Comparelcms simengine

From Eigenvector Research Documentation Wiki
Revision as of 14:24, 3 September 2008 by imported>Jeremy (Importing text file)
Jump to navigation Jump to search

Purpose

Select variables that are different between related data sets, e.g. mass chromatograms from LC/MS data of different batches.

Synopsis

y=comparelcms_simengine(data,filter_width)

Description

COMPARELCMS_SIMENGINE determines which variables are different between different data sets. For example, after applying coda_dw to LC/MS data sets of highly related samples, such as the data of a good and a bad batch, the results will be very similar. comparelcms_engine takes the next step and extracts the mass chromatograms that are different. This function is normally not called by itself but by the function comparelcms_sim_interactive. The input argument data is a data cube with mode 1 the number of samples, mode two the number of spectra and mode 3 the number of variables, The optional input argument filter_width is used to smooth the columns of the data set in order to minimize the effect of small shifts, The output argument y contains the similarity indices of the variables. Variables with a low similarity index show the differences between the data sets.

Examples

Determination of similarity indices with a filter of 7 data points.

y=comparelcms_simengine(data,7)

Algorithm

The calculations are based on a similarity index of the minimum of the chromatograms (across the samples) and the maximum of the chromatograms.

See Also

comparelcms_sim_interactive