Dataset Information

Removing System Noise from Comparative Genomic Hybridization Data by Self-Self Analysis

ABSTRACT: To minimize the distortion of genetic signal by system noise, we have explored the latter in an archive of hybridizations in which no genetic signal is expected. This archive is obtained by comparative genomic hybridization (CGH) of a reference sample in one channel to the same sample in the other channel, which we have termed ‘self-self’ data. We show that these self-self hybridizations trap a variety of system noise inherent in sample-reference (test) data. Through singular value decomposition (SVD) of self-self data, we are able to determine the principal components of system noise. Assuming simple linear models of noise generation, we present evidence that the linear correction of test data with self-self data—which we call system normalization—reduces local and long-range correlations as well as improves signal-to-noise metrics, yet does not introduce detectable spurious signal.

ORGANISM(S): Homo sapiens

PROVIDER: GSE23682 | GEO | 2011/06/09

SECONDARY ACCESSION(S): PRJNA130799

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:Comparative genomic hybridization microarrays (array CGH or molecular karyotyping) for the detection of congenital chromosomal aberrations is the application of microarray technology that is coming fastest into routine clinical application. When using a two-channel microarray of genomic DNA probes for array CGH, the basic setup consists in hybridizing a patient sample against a normal reference sample and detecting copy number variations through the deviation of fluorescent signal intensity between patient and normal reference. Two major disadvantages of this setup are (1) the use of half of the resources to measure a (little informative) reference sample and (2) the possibility that deviating signals are caused by benign copy number variation in the ânormalâ reference instead of a patient aberration. We therefore propose a new experimental loop design that compares three patients in three hybridizations (Patient 1 vs. Patient 3, Patient 3 vs. Patient 2, and Patient 2 vs. Patient 1). We develop and compare two statistical methods (linear models of log ratios and mixed models of absolute measurements). In an analysis of data from 27 patients seen at our genetics center, this new setup together with the linear model analysis significantly overcomes the limitations of the classical setup. Furthermore, we observed that the linear models of the log-ratios had a higher signal-to-noise ratio than the mixed models of the absolute intensities. These improvements are important to guarantee a maximal efficiency of array CGH in a clinical setting and will therefore contribute to its quick adoption as a routine diagnostic tool. The method is implemented as a web application and is available at www.esat.kuleuven.be/loop. Keywords: comparative genomic hybridization The data set consists out of nine loop designs or 27 patients with mental retardation (MR) and multiple congenital anomalies (MCA). The patients were seen at our genetics center (Center for Human Genetics, U.Z.Leuven).

Dataset Information

Removing System Noise from Comparative Genomic Hybridization Data by Self-Self Analysis

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure