DNA microarray data imputation and significance analysis of differential expression

MOTIVATION Significance analysis of differential expression in DNA microarray data is an important task. Much of the current research is focused on developing improved tests and software tools. The task is difficult not only owing to the high dimensionality of the data (number of genes), but also because of the often non-negligible presence of missing values. There is thus a great need to reliably impute these missing values prior to the statistical analyses. Many imputation methods have been developed for DNA microarray data, but their impact on statistical analyses has not been well studied. In this work we examine how missing values and their imputation affect significance analysis of differential expression. RESULTS We develop a new imputation method (LinCmb) that is superior to the widely used methods in terms of normalized root mean squared error. Its estimates are the convex combinations of the estimates of existing methods. We find that LinCmb adapts to the structure of the data: If the data are heterogeneous or if there are few missing values, LinCmb puts more weight on local imputation methods; if the data are homogeneous or if there are many missing values, LinCmb puts more weight on global imputation methods. Thus, LinCmb is a useful tool to understand the merits of different imputation methods. We also demonstrate that missing values affect significance analysis. Two datasets, different amounts of missing values, different imputation methods, the standard t-test and the regularized t-test and ANOVA are employed in the simulations. We conclude that good imputation alleviates the impact of missing values and should be an integral part of microarray data analysis. The most competitive methods are LinCmb, GMC and BPCA. Popular imputation schemes such as SVD, row mean, and KNN all exhibit high variance and poor performance. The regularized t-test is less affected by missing values than the standard t-test. AVAILABILITY Matlab code is available on request from the authors.

[1]  Tommi S. Jaakkola,et al.  Continuous Representations of Time-Series Gene Expression Data , 2003, J. Comput. Biol..

[2]  Russ B. Altman,et al.  Missing value estimation methods for DNA microarrays , 2001, Bioinform..

[3]  David Botstein,et al.  The Stanford Microarray Database: data access and quality assessment tools , 2003, Nucleic Acids Res..

[4]  Pierre Baldi,et al.  A Bayesian framework for the analysis of microarray expression data: regularized t -test and statistical inferences of gene changes , 2001, Bioinform..

[5]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[6]  Gene H. Golub,et al.  Missing value estimation for DNA microarray gene expression data: local least squares imputation , 2005, Bioinform..

[7]  John Quackenbush.  Microarray data normalization and transformation , 2002, Nature Genetics.

[8]  Y. Benjamini,et al.  Controlling the false discovery rate: a practical and powerful approach to multiple testing , 1995 .

[9]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[10]  Rebecka Jörnsten,et al.  Screening anti-inflammatory compounds in injured spinal cord with microarrays: a comparison of bioinformatics analysis approaches. , 2004, Physiological genomics.

[11]  Ming Ouyang,et al.  Gaussian mixture clustering and imputation of microarray data , 2004, Bioinform..

[12]  Shin Ishii,et al.  A Bayesian missing value estimation method for gene expression profile data , 2003, Bioinform..

[13]  X. Cui,et al.  Statistical tests for differential expression in cDNA microarray experiments , 2003, Genome Biology.

[14]  Iqbal Gondal,et al.  Collateral missing value imputation: a new robust missing value estimation algorithm for microarray data , 2005, Bioinform..

[15]  Gene H. Golub,et al.  Missing value estimation for DNA microarray gene expression data: local least squares imputation , 2005, Bioinform..

[16]  D. Botstein,et al.  Gene expression patterns in human liver cancers. , 2002, Molecular biology of the cell.

[17]  T. H. Bø,et al.  LSimpute: accurate estimation of missing values in microarray data with least squares methods. , 2004, Nucleic acids research.

[18]  Xiaobo Zhou,et al.  Missing-value estimation using linear and non-linear regression with Bayesian gene selection , 2003, Bioinform..

[19]  D. Botstein,et al.  Cluster analysis and display of genome-wide expression patterns. , 1998, Proceedings of the National Academy of Sciences of the United States of America.

[20]  A. Raftery,et al.  Model-based Gaussian and non-Gaussian clustering , 1993 .

[21]  S. P. Fodor,et al.  High density synthetic oligonucleotide arrays , 1999, Nature Genetics.

[22]  Alexander Schliep,et al.  Using hidden Markov models to analyze gene expression time course data , 2003, ISMB.