Prediction of skin penetration using machine learning methods

Yi Sun, Gary Moss, M. Prapopoulou, Roderick Adams, Marc Brown, N. Davey

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    5 Citations (Scopus)
    164 Downloads (Pure)

    Abstract

    Improving predictions of the skin permeability coefficient is a difficult problem. It is also an important issue with the increasing use of skin patches as a means of drug delivery. In this work, we applyK-nearest-neighbour regression, single
    layer networks, mixture of experts and Gaussian processes to predict the permeability coefficient. We obtain a considerable improvement over the quantitative structureactivity relationship (QSARs) predictors. We show that using
    five features, which are molecular weight, solubility parameter, lipophilicity, the number of hydrogen bonding acceptor and donor groups, can produce better predictions than the one using only lipophilicity and the molecular weight. The Gaussian process regression with five compound features gives the best performance in this work.
    Original languageEnglish
    Title of host publicationProcs of the 8th IEEE International Conference on Data Mining
    Subtitle of host publication(ICDM'08)
    EditorsD Gunopulos, F Turini, C Zaniolo, N Ramakrishnan, XD Wu
    PublisherInstitute of Electrical and Electronics Engineers (IEEE)
    Pages1049-1054
    Number of pages6
    ISBN (Print)978-0-7695-3502-9
    DOIs
    Publication statusPublished - 2008
    Event8th IEEE International Conference on Data Mining - Pisa
    Duration: 15 Dec 200819 Dec 2008

    Conference

    Conference8th IEEE International Conference on Data Mining
    CityPisa
    Period15/12/0819/12/08

    Keywords

    • STRUCTURE-PERMEABILITY RELATIONSHIPS
    • PERCUTANEOUS-ABSORPTION

    Fingerprint

    Dive into the research topics of 'Prediction of skin penetration using machine learning methods'. Together they form a unique fingerprint.

    Cite this