Location: Genomics and Bioinformatics Research
Project Number: 6066-21310-005-032-A
Project Type: Cooperative Agreement
Start Date: Sep 2, 2018
End Date: Sep 1, 2020
Objective:
Agriculture production is always under the threat of introduced or changing pests. Within the U.S.A. agricultural system, these threats are even greater due to our large and concentrated production areas be they animals or crops, the total output of world agricultural production needs to be protected. Disease, especially foreign animal disease and non-native pests can be extremely debilitating to production systems due to the lack of natural resistance or native predators that can control the pests. Also agronomic practices that can mitigate the problem are often not developed or put into practice before the disease or pest becomes a major issue and threatens agricultural production.
Equally bad are those pathogens that are native to the production area. With domestic diseases and pests, the natural occurrence of mutations that can enhance their pathogenicity can occur. Natural environmental events or bad agronomic practices can cause population shifts that can increase more pathogenic subpopulations to increase and negatively impact agricultural production. The threat from these introductions can be long or quick in development.
Genomic tools allow for the understanding of the pests and their increased pathogenicity. The same tools can be used for early detection and identification of pests or characterization of populations. Genomics is used to discover the location and subsequent transmission pathway of a pathogen outbreak. Proteomics can be similarly used for some analysis. However, the study of the disease/pest is only half the issue. The development of superior cultivars or genetic stock that can resist or overcome these diseases/pests or other limitations is also required.
To fully utilize these technologies computational analysis of Big Data is needed. In turn, the techniques need access to high performance computing (HPC) for complex data analysis and modeling. Also, computational tools need to run optimally on an HPC environment and this takes specialized development. All of this needs to be developed and put into place with training for proper analysis.
Approach:
USDA has extensive experience researching and controlling diseases/pest, especially foreign diseases and non-native invasive pests. The USDA also leads in the development of genomic tools for most agricultural production systems. Mississippi State University (MSU) has expertise in the operations of high performance computing (HPC) and developing/altering computational software to run in an HPC environment. MSU also has expertise in bioinformatic analysis related to agriculture. In this project, ARS and MSU will work together to develop, implement, and utilize HPC and will work together to analyze Big Data generated by ARS. Types of analysis include genomics, bioinformatics, proteomics, modeling and machine learning (also known as artificial intelligence). These types of analysis will require a non-standard HPC configuration of high memory nodes, normal nodes, graphical processing unit (GPU) nodes and data storage.
