Blind Source Separation and the Analysis of Microarray Data

Abstract

We develop an approach for the exploratory analysis of gene expression data, based upon blind source separation techniques. This approach exploits higher-order statistics to identify a linear model for (logarithms of) expression profiles, described as linear combinations of "independent sources." As a result, it yields "elementary expression patterns" (the "sources"), which may be interpreted as potential regulation pathways. Further analysis of the so-obtained sources show that they are generally characterized by a small number of specific coexpressed or antiexpressed genes. In addition, the projections of the expression profiles onto the estimated sources often provides significant clustering of conditions. The algorithm relies on a large number of runs of "independent component analysis" with random initializations, followed by a search of "consensus sources." It then provides estimates for independent sources, together with an assessment of their robustness. The results obtained on two datasets (namely, breast cancer data and Bacillus subtilis sulfur metabolism data) show that some of the obtained gene families correspond to well known families of coregulated genes, which validates the proposed approach.

Blind source separation methods for deconvolution of complex signals in cancer biology

“…For the particular case of breast cancer, independent components extracted for gene expression profiles of breast cancer cell lines were characterized in [26] for cell lines and in [27] for tumors. In [28], several ICA algorithms were tested on 6 breast cancer datasets and the components were systematically interpreted using pathway and Gene Ontology annotations.…”

Blind Principles Based Interference and Noise Reduction Schemes for OFDM

“…An adaptive BSS algorithm entirely based on second order statistics which can run in offline or online [12] is given as the solution. A statistical analysis of microarray data using BSS theories for identifying "elementary" independent expression patterns is unveiled in [13]. In addition to all those applications, basic techniques and formulas that can be tested for specific applications are also widely available [14].…”

Comments on selected fundamental aspects of microarray analysis

“…(a) The fact that the data often consist of a very large amount of small values and a few, extreme points, something which affects most data analysis techniques strongly (Chiappetta et al, 2004). (b) Some effects being studied may have a multiplicative behaviour.…”

“…2), whilst for the second problem taking the log is preferable (Chiappetta et al, 2004;Hoyle et al, 2002;Thygesen and Zwinderman, 2004;Tusher et al, 2001). Fig.…”

“…The figure shows that a spreadsheet (MS-Excel) attributes the same "weight", or importance, to each experimental condition, whilst PCA and ICA do not. in absolute terms: the algorithm gets stuck with a local maximum (Chiappetta et al, 2004). Launching ICA a large number of times, typically 100, circumvents this problem and only those directions or solutions that have been most frequently found are kept.…”