TAMO.Clustering.UPGMA | index /home/David_Gordon/docs/TAMO/Clustering/UPGMA.py |
This module uses a variant of the UPGMA algorithm for hierarchical clustering to
organize motifs into groups. The "A" step (averaging) is performed by avergaing the
probability values in the aligned PSSMs within a cluster.
1) Initialization
1.1) Assign each motif to a cluster
1.2) Compute Dmat of all clusters
2) Iteration
2.1) Find the i and j with the smallest Distance
2.2) Create a new cluster (ij) which has n_i + n_j members
2.3) "Connect" i and j to (ij) and give each of the branchs D_ij/2 (better distance?)
2.4) Compute the distance from (ij) to all other clusters (except i and j)
2.5) Delete columns/rows in Dmat corresponding to i and j
Copyright (2005) Whitehead Institute for Biomedical Research (except as noted below)
All Rights Reserved
Author: David Benjamin Gordon
Modules | ||||||
|
Functions | ||
|
Data | ||
ACGT = ['A', 'C', 'G', 'T'] DFUNC = None DMAX = 0.20000000000000001 GLOBALS = {} rcmemo = {} |