Machine Learning‐Assisted Prediction and Generation of Antimicrobial Peptides

Antimicrobial peptides (AMPs) offer a highly potent alternative solution due to their broad‐spectrum activity and minimum resistance development against the rapidly evolving antibiotic‐resistant pathogens. Herein, to accelerate the discovery process of new AMPs, a predictive and generative algorithm...

Full description

Saved in:
Bibliographic Details
Main Authors: Sukhvir Kaur Bhangu, Nicholas Welch, Morgan Lewis, Fanyi Li, Brint Gardner, Helmut Thissen, Wioleta Kowalczyk
Format: Article
Language:English
Published: Wiley-VCH 2025-06-01
Series:Small Science
Subjects:
Online Access:https://doi.org/10.1002/smsc.202400579
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849335780054401024
author Sukhvir Kaur Bhangu
Nicholas Welch
Morgan Lewis
Fanyi Li
Brint Gardner
Helmut Thissen
Wioleta Kowalczyk
author_facet Sukhvir Kaur Bhangu
Nicholas Welch
Morgan Lewis
Fanyi Li
Brint Gardner
Helmut Thissen
Wioleta Kowalczyk
author_sort Sukhvir Kaur Bhangu
collection DOAJ
description Antimicrobial peptides (AMPs) offer a highly potent alternative solution due to their broad‐spectrum activity and minimum resistance development against the rapidly evolving antibiotic‐resistant pathogens. Herein, to accelerate the discovery process of new AMPs, a predictive and generative algorithm is build, which constructs new peptide sequences, scores their antimicrobial activity using a machine learning (ML) model, identifies amino acid motifs, and assembles high‐ranking motifs into new peptide sequences. The eXtreme Gradient Boosting model achieves an accuracy of ≈87% in distinguishing between AMPs and non‐AMPs. The generated peptide sequences are experimentally validated against the bacterial pathogens, and an accuracy of ≈60% is achieved. To refine the algorithm, the physicochemical features are analyzed, particularly charge and hydrophobicity of experimentally validated peptides. The peptides with specific range of charge and hydrophobicity are then removed, which lead to a substantial increase in an experimental accuracy, from ≈60% to ≈80%. Furthermore, generated peptides are active against different fungal strains with minimal off‐target toxicity. In summary, in silico predictive and generative models for functional motif and AMP discovery are powerful tools for engineering highly effective AMPs to combat multidrug resistant pathogens.
format Article
id doaj-art-ae6dc83f81e84a228d8f62fbe46040b8
institution Kabale University
issn 2688-4046
language English
publishDate 2025-06-01
publisher Wiley-VCH
record_format Article
series Small Science
spelling doaj-art-ae6dc83f81e84a228d8f62fbe46040b82025-08-20T03:45:10ZengWiley-VCHSmall Science2688-40462025-06-0156n/an/a10.1002/smsc.202400579Machine Learning‐Assisted Prediction and Generation of Antimicrobial PeptidesSukhvir Kaur Bhangu0Nicholas Welch1Morgan Lewis2Fanyi Li3Brint Gardner4Helmut Thissen5Wioleta Kowalczyk6CSIRO Manufacturing Research Way Clayton Victoria 3168 AustraliaCSIRO Manufacturing Research Way Clayton Victoria 3168 AustraliaCSIRO Information Management & Technology Kensington Western Australia 6151 AustraliaCSIRO Manufacturing Research Way Clayton Victoria 3168 AustraliaCSIRO Information Management & Technology Research Way Clayton Victoria 3168 AustraliaCSIRO Manufacturing Research Way Clayton Victoria 3168 AustraliaCSIRO Manufacturing Research Way Clayton Victoria 3168 AustraliaAntimicrobial peptides (AMPs) offer a highly potent alternative solution due to their broad‐spectrum activity and minimum resistance development against the rapidly evolving antibiotic‐resistant pathogens. Herein, to accelerate the discovery process of new AMPs, a predictive and generative algorithm is build, which constructs new peptide sequences, scores their antimicrobial activity using a machine learning (ML) model, identifies amino acid motifs, and assembles high‐ranking motifs into new peptide sequences. The eXtreme Gradient Boosting model achieves an accuracy of ≈87% in distinguishing between AMPs and non‐AMPs. The generated peptide sequences are experimentally validated against the bacterial pathogens, and an accuracy of ≈60% is achieved. To refine the algorithm, the physicochemical features are analyzed, particularly charge and hydrophobicity of experimentally validated peptides. The peptides with specific range of charge and hydrophobicity are then removed, which lead to a substantial increase in an experimental accuracy, from ≈60% to ≈80%. Furthermore, generated peptides are active against different fungal strains with minimal off‐target toxicity. In summary, in silico predictive and generative models for functional motif and AMP discovery are powerful tools for engineering highly effective AMPs to combat multidrug resistant pathogens.https://doi.org/10.1002/smsc.202400579antimicrobial peptidesantimicrobial resistancesbacteriamachine learningmultidrug resistances
spellingShingle Sukhvir Kaur Bhangu
Nicholas Welch
Morgan Lewis
Fanyi Li
Brint Gardner
Helmut Thissen
Wioleta Kowalczyk
Machine Learning‐Assisted Prediction and Generation of Antimicrobial Peptides
Small Science
antimicrobial peptides
antimicrobial resistances
bacteria
machine learning
multidrug resistances
title Machine Learning‐Assisted Prediction and Generation of Antimicrobial Peptides
title_full Machine Learning‐Assisted Prediction and Generation of Antimicrobial Peptides
title_fullStr Machine Learning‐Assisted Prediction and Generation of Antimicrobial Peptides
title_full_unstemmed Machine Learning‐Assisted Prediction and Generation of Antimicrobial Peptides
title_short Machine Learning‐Assisted Prediction and Generation of Antimicrobial Peptides
title_sort machine learning assisted prediction and generation of antimicrobial peptides
topic antimicrobial peptides
antimicrobial resistances
bacteria
machine learning
multidrug resistances
url https://doi.org/10.1002/smsc.202400579
work_keys_str_mv AT sukhvirkaurbhangu machinelearningassistedpredictionandgenerationofantimicrobialpeptides
AT nicholaswelch machinelearningassistedpredictionandgenerationofantimicrobialpeptides
AT morganlewis machinelearningassistedpredictionandgenerationofantimicrobialpeptides
AT fanyili machinelearningassistedpredictionandgenerationofantimicrobialpeptides
AT brintgardner machinelearningassistedpredictionandgenerationofantimicrobialpeptides
AT helmutthissen machinelearningassistedpredictionandgenerationofantimicrobialpeptides
AT wioletakowalczyk machinelearningassistedpredictionandgenerationofantimicrobialpeptides