Two-group comparisons of zero-inflated intensity values: the choice of test statistic matters

A special characteristic of data from molecular biology is the frequent occurrence of zero intensity values which can arise either by true absence of a compound or by a signal that is below a technical limit of detection. While so-called two-part tests compare mixture distributions between groups, o...

Full description

Saved in:
Bibliographic Details
Published in:Bioinformatics 2015-07, Vol.31 (14), p.2310-2317
Main Authors: Gleiss, Andreas, Dakna, Mohammed, Mischak, Harald, Heinze, Georg
Format: Article
Language:English
Subjects:
Bioinformatics
Computer programs
Computer simulation
Estimates
Likelihood Functions
Likelihood ratio
Models, Statistical
Molecular biology
Peptides - metabolism
Software
Statistical tests
Statistics, Nonparametric
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A special characteristic of data from molecular biology is the frequent occurrence of zero intensity values which can arise either by true absence of a compound or by a signal that is below a technical limit of detection. While so-called two-part tests compare mixture distributions between groups, one-part tests treat the zero-inflated distributions as left-censored. The left-inflated mixture model combines these two approaches. Both types of distributional assumptions and combinations of both are considered in a simulation study to compare power and estimation of log fold change. We discuss issues of application using an example from peptidomics.The considered tests generally perform best in scenarios satisfying their respective distributional assumptions. In the absence of distributional assumptions, the two-part Wilcoxon test or the empirical likelihood ratio test is recommended. Assuming a log-normal subdistribution the left-inflated mixture model provides estimates for the proportions of the two considered types of zero intensities. R code is available at http://cemsiis.meduniwien.ac.at/en/kb/science-research/software/
ISSN:1367-4803
1367-4811
1460-2059
DOI:10.1093/bioinformatics/btv154