Neural Information Processing - Letters and Reviews

Vol. 12, Nos. 1-3, January-March 2008


pp. 1-10


Automatic Factorization of Biological Signals Measured by Fluorescence Correlation Spectroscopy using Non-negative Matrix Factorization


Kenji Watanabe

Department of Computer Science, Graduate School of Systems and Informatoin Engineering,

University of Tsukuba,

1-1-1 Tennodai, Tsukuba-shi, Ibaraki-ken, 305-8577 Japan



Takio Kurita

National Institute of Advanced Industrial Science and Technology (AIST),

AIST Central 2, 1-1-1 Umezono, Tsukuba-shi, Ibaraki-ken, 305-8568 Japan




This paper proposes an automatic factorization method of the biological signals measured by Fluorescence Correlation Spectroscopy (FCS). Since the signals are composed from several positive components, the signals are decomposed by using the idea of Nonnegative matrix factorization (NMF). Each component is approximated by a model function and the signals are factorized as the non-negative sum of a few model functions. Analytical accuracy of the proposed method was verified by using biological data that were measured by FCS. The experimental results showed that the proposed method could automatically factorize the signals and could succeed to obtain the similar results with the manual investigations.


Keywords − Signal processing, NMF, Pattern recognition, Protein dynamics