annotate nxn_clustering.py @ 5:57a1a58056a6 draft

planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit d786052cd04f8b25eb4aff80b1b9724f62031b61
author bgruening
date Sat, 20 May 2017 12:57:06 -0400
parents 70b071de9bee
children 0d88631bb7de
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
2
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
1 #!/usr/bin/env python
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
2 """
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
3 Modified version of code examples from the chemfp project.
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
4 http://code.google.com/p/chem-fingerprints/
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
5 Thanks to Andrew Dalke of Andrew Dalke Scientific!
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
6 """
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
7 import matplotlib
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
8 matplotlib.use('Agg')
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
9 import argparse
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
10 import os
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
11 import chemfp
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
12 import scipy.cluster.hierarchy as hcluster
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
13 import pylab
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
14 import numpy
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
15
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
16 def distance_matrix(arena, tanimoto_threshold = 0.0):
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
17 n = len(arena)
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
18 # Start off a similarity matrix with 1.0s along the diagonal
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
19 try:
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
20 similarities = numpy.identity(n, "d")
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
21 except:
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
22 raise Exception('Input dataset is to large!')
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
23 chemfp.set_num_threads( args.processors )
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
24
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
25 ## Compute the full similarity matrix.
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
26 # The implementation computes the upper-triangle then copies
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
27 # the upper-triangle into lower-triangle. It does not include
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
28 # terms for the diagonal.
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
29 results = chemfp.search.threshold_tanimoto_search_symmetric(arena, threshold=tanimoto_threshold)
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
30
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
31 # Copy the results into the NumPy array.
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
32 for row_index, row in enumerate(results.iter_indices_and_scores()):
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
33 for target_index, target_score in row:
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
34 similarities[row_index, target_index] = target_score
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
35
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
36 # Return the distance matrix using the similarity matrix
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
37 return 1.0 - similarities
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
38
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
39
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
40 if __name__ == "__main__":
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
41 parser = argparse.ArgumentParser(description="""NxN clustering for fps files.
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
42 For more details please see the chemfp documentation:
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
43 https://chemfp.readthedocs.org
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
44 """)
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
45
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
46 parser.add_argument("-i", "--input", dest="input_path",
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
47 required=True,
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
48 help="Path to the input file.")
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
49
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
50 parser.add_argument("-c", "--cluster", dest="cluster_image",
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
51 help="Path to the output cluster image.")
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
52
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
53 parser.add_argument("-s", "--smatrix", dest="similarity_matrix",
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
54 help="Path to the similarity matrix output file.")
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
55
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
56 parser.add_argument("-t", "--threshold", dest="tanimoto_threshold",
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
57 type=float, default=0.0,
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
58 help="Tanimoto threshold [0.0]")
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
59
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
60 parser.add_argument("--oformat", default='png', help="Output format (png, svg)")
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
61
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
62 parser.add_argument('-p', '--processors', type=int,
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
63 default=4)
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
64
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
65 args = parser.parse_args()
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
66
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
67 targets = chemfp.open( args.input_path, format='fps' )
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
68 arena = chemfp.load_fingerprints( targets )
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
69 distances = distance_matrix( arena, args.tanimoto_threshold )
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
70
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
71 if args.similarity_matrix:
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
72 distances.tofile( args.similarity_matrix )
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
73
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
74 if args.cluster_image:
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
75 linkage = hcluster.linkage( distances, method="single", metric="euclidean" )
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
76
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
77 hcluster.dendrogram(linkage, labels=arena.ids)
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
78
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
79 pylab.savefig( args.cluster_image, format=args.oformat )
70b071de9bee planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/chemfp commit 01da22e4184a5a6f6a3dd4631a7b9c31d1b6d502
bgruening
parents:
diff changeset
80