skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


  • NSF Public Access
  • Search Results
  • SAMP: Identifying antimicrobial peptides by an ensemble learning model based on proportionalized split amino acid composition
Title: SAMP: Identifying antimicrobial peptides by an ensemble learning model based on proportionalized split amino acid composition
Abstract It is projected that 10 million deaths could be attributed to drug-resistant bacteria infections in 2050. To address this concern, identifying new-generation antibiotics is an effective way. Antimicrobial peptides (AMPs), a class of innate immune effectors, have received significant attention for their capacity to eliminate drug-resistant pathogens, including viruses, bacteria, and fungi. Recent years have witnessed widespread applications of computational methods especially machine learning (ML) and deep learning (DL) for discovering AMPs. However, existing methods only use features including compositional, physiochemical, and structural properties of peptides, which cannot fully capture sequence information from AMPs. Here, we present SAMP, an ensemble random projection (RP) based computational model that leverages a new type of feature called proportionalized split amino acid composition (PSAAC) in addition to conventional sequence-based features for AMP prediction. With this new feature set, SAMP captures the residue patterns like sorting signals at both the N-terminal and the C-terminal, while also retaining the sequence order information from the middle peptide fragments. Benchmarking tests on different balanced and imbalanced datasets demonstrate that SAMP consistently outperforms existing state-of-the-art methods, such as iAMPpred and AMPScanner V2, in terms of accuracy, Matthews correlation coefficient (MCC), G-measure, and F1-score. In addition, by leveraging an ensemble RP architecture, SAMP is scalable to processing large-scale AMP identification with further performance improvement, compared to those models without RP. To facilitate the use of SAMP, we have developed a Python package that is freely available at https://github.com/wan-mlab/SAMP.  more » « less
Award ID(s):
2044049
PAR ID:
10636358
Author(s) / Creator(s):
; ; ; ; ; ; ;
Publisher / Repository:
Oxford
Date Published:
Journal Name:
Briefings in Functional Genomics
Volume:
23
Issue:
6
ISSN:
2041-2649
Page Range / eLocation ID:
879 to 890
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; (, Advances in Polymer Sciences and Technology)
    Bhuvanesh Gupta; Anup K.Ghosh; Atsushi Suzuki; Sunita Rattan (Ed.)
    Antibiotic resistance in bacteria is a major health concern. Antimicrobial Peptides (AMPs) are efficient in killing most microbes and yet the development of resistance to AMPs is rare. Although AMPs show promising antimicrobial activities, commercializing them as antibiotics is difficult as in vitro extraction and purification of AMPs is complicated and expensive. AMP mimicking antimicrobial polymers can overcome such problems while maintaining the necessary features of AMPs. Here, we have developed meth-acrylamide based polymers to mimic AMPs which possess high antimicrobial activities with low cytotoxicity. Bactericidal and scanning electron microscopy studies show that the synthesized polymers are effective against Gram-positive and Gram-negative bacteria. We find that these polymers are lethal to bacteria and at the same time, they are also non-cytotoxic to mammalian cells, thereby increasing the potential of these polymers to be used as antibiotics. 
    more » « less
  2. ; ; ; ; ; ; ; ; ; (, Advanced NanoBiomed Research)
    Antibiotics are losing effectiveness as bacteria become resistant to conventional drugs. To find new alternatives, antimicrobial peptides (AMPs) are rationally designed with different lengths, charges, hydrophobicities (H), and hydrophobic moments (μH), containing only three types of amino acids: arginine, tryptophan, and valine. Six AMPs with low minimum inhibitory concentrations (MICs) and <25% toxicity to mammalian cells are selected for biophysical studies. Their secondary structures are determined using circular dichroism (CD), which finds that the % α‐helicity of AMPs depends on composition of the lipid model membranes (LMMs): gram‐negative (G(−)) inner membrane (IM) >gram‐positive (G(+))> Euk33 (eukaryotic with 33 mol% cholesterol). The two most effective peptides, E2‐35 (16 amino acid [AA] residues) and E2‐05 (22 AAs), are predominantly helical in G(–) IM and G(+) LMMs. AMP/membrane interactions such as membrane elasticity, chain order parameter, and location of the peptides in the membrane are investigated by low‐angle and wide‐angle X‐ray diffuse scattering (XDS). It is found that headgroup location correlates with efficacy and toxicity. The membrane bending modulusKCdisplays nonmonotonic changes due to increasing concentrations of E2‐35 and E2‐05 in G(–) and G(+) LMMs, suggesting a bacterial killing mechanism where domain formation causes ion and water leakage. 
    more » « less
  3. ; ; ; ; ; ; ; ; (, Frontiers in Microbiology)
    The threat of antibiotic resistance warrants the discovery of agents with novel antimicrobial mechanisms. Antimicrobial peptides (AMPs) directly disrupting bacterial membranes may overcome resistance to traditional antibiotics. AMP development for clinical use has been mostly limited to topical application to date. We developed a rational framework for systematically addressing this challenge using libraries composed of 86 novel Trp- and Arg-rich engineered peptides tested against clinical strains of the most common multidrug-resistant bacteria known as ESKAPE pathogens. Structure-function correlations revealed minimum lengths (as low as 16 residues) and Trp positioning for maximum antibacterial potency with mean minimum inhibitory concentration (MIC) of 2–4 μM and corresponding negligible toxicity to mammalian cells. Twelve peptides were selected based on broad-spectrum activity against both gram-negative and -positive bacteria and <25% toxicity to mammalian cells at maximum test concentrations. Most of the selected PAX remained active against the colistin-resistant clinical strains. Of the selected peptides, the shortest (the 16-residue E35) was further investigated for antibacterial mechanism and proof-of-concept in vivo efficacy. E35 killed an extensively-resistant isolate of Pseudomonas aeruginosa (PA239 from the CDC, also resistant to colistin) by irreversibly disrupting the cell membranes as shown by propidium iodide incorporation, using flow cytometry and live cell imaging. As proof of concept, in vivo toxicity studies showed that mice tolerated a systemic dose of up to 30 mg/kg peptide and were protected with a single 5 mg/kg intravenous (IV) dose against an otherwise lethal intraperitoneal injection of PA239. Efficacy was also demonstrated in an immune-compromised Klebsiella pneumoniae infection model using a daily dose of 4mg/kg E35 systemically for 2 days. This framework defines the determinants of efficacy of helical AMPs composed of only cationic and hydrophobic amino acids and provides a path for a potential departure from the restriction to topical use of AMPs toward systemic application. 
    more » « less
  4. ; ; (, Journal of The Royal Society Interface)
    null (Ed.)
    Antimicrobial peptides (AMPs) produced by multi-cellular organisms as their immune system’s defence against microbes are actively considered as natural alternatives to conventional antibiotics. Although substantial progress has been achieved in studying the AMPs, the microscopic mechanisms of their functioning remain not well understood. Here, we develop a new theoretical framework to investigate how the AMPs are able to efficiently neutralize bacteria. In our minimal theoretical model, the most relevant processes, AMPs entering into and the following inhibition of the single bacterial cell, are described stochastically. Using complementary master equations approaches, all relevant features of bacteria clearance dynamics by AMPs, such as the probability of inhibition and the mean times before the clearance, are explicitly evaluated. It is found that both processes, entering and inhibition, are equally important for the efficient functioning of AMPs. Our theoretical method naturally explains a wide spectrum of efficiencies of existing AMPs and their heterogeneity at the single-cell level. Theoretical calculations are also consistent with existing single-cell measurements. Thus, the presented theoretical approach clarifies some microscopic aspects of the action of AMPs on bacteria. 
    more » « less
  5. ; ; ; (, Biology)
    null (Ed.)
    Machine learning algorithms can learn mechanisms of antimicrobial resistance from the data of DNA sequence without any a priori information. Interpreting a trained machine learning algorithm can be exploited for validating the model and obtaining new information about resistance mechanisms. Different feature extraction methods, such as SNP calling and counting nucleotide k-mers have been proposed for presenting DNA sequences to the model. However, there are trade-offs between interpretability, computational complexity and accuracy for different feature extraction methods. In this study, we have proposed a new feature extraction method, counting amino acid k-mers or oligopeptides, which provides easier model interpretation compared to counting nucleotide k-mers and reaches the same or even better accuracy in comparison with different methods. Additionally, we have trained machine learning algorithms using different feature extraction methods and compared the results in terms of accuracy, model interpretability and computational complexity. We have built a new feature selection pipeline for extraction of important features so that new AMR determinants can be discovered by analyzing these features. This pipeline allows the construction of models that only use a small number of features and can predict resistance accurately. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email