in efficacy and must be replaced, distinguishing antibiotics from most other forms of drug development. Together with a slow and expensive antibiotic development pipeline, the proliferation of drug-resistant pathogens drives urgent interest in computational methods that promise to expedite candidate discovery. Strides in artificial intelligence (AI) have encouraged its application to multiple dimensions of computer-aided drug design, with increasing application to antibiotic discovery. This review describes AI-facilitated advances in the discovery of both small molecule antibiotics and antimicrobial peptides. Beyond the essential prediction of antimicrobial activity, emphasis is also given to antimicrobial compound representation, determination of drug-likeness traits, antimicrobial resistance, and de novo molecular design. Given the urgency of the antimicrobial resistance crisis, we analyze uptake of open science best practices in AI-driven antibiotic discovery and argue for openness and reproducibility as a means of accelerating preclinical research. Finally, trends in the literature and areas for future inquiry are discussed, as artificially intelligent enhancements to drug discovery at large offer many opportunities for future applications in antibiotic development.
Highlights
1234567890():,; By targeting invasive organisms, antibiotics insert themselves into the ancient struggle of the host-pathogen evolutionary arms race
Protein structures are probed for binding sites, allowing large libraries of compounds to be used for automated large-scale docking and binding affinity studies in a process known as virtual screening (VS)
This practice is integral to many drug development pipelines, receiving ample attention from the machine learning (ML) community[12]
Summary
In a recent effort to repurpose previously developed drugs as antibiotics[16], a combination of neural network models was used to create a new representation for chemical compounds, and assess their antimicrobial potential. This effort made use of ensemble learning[34], a technique that combines multiple copies of a model (with different weights or architectures) and takes a weighted vote of each model into consideration to achieve the final prediction[35]. A 2020 study created a deep convolutional neural network model based on a simplified amino acid vocabulary that translated the natural 20 amino acids into pseudo residue types[41] This model predicts antimicrobial activity in small peptides and is available in a web server.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
