ICA - Independent Component Analysis


ICA and Bioinformatics

ICA - Independent Component Analysis

Matthias Scholz, Ph.D. thesis

Applying ICA is motivated by the idea that the variation in molecular data is generated by divers factors s. These may include internal biological factors as well as external environmental or technical factors. Each observed variable x (e.g., gene) can therefore be seen as derived from a specific combination of these factors. The illustrated factors may represent an increase of temperature (s1), an internal circadian rhythm (s2), and different ecotypes (s3).
With the assumption that the factors are independent of each other, ICA can be applied to a data set X in order to identify the original factors s and the dependencies given by the matrix A.

Bioinformatics publication | matlab code



Resources

Tutorials

Books

Conferences

  • International Workshop on Independent Component Analysis and Blind Signal Separation
    [2000 | 2003 | 2004 | 2006 | 2007 | 2010 ]
  • European Meeting on Independent Component Analysis
    [ 2002 | 2003 ]

People

  • list by Paris Smaragdis
  • list by Allan Kardec Barros

Mailing List

Algorithms

see also: Principal Component Analysis (PCA)

www.matthias-scholz.de