Assessment of data transformations for model-based clustering of RNA-Seq data

Janelle R. Noel-MacDonnell, Joseph Usset, Ellen L Goode, Brooke L. Fridley

Research output: Contribution to journalArticle

Abstract

Quality control, global biases, normalization, and analysis methods for RNA-Seq data are quite different than those for microarray-based studies. The assumption of normality is reasonable for microarray based gene expression data; however, RNA-Seq data tend to follow an over-dispersed Poisson or negative binomial distribution. Little research has been done to assess how data transformations impact Gaussian model-based clustering with respect to clustering performance and accuracy in estimating the correct number of clusters in RNASeq data. In this article, we investigate Gaussian model-based clustering performance and accuracy in estimating the correct number of clusters by applying four data transformations (i.e., naïve, logarithmic, Blom, and variance stabilizing transformation) to simulated RNASeq data. To do so, an extensive simulation study was carried out in which the scenarios varied in terms of: how genes were selected to be included in the clustering analyses, size of the clusters, and number of clusters. Following the application of the different transformations to the simulated data, Gaussian model-based clustering was carried out. To assess clustering performance for each of the data transformations, the adjusted rand index, clustering error rate, and concordance index were utilized. As expected, our results showed that clustering performance was gained in scenarios where data transformations were applied to make the data appear "more" Gaussian in distribution.

Original languageEnglish (US)
Article numbere0191758
JournalPLoS One
Volume13
Issue number2
DOIs
StatePublished - Feb 1 2018

Fingerprint

Cluster Analysis
RNA
Microarrays
Gene expression
quality control
Quality control
Genes
gene expression
Binomial Distribution
genes
Normal Distribution
Quality Control
methodology
Gene Expression
Research
normal distribution
binomial distribution

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

Assessment of data transformations for model-based clustering of RNA-Seq data. / Noel-MacDonnell, Janelle R.; Usset, Joseph; Goode, Ellen L; Fridley, Brooke L.

In: PLoS One, Vol. 13, No. 2, e0191758, 01.02.2018.

Research output: Contribution to journalArticle

Noel-MacDonnell, Janelle R. ; Usset, Joseph ; Goode, Ellen L ; Fridley, Brooke L. / Assessment of data transformations for model-based clustering of RNA-Seq data. In: PLoS One. 2018 ; Vol. 13, No. 2.
@article{dd4b12b320234c9c99c0737b894ddb97,
title = "Assessment of data transformations for model-based clustering of RNA-Seq data",
abstract = "Quality control, global biases, normalization, and analysis methods for RNA-Seq data are quite different than those for microarray-based studies. The assumption of normality is reasonable for microarray based gene expression data; however, RNA-Seq data tend to follow an over-dispersed Poisson or negative binomial distribution. Little research has been done to assess how data transformations impact Gaussian model-based clustering with respect to clustering performance and accuracy in estimating the correct number of clusters in RNASeq data. In this article, we investigate Gaussian model-based clustering performance and accuracy in estimating the correct number of clusters by applying four data transformations (i.e., na{\"i}ve, logarithmic, Blom, and variance stabilizing transformation) to simulated RNASeq data. To do so, an extensive simulation study was carried out in which the scenarios varied in terms of: how genes were selected to be included in the clustering analyses, size of the clusters, and number of clusters. Following the application of the different transformations to the simulated data, Gaussian model-based clustering was carried out. To assess clustering performance for each of the data transformations, the adjusted rand index, clustering error rate, and concordance index were utilized. As expected, our results showed that clustering performance was gained in scenarios where data transformations were applied to make the data appear {"}more{"} Gaussian in distribution.",
author = "Noel-MacDonnell, {Janelle R.} and Joseph Usset and Goode, {Ellen L} and Fridley, {Brooke L.}",
year = "2018",
month = "2",
day = "1",
doi = "10.1371/journal.pone.0191758",
language = "English (US)",
volume = "13",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "2",

}

TY - JOUR

T1 - Assessment of data transformations for model-based clustering of RNA-Seq data

AU - Noel-MacDonnell, Janelle R.

AU - Usset, Joseph

AU - Goode, Ellen L

AU - Fridley, Brooke L.

PY - 2018/2/1

Y1 - 2018/2/1

N2 - Quality control, global biases, normalization, and analysis methods for RNA-Seq data are quite different than those for microarray-based studies. The assumption of normality is reasonable for microarray based gene expression data; however, RNA-Seq data tend to follow an over-dispersed Poisson or negative binomial distribution. Little research has been done to assess how data transformations impact Gaussian model-based clustering with respect to clustering performance and accuracy in estimating the correct number of clusters in RNASeq data. In this article, we investigate Gaussian model-based clustering performance and accuracy in estimating the correct number of clusters by applying four data transformations (i.e., naïve, logarithmic, Blom, and variance stabilizing transformation) to simulated RNASeq data. To do so, an extensive simulation study was carried out in which the scenarios varied in terms of: how genes were selected to be included in the clustering analyses, size of the clusters, and number of clusters. Following the application of the different transformations to the simulated data, Gaussian model-based clustering was carried out. To assess clustering performance for each of the data transformations, the adjusted rand index, clustering error rate, and concordance index were utilized. As expected, our results showed that clustering performance was gained in scenarios where data transformations were applied to make the data appear "more" Gaussian in distribution.

AB - Quality control, global biases, normalization, and analysis methods for RNA-Seq data are quite different than those for microarray-based studies. The assumption of normality is reasonable for microarray based gene expression data; however, RNA-Seq data tend to follow an over-dispersed Poisson or negative binomial distribution. Little research has been done to assess how data transformations impact Gaussian model-based clustering with respect to clustering performance and accuracy in estimating the correct number of clusters in RNASeq data. In this article, we investigate Gaussian model-based clustering performance and accuracy in estimating the correct number of clusters by applying four data transformations (i.e., naïve, logarithmic, Blom, and variance stabilizing transformation) to simulated RNASeq data. To do so, an extensive simulation study was carried out in which the scenarios varied in terms of: how genes were selected to be included in the clustering analyses, size of the clusters, and number of clusters. Following the application of the different transformations to the simulated data, Gaussian model-based clustering was carried out. To assess clustering performance for each of the data transformations, the adjusted rand index, clustering error rate, and concordance index were utilized. As expected, our results showed that clustering performance was gained in scenarios where data transformations were applied to make the data appear "more" Gaussian in distribution.

UR - http://www.scopus.com/inward/record.url?scp=85042775636&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042775636&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0191758

DO - 10.1371/journal.pone.0191758

M3 - Article

C2 - 29485993

AN - SCOPUS:85042775636

VL - 13

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 2

M1 - e0191758

ER -