Parametric Time Warping

Parametric Time Warping aligns patterns, i.e. it aims to put corresponding features at the same locations. The algorithm searches for an optimal polynomial describing the warping. It is possible to align one sample to a reference, several samples to the same reference, or several samples to several references. One can choose between calculating individual warpings, or one global warping for a set of samples and one reference. Two optimization criteria are implemented: RMS (Root Mean Square error) and WCC (Weighted Cross Correlation). Both warping of peak profiles and of peak lists are supported. A vignette for the latter is contained in the inst/doc directory of the source package - the vignette source can be found on the package github site.


version 1.9-13:

  • fixed prediction for forward warping (contribution by Sergio Oller)
  • updated author list in DESCRIPTION

version 1.9-12:

  • implementation of ptw for peak lists: stptw and associated functions
  • addition of vignette
  • removal of obsolete function interpol, replaced by approx
  • new maintainer

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


1.9-15 by Ron Wehrens, 2 years ago

Browse source code at

Authors: Jan Gerretzen [ctb] , Paul Eilers [aut] , Hans Wouters [ctb] , Tom Bloemberg [aut] , Ron Wehrens [aut, cre]

Documentation:   PDF Manual  

Task views: Time Series Analysis

GPL (>= 2) license

Imports nloptr, graphics, grDevices, stats

Imported by Rnmr1D, chipPCR, spant.

Depended on by RGCxGC, RpeakChrom.

See at CRAN