QT Clustering using Euclidean Distance

Implement QT Clustering using Euclidean Distance

Description

QT clustering algorithm as described in:

Heyer, L. J., Kruglyak, S., Yooseph, S. (1999). Exploring expression data: Identification and analysis of coexpressed genes. Genome Research 9, 1106–1115.

http://genome.cshlp.org/content/9/11/1106.full
http://genome.cshlp.org/content/9/11/1106/F5.large.jpg

if two sets A{i} have same cardinality, we pick first one
our distance measure is Euclidean distance

input:
G-nxp data to cluster
d-diameter threshold
D-Euclidean distance for all 0<i<j<=n

output:
idx-nx1 vector containing cluster indices

Misha Koshelev
January 20th, 2009
Montague Laboratory
    
Advantage: always returns same result.

• MATLAB release: MATLAB 7.6 (R2008a)

[Aadarsh]

No matter what I put as D, I keep getting the following error message:
??? Attempted to access D(2,1); index out of bounds because numel(D)=1.

Error in ==> qtclusteuclid at 51
                jdiam(pidx)=max(D(pts(pidx),A));

I am trying to use this function with the following call:
qtclusteuclid(new,5,4)

where new is a matrix of size 2x548.

It still gives the same error message if i flip the matrix, and also if i change the parameters for d and D.

Can you suggest how I can get it to work?

Thanks