Normalize Compositional Distance

[NCD,DP] = NCD(files) calculates the Normalized Compositional Difference (NCD) (aka Normalized Difference) between compositions and the Differentiation Potential (DP) between groups of compositions.

INPUTS:

NCD(files) is a 1 by nf cell array of the input file paths.
If files is not specified they must be selected manually
using a file open interface dialog.

Compositions are given in CSV files as row-ordered arrays, i.e. 1 composition per row.

All compositions must be vectors of the same length, i.e. all arrays have the same number of columns. The number of compositions (i.e. rows) can vary between files and is not limited.

Groups of compositions are stored in seperate CSV files, each with an identical number of columns.

The contents of CSV files should be numeric and contain no characters.

OUTPUTS

Results are written to 'NCD_analysis.csv', containing:

1. Mean, median and standard deviation of NCD between compositions in each file (group).

2. Median NCD between files (groups), DP scores and significance at 95% confidence level (1 == pass, 0 == fail). The median NCD between files is calculated pairwise, so that each composition in File A is compared to each composition in File B.