|
Activity Number:
|
248
- Machine Learning in Science and Industry
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 29, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #306479
|
|
|
Title:
|
A Novel Method for Evaluating Co-Dependencies of Phenotypic Susceptibility to Multiple Antimicrobials Within and Between Bacterial Species in an Ecological Niche
|
|
Author(s):
|
Heman Shakeri* and Victoriya Volkova and Majid Jaberi-Douraki
|
|
Companies:
|
Kansas State University and Kansas State University and Kansas State University
|
|
Keywords:
|
Antimicrobial resistance;
Minimum inhibitory concentration;
censored data;
Hypothesis testing;
Markov fields;
Temporal patterns
|
|
Abstract:
|
In this work, we focus on better understanding the antimicrobial resistance and its complex nature through studying the joint distributions of measured minimum inhibitory concentration (MIC) for a set of agents in different hosts--MIC is an interval-censored quantity. Patterns of co-dependencies of phenotypic susceptibility (CDS) combined with the existing knowledge enable us to make predictions and improve antimicrobial use policies. Networks obtained from these CDS patterns are distilled from data and help us strengthen the existing hypothesis and also design new ones. Moreover, we study the role of resistance genes that can confer phenotypic antimicrobial resistance to further understand the genotypic sides of the problem. We use the patterns of antimicrobial resistance in MDR isolates with the presence of resistance genes that expressed in these isolates to look into multi-drug resistance mechanisms and the confounding effects of resistance genes to complement our CDS analysis. Our consolidated findings have implications in the use of antibiotics for health organizations and food processing units.
|
Authors who are presenting talks have a * after their name.
