AMPlify: attentive deep learning model for discovery of novel antimicrobial peptides effective against WHO priority pathogens

Chenkai Li,Chen Yang,Caren C Helbing,Caroline E Cameron,Simon Houston,Lauren Bergman,Darcy Sutherland,René L Warren,Linda M N Hoang,Titus Wong,Figali Taho,Inanc Birol,S Austin Hammond

doi:10.1186/s12864-022-08310-4

Abstract

BackgroundAntibiotic resistance is a growing global health concern prompting researchers to seek alternatives to conventional antibiotics. Antimicrobial peptides (AMPs) are attracting attention again as therapeutic agents with promising utility in this domain, and using in silico methods to discover novel AMPs is a strategy that is gaining interest. Such methods can sift through large volumes of candidate sequences and reduce lab screening costs.ResultsHere we introduce AMPlify, an attentive deep learning model for AMP prediction, and demonstrate its utility in prioritizing peptide sequences derived from the Rana [Lithobates] catesbeiana (bullfrog) genome. We tested the bioactivity of our predicted peptides against a panel of bacterial species, including representatives from the World Health Organization’s priority pathogens list. Four of our novel AMPs were active against multiple species of bacteria, including a multi-drug resistant isolate of carbapenemase-producing Escherichia coli.ConclusionsWe demonstrate the utility of deep learning based tools like AMPlify in our fight against antibiotic resistance. We expect such tools to play a significant role in discovering novel candidates of peptide-based alternatives to classical antibiotics.

Highlights

Antibiotic resistance is a growing global health concern prompting researchers to seek alternatives to conventional antibiotics
We introduce AMPlify, an attentive deep learning model that improves in silico Antimicrobial peptide (AMP) prediction by applying two types of attention mechanisms layered on a bidirectional long short-term memory [18,19,20] (BiLSTM) layer (Fig. 1)
Evaluation of model architecture To demonstrate the effectiveness of each component within our model, we evaluated the model architecture starting from a single bidirectional long short-term memory [18–20] (Bi-long shortterm memory (LSTM)) layer and gradually adding attention layers over it

Summary

Introduction

Antibiotic resistance is a growing global health concern prompting researchers to seek alternatives to conventional antibiotics. Antimicrobial peptides (AMPs) are attracting attention again as therapeutic agents with promising utility in this domain, and using in silico methods to discover novel AMPs is a strategy that is gaining interest. Such methods can sift through large volumes of candidate sequences and reduce lab screening costs. Large scale discovery of novel AMPs through wet lab screening is time-consuming, labor-intensive and costly [12] For these reasons, various computational models have been developed over the last few years [12] to streamline in silico AMP prediction. Despite the rapid progress in the field, currently available models still have substantial room for improvement

Methods

Results

Discussion

Conclusion