DesignNotes for new RKColocal ImageJ Plugin

Welcome to the RKColocal wiki!

This is a place for us to gather our thoughts on design / layout of the new ImageJ(2) RKColocal plugin.

General Plugin Structure

INPUT: two images, same size (perhaps give user option to choose two channels?); 'other parameters'

PRE-PROCESSING: Costes AutoThresholdRegression (method from Coloc2), before Pearsons/Manders/others? - should it be mandatory?... ie) default is to run Costes AutoThresholdRegression unless user applies different threshold method? - so also perhaps offer other thresholding methods as well (should we include manual settings?)

COLOC ALGORITHMS: ICQ; KendallTauBRank; Manders; Costes-Pearsons; MTKT ...

BLOCK SHUFFLE to calculate P-VALUE

OUTPUT: calculated coloc measures (both with and without thresholding?); p-values

TO DO

Ellen

1. Continue to write MTKT test Op
2. Update AutoThreshold Op ... fix tests
3. Update Pearsons methods to optionally include threshold... also, rename as they are Costes methods...

Random Thoughts

1. Provide users opportunity to pre-process images... most importantly, Costes AutoThreshold method (as in Coloc2) before Pearsons and Manders calculations...
2. Include warnings... ex) did auto thresh method work, is Pearsons<thresholds close to zero? ex) intercept far from 0? ex) regression slope far from 1 (could cause numeral problems in auto threshold)?