Senior Design Projects

ECS193 A/B Winter & Spring 2019

Using machine learning to find emerging trends and common patterns in antimicrobial resistance in people, animals and food

Email **********
Name Beatriz Martinez Lopez
Affiliation VM: Medicine & Epidemiology, UC Davis

Project's details

Project title Using machine learning to find emerging trends and common patterns in antimicrobial resistance in people, animals and food
Background The emergence of antimicrobial resistant ( AMR ) pathogens is considered to be a major threat for public health due to the increased incidence of treatment failure and severity of diseases in humans. Moreover, AMR has been associated with devastating consequences in animal health, welfare and production as well as severe losses in agriculture due to the failure to control AMR plant-pathogenic bacteria. Notably, 1 in 5 AMR human infections have been associated with food and animals. Since 1996, the Food and Drug Administration ( FDA ), the Centers for Disease Control and Prevention ( CDC ), and the U.S. Department of Agriculture ( USDA ) have been testing antimicrobial susceptibility and conducting whole genome sequencing of enteric ( intestinal ) bacteria found in ill people ( CDC ), retail meats ( FDA ), and food animals ( USDA ) in the United States. In this project, we propose to use machine learning to better understand AMR patterns and potential spatio-temporal clusters or associations of AMR bacteria found in animals, food and people in the US.
Description Goal: To use machine learning to evaluate spatio-temporal dynamics of antimicrobial resistant bacteria, identify similar resistance patterns in animals, food and humans as a first step to support prioritization and risk-reduction interventions for zoonotic pathogens. The project will make use of an available comprehensive diagnostic database of thousands of food-borne pathogens collected during the last 15 years. Approach: Perform unsupervised learning on the whole genome sequences and/or the resistant genes profile of thousands of bacterial food borne pathogens isolated from animals, food and humans during the last 15 years in the US. Generate a web-based visualization tool to facilitate the data mining and visualization of spatio-temporal changes in resistant genes and facilitate the identification of most important resistant pathogens shared by animals, food and humans.
Deliverable Students are expected to provide a working prototype of the web-based visualization tool ( with the corresponding source code and documentation in Github ).
Skill set desirable the ideal team will have experience in Javascript, HTML/CSS, SQL, Git, JSON
Phone number **********
Client time availability 30-60 min weekly or more
IP requirement Open source project
Attachment N/A
Selected No
Team members N/A
TA N/A