Comparative analysis of machine learning algorithms on the microbial strain-specific AMP prediction

Abstract

The evolution of drug-resistant pathogenic microbial species is a major global health concern. Naturally occurring, antimicrobial peptides (AMPs) are considered promising candidates to address antibiotic resistance problems. A variety of computational methods have been developed to accurately predict AMPs. The majority of such methods are not microbial strain specific (MSS): they can predict whether a given peptide is active against some microbe, but cannot accurately calculate whether such peptide would be active against a particular MS. Due to insufficient data on most MS, only a few MSS predictive models have been developed so far. To overcome this problem, we developed a novel approach that allows to improve MSS predictive models (MSSPM), based on properties, computed for AMP sequences and characteristics of genomes, computed for target MS. New models can perform predictions of AMPs for MS that do not have data on peptides tested on them. We tested various types of feature engineering as well as different machine learning (ML) algorithms to compare the predictive abilities of resulting models. Among the ML algorithms, Random Forest and AdaBoost performed best. By using genome characteristics as additional features, the performance for all models increased relative to models relying on AMP sequence-based properties only. Our novel MSS AMP predictor is freely accessible as part of DBAASP database resource at http://dbaasp.org/prediction/genome

Antimicrobial Peptides in the Global Microbiome: Biosynthetic Genes and Resistance Determinants

“…The diversity of AMPs probably reflects the ability of species to adapt to the unique microbial environment of the niche they occupy, as a single mutation can greatly change the biological activity of each peptide. Complex microbial ecosystems such as the human gut are locations where long-term competition and co-evolution could generate antimicrobial compounds such as AMPs . Therefore, the human gut microbiome could serve as a potential source of AMPs that could be used to address the antibiotic resistance crisis.…”

“…Complex microbial ecosystems such as the human gut are locations where long-term competition and co-evolution could generate antimicrobial compounds such as AMPs. 52 Therefore, the human gut microbiome could serve as a potential source of AMPs that could be used to address the antibiotic resistance crisis.…”

Diversity of the Antimicrobial Peptide Genes in Collembola

“…The heatmap diagram was made using Microsoft Excel. We used the DBAASP database [ 52 , 53 ] to predict the antimicrobial activities of collembolan AMPs against specific strains of bacteria and fungi. Three available methods were utilized to predict antibacterial activities against five bacterial strains: Escherichia coli ATCC 25922, Pseudomonas aeruginosa ATCC 27853, Klebsiella pneumoniae , Staphylococcus aureus ATC 25923, and Bacillus subtilis .…”

Analysis, Modeling, and Target-Specific Predictions of Linear Peptides Inhibiting Virus Entry

“…Many authors have developed AMP predictive models based solely on peptide sequence properties, but information about targets is very important for developing prediction models. Some articles use information about targets using multitarget or multitasking models, − and tools for developing the corresponding prediction models are also now available. , Our approaches use attributes of the target strain to develop predictive models . Such an approach allows us to predict peptide’s activity against new strains for which the information about the antimicrobial activity is not available, but data on the envelope protein structure are known.…”

“…In such a case, available data on activities and data on similarity between structures of envelope proteins allow us to predict the potency of peptides against strain with untested susceptibility. Genome or other data can be used as attributes. Taking into account that envelope proteins play an essential role in the fusion process, and the fact that many peptides derived from envelope proteins have the anti-VEIP activity, − properties of the envelope proteins of the target viruses can be used as attributes to develop models.…”