Graph-Based Active Machine Learning Method for Diverse and Novel Antimicrobial Peptides Generation and Selection

TL;DR

This paper introduces a graph-based active machine learning framework that efficiently generates and selects diverse, novel antimicrobial peptides, reducing experimental costs while maintaining high diversity and effectiveness.

Contribution

It presents a novel integration of recurrent neural networks and graph-based filtering to improve AMP candidate diversity and reduce experimental screening efforts.

Findings

The method produces more diverse AMP candidates.

It reduces the number of wet-lab experiments needed.

The approach outperforms existing methods in diversity and novelty metrics.

Abstract

As antibiotic-resistant bacterial strains are rapidly spreading worldwide, infections caused by these strains are emerging as a global crisis causing the death of millions of people every year. Antimicrobial Peptides (AMPs) are one of the candidates to tackle this problem because of their potential diversity, and ability to favorably modulate the host immune response. However, large-scale screening of new AMP candidates is expensive, time-consuming, and now affordable in developing countries, which need the treatments the most. In this work, we propose a novel active machine learning-based framework that statistically minimizes the number of wet-lab experiments needed to design new AMPs, while ensuring a high diversity and novelty of generated AMPs sequences, in multi-rounds of wet-lab AMP screening settings. Combining recurrent neural network models and a graph-based filter (GraphCC), our proposed approach delivers novel and diverse candidates and demonstrates better performances according to our defined metrics.