GenTχWarper is an application tool for alignment, analysis and mining of gene expression time series. The core algorithm is based on dynamic time warping techniques used in the speech recognition field⁽¹⁾. These techniques allow for non-linear (elastic) alignment of temporal sequences of feature vectors and consequently enable detection of similar shapes with different phases, as demonstrated on the figure below.

To get started with some examples, please read the tutorial section.

⁽¹⁾ Sakoe,H. and Chiba,S. Dynamic programming algorithm optimization for spoken word recognition. IEEE Trans. on Acoust., Speech, and Signal Process., ASSP 26, 43-49 (1978).