Premium
Efficiency of Functional Regression Estimators for Combining Multiple Laser Scans of cDNA Microarrays
Author(s) -
Glasbey C. A.,
Khondoker M. R.
Publication year - 2009
Publication title -
biometrical journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.108
H-Index - 63
eISSN - 1521-4036
pISSN - 0323-3847
DOI - 10.1002/bimj.200710444
Subject(s) - estimator , gaussian , multivariate statistics , mathematics , statistics , dna microarray , regression , pattern recognition (psychology) , regression analysis , algorithm , expression (computer science) , computer science , computational biology , artificial intelligence , biology , gene , gene expression , genetics , physics , quantum mechanics , programming language
The first stage in the analysis of cDNA microarray data is estimation of the level of expression of each gene, from laser scans of hybridised microarrays. Typically, data are used from a single scan, although, if multiple scans are available, there is the opportunity to reduce sampling error by using all of them. Combining multiple laser scans can be formulated as multivariate functional regression through the origin. Maximum likelihood estimation fails, but many alternative estimators exist, one of which is to maximise the likelihood of a Gaussian structural regression model. We show by simulation that, surprisingly, this estimator is efficient for our problem, even though the distribution of gene expression values is far from Gaussian. Further, it performs well if errors have a heavier tailed distribution or the model includes intercept terms, but not necessarily in other regions of parameter space. Finally, we show that by combining multiple laser scans we increase the power to detect differential expression of genes. (© 2009 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)