STP1306

    Protein Patterns and Toxicity Identification Using Artificial Neural Network Models

    Published: Jan 1996


      Format Pages Price  
    PDF (316K) 16 $25   ADD TO CART
    Complete Source PDF (8.1M) 16 $100   ADD TO CART


    Abstract

    Proteins extracted, separated, and visualized can provide detailed information about an organism and its environment. We have used an artificial neural network model to identify significant exposures of a cladoceran (Daphnia magnet) to alcohol and pesticides, of a copepod (Eurytemora affinis) to heat and salinity, of an earthworm (Lumbricus terrestris) to sulfur mustard and of a small fish (Oryzias latipes) to groundwater concentrations. The method depends on systematic differences or tendencies in numbers and amounts of proteins present in different treatments or environments. We illustrate how neural computing might be useful in retrieving the information contained in the hundreds or thousands of proteins expressed in test organisms. Such information could apply to prediction of toxicity, identification of toxicity and to characterizing environments in general.

    Keywords:

    protein patterns, toxicity, neural networks


    Author Information:

    Bradley, BP
    Professor, University of Maryland Baltimore County, Baltimore, MD

    Brown, DC
    Research Associate, University of Maryland Baltimore County, Baltimore, MD

    Iamonte, TN
    Graduate Student, University of Maryland Baltimore County, Baltimore, MD

    Boyd, SM
    Graduate Student, University of Maryland Baltimore County, Baltimore, MD

    O'Neill, MC
    Associate Professor, University of Maryland Baltimore County, Baltimore, MD


    Paper ID: STP11700S

    Committee/Subcommittee: E47.09

    DOI: 10.1520/STP11700S


    CrossRef ASTM International is a member of CrossRef.