Antimicrobial resistance (AMR) is a global issue due to improper drug use in humans and animals. Antimicrobial peptides (AMPs) show promise in targeting bacteria with minimal harm to host cells and low risk of resistance development. Machine learning enhances accuracy in predicting AMPs. Common classifiers include SVM, RF, ANN, LGBM, and DT. This review compares peptide prediction tools based on machine learning, assessing performance using cross-validation. Carefully chosen independent datasets were used to evaluate predictive efficiency. By utilizing a variety of ML methods, the best techniques for predicting Antimicrobial peptides, Antibacterial peptides, Antifungal peptides can be developed quickly
Top1. Introduction
Antibiotic resistance is a global threat that could lead to a worldwide pandemic. Antimicrobial resistance is caused by excessive and improper use of drugs in humans and animals. Unnecessary use or incomplete treatment can lead to microbial resistance (Ayukekbong et al., 2017). Regular antibiotics are losing their effectiveness very quickly around the world. As a result, a lot of studies are being done to find new, non-standard antibiotics that can fight infections. Antimicrobial use in agriculture can also lead to resistant bacteria spreading to humans through food (Economou et al., 2015). Antimicrobial peptides (AMPs), also known as host defense peptides, are promising infectious disease medication and immunomodulatory therapeutics (Mahlapuu et al., 2020). For that, the development of a new form of antimicrobial drugs is the only solution to prevent infections and save countless lives, just like antibiotics have done since their initial discovery and production (Ventola et al., 2015; Nallakaruppan et al., 2023). It has been recognized as one of humanity's most significant accomplishments of the 20th century. However, pathogens can counterattack and reduce the effectiveness of conventional antibiotics. Proteases are used to break down antibiotics, which are then pumped out of the cells and altered to decrease drug affinity towards specific targets by the bacteria (Monserrat et al., 2019).
Antimicrobial peptides (AMPs) are short, cationic molecules found in various organisms, exhibiting selectivity for bacteria, low cytotoxicity, and potential for therapeutic applications (Olga et al., 2021; Junker et al., 2013). Rogelio in 2020 stated that AMPs target a range of pathogens, making them promising for wound healing. Antimicrobial peptides (AMPs) are nature's mini-warriors, acting as the first line of defense against microbes in diverse organisms. These short, potent molecules punch through bacterial membranes, disrupt vital functions, and even neutralize biofilms, offering a promising weapon against antibiotic-resistant bacteria. Their natural origin minimizes toxicity concerns, and their diverse modes of action make it harder for microbes to develop resistance. While still under development, AMPs hold immense potential as future antibiotics, offering a glimpse into a new era of targeted and effective treatments (Lei et al., 2019). Evidence suggests AMPs act beyond membrane disruption, inhibiting crucial intracellular processes (Moghaddamet al., 2014).
Li and Brogden introduced antibacterial peptides (ABPs) which are a subset of AMPs, and show broad inhibitory effects on bacteria. Antibacterial peptides (ABPs) show immense potential as future antibiotics due to their unique biological properties, exhibiting broad inhibitory effects on pathogenic bacteria. This not only addresses current infections but also provides a potential solution to the growing challenge of antibiotic resistance. Nonetheless, ongoing research aims to refine ABPs and harness their full potential as potent natural weapons against bacterial infections (Huan et al., 2020).