Using machine learning algorithms to detect cellular stress of Listeria monocytogenes from cDNA microarray data
Microarrays are useful tools for measuring gene expression, and can be used to determine if a cell population undergoes specific biological processes. Machine learning algorithms can use a dataset derived from microarrays to learn a classifier that can later identify if a novel cell population is involved in this proposed biological process. While these algorithms [including Bayesian Net, J48 Decision Tree, Random Forest and Support Vector Machine (SVM)] are often used to classify eukaryote microarray experiments, this study focuses on a prokaryotic application. Our objectives were to explore if any machine learning algorithm can learn a classifier that can predict whether a L. monocytogenes population is under stress from an antimicrobial. One task was to distinguish cefuroxime treated versus untreated L. monocytogenes EGE-e, based on the expression level (represented as fluorescence intensity) for each gene from 32 samples (GEO accession GPL14687). The other task was to distinguish between L. monocytogenes 08-5923 treated with carnocyclin A and untreated L. monocytogenes 08-5923, based on expression level of 15 selected genes that were ≥ 2-fold up or down-regulated in the presence of carnocyclin A. We selected features using in-fold feature selection. Results showed that J48 Decision Tree was the most accurate algorithm for predicting cefuroxime stress (96.9% accuracy with leave-one-out cross validation), and both the J48 Decision Tree and Bayesian Network were equally effective for predicting whether L. monocytogenes was under stress from carnocyclin A (90.0% accuracy with 5-fold cross validation). This work demonstrated that Bayesian Nets and J48 Decision Tree could be applied to detect the presence of cellular stress in prokaryotes using data from DNA microarrays.
Citation
X. Liu,
P. Miller,
U. Basu,
N. Asgarian,
R. Greiner,
L. McMullen.
"Using machine learning algorithms to detect cellular stress of Listeria monocytogenes from cDNA microarray data". RiboWest, pp n/a, June 2012.
| Keywords: |
Machine learning, microarray, gene expression, Listeria |
| Category: |
In Conference |
BibTeX
@incollection{Liu+al:12,
author = {Xiaoji Liu and Petr Miller and Urmila Basu and Nasimeh Asgarian and
Russ Greiner and Lynn M. McMullen},
title = {Using machine learning algorithms to detect cellular stress of
Listeria monocytogenes from cDNA microarray data},
Pages = {n/a},
booktitle = {RiboWest},
year = 2012,
}Last Updated: February 12, 2020
Submitted by Sabina P 
