Optimization of antibacterial peptides by genetic algorithms and cheminformatics

Christopher D. Fjell, Håvard Jenssen, Warren A. Cheung, Robert E. W. Hancock, Artem Cherkasov

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

    Resumé

    Pathogens resistant to available drug therapies are a pressing global health problem. Short, cationic peptides represent a novel class of agents that have lower rates of drug resistance than derivatives of current antibiotics. Previously, we created a software system utilizing artificial neural networks that were trained on quantitative structure-activity relationship descriptors calculated for a total of 1400 synthetic peptides for which antibacterial activity was determined. Using the trained system, we correctly identified additional peptides with activity of 94% accuracy; active peptides were 47 of the top rated 50 peptides chosen from an in silico library of nearly 100 000 sequences. Here, we report a method of generating candidate peptide sequences using the heuristic evolutionary programming method of genetic algorithms (GA), which provided a large (19-fold) improvement in identification of novel antibacterial peptides. Approximately 0.50% of peptides evaluated during the GA method were classified as highly active, while only 0.026% of the nearly 100 000 sequences we previously screened were classified as highly active. A selection of these peptides was tested in vitro and activities reported here. While GA significantly improves the possibility of identifying candidate peptides, we encountered important pitfalls to this method that should be considered when using GA.
    OriginalsprogEngelsk
    TidsskriftChemical Biology and Drug Design (Print)
    Vol/bind77
    Udgave nummer1
    Sider (fra-til)48-56
    ISSN1747-0277
    DOI
    StatusUdgivet - 2011

    Citer dette

    Fjell, Christopher D. ; Jenssen, Håvard ; Cheung, Warren A. ; Hancock, Robert E. W. ; Cherkasov, Artem . / Optimization of antibacterial peptides by genetic algorithms and cheminformatics. I: Chemical Biology and Drug Design (Print). 2011 ; Bind 77, Nr. 1. s. 48-56.
    @article{a1043b1f0c644e7c9915ae1bc89210b9,
    title = "Optimization of antibacterial peptides by genetic algorithms and cheminformatics",
    abstract = "Pathogens resistant to available drug therapies are a pressing global health problem. Short, cationic peptides represent a novel class of agents that have lower rates of drug resistance than derivatives of current antibiotics. Previously, we created a software system utilizing artificial neural networks that were trained on quantitative structure-activity relationship descriptors calculated for a total of 1400 synthetic peptides for which antibacterial activity was determined. Using the trained system, we correctly identified additional peptides with activity of 94{\%} accuracy; active peptides were 47 of the top rated 50 peptides chosen from an in silico library of nearly 100 000 sequences. Here, we report a method of generating candidate peptide sequences using the heuristic evolutionary programming method of genetic algorithms (GA), which provided a large (19-fold) improvement in identification of novel antibacterial peptides. Approximately 0.50{\%} of peptides evaluated during the GA method were classified as highly active, while only 0.026{\%} of the nearly 100 000 sequences we previously screened were classified as highly active. A selection of these peptides was tested in vitro and activities reported here. While GA significantly improves the possibility of identifying candidate peptides, we encountered important pitfalls to this method that should be considered when using GA.",
    keywords = "antibacterial peptides, cheminformatics, genetic algorithms, quantitative structure-activity relationship",
    author = "Fjell, {Christopher D.} and H{\aa}vard Jenssen and Cheung, {Warren A.} and Hancock, {Robert E. W.} and Artem Cherkasov",
    year = "2011",
    doi = "10.1111/j.1747-0285.2010.01044.x",
    language = "English",
    volume = "77",
    pages = "48--56",
    journal = "Chemical Biology and Drug Design (Print)",
    issn = "1747-0277",
    publisher = "Wiley-Blackwell Publishing, Inc",
    number = "1",

    }

    Optimization of antibacterial peptides by genetic algorithms and cheminformatics. / Fjell, Christopher D. ; Jenssen, Håvard; Cheung, Warren A. ; Hancock, Robert E. W. ; Cherkasov, Artem .

    I: Chemical Biology and Drug Design (Print), Bind 77, Nr. 1, 2011, s. 48-56.

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

    TY - JOUR

    T1 - Optimization of antibacterial peptides by genetic algorithms and cheminformatics

    AU - Fjell, Christopher D.

    AU - Jenssen, Håvard

    AU - Cheung, Warren A.

    AU - Hancock, Robert E. W.

    AU - Cherkasov, Artem

    PY - 2011

    Y1 - 2011

    N2 - Pathogens resistant to available drug therapies are a pressing global health problem. Short, cationic peptides represent a novel class of agents that have lower rates of drug resistance than derivatives of current antibiotics. Previously, we created a software system utilizing artificial neural networks that were trained on quantitative structure-activity relationship descriptors calculated for a total of 1400 synthetic peptides for which antibacterial activity was determined. Using the trained system, we correctly identified additional peptides with activity of 94% accuracy; active peptides were 47 of the top rated 50 peptides chosen from an in silico library of nearly 100 000 sequences. Here, we report a method of generating candidate peptide sequences using the heuristic evolutionary programming method of genetic algorithms (GA), which provided a large (19-fold) improvement in identification of novel antibacterial peptides. Approximately 0.50% of peptides evaluated during the GA method were classified as highly active, while only 0.026% of the nearly 100 000 sequences we previously screened were classified as highly active. A selection of these peptides was tested in vitro and activities reported here. While GA significantly improves the possibility of identifying candidate peptides, we encountered important pitfalls to this method that should be considered when using GA.

    AB - Pathogens resistant to available drug therapies are a pressing global health problem. Short, cationic peptides represent a novel class of agents that have lower rates of drug resistance than derivatives of current antibiotics. Previously, we created a software system utilizing artificial neural networks that were trained on quantitative structure-activity relationship descriptors calculated for a total of 1400 synthetic peptides for which antibacterial activity was determined. Using the trained system, we correctly identified additional peptides with activity of 94% accuracy; active peptides were 47 of the top rated 50 peptides chosen from an in silico library of nearly 100 000 sequences. Here, we report a method of generating candidate peptide sequences using the heuristic evolutionary programming method of genetic algorithms (GA), which provided a large (19-fold) improvement in identification of novel antibacterial peptides. Approximately 0.50% of peptides evaluated during the GA method were classified as highly active, while only 0.026% of the nearly 100 000 sequences we previously screened were classified as highly active. A selection of these peptides was tested in vitro and activities reported here. While GA significantly improves the possibility of identifying candidate peptides, we encountered important pitfalls to this method that should be considered when using GA.

    KW - antibacterial peptides

    KW - cheminformatics

    KW - genetic algorithms

    KW - quantitative structure-activity relationship

    U2 - 10.1111/j.1747-0285.2010.01044.x

    DO - 10.1111/j.1747-0285.2010.01044.x

    M3 - Journal article

    VL - 77

    SP - 48

    EP - 56

    JO - Chemical Biology and Drug Design (Print)

    JF - Chemical Biology and Drug Design (Print)

    SN - 1747-0277

    IS - 1

    ER -