comparison chemfp_clustering/nxn_clustering.py @ 22:6c496b524b41

ChemicalToolBoX update.

21:7c84cfa515e0

import os
import chemfp
import scipy.cluster.hierarchy as hcluster
import pylab
import numpy
import tempfile


def distance_matrix(arena, tanimoto_threshold = 0.0):
    n = len(arena)
    # Start off a similarity matrix with 1.0s along the diagonal
    try:

        for target_index, target_score in row:
            similarities[row_index, target_index] = target_score

    # Return the distance matrix using the similarity matrix
    return 1.0 - similarities



if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="""NxN clustering for fps files.
For more details please see the chemfp documentation:

    parser.add_argument('-p', '--processors', type=int, 
        default=4)

    args = parser.parse_args()

    # make sure that the file ending is fps
    temp_file = tempfile.NamedTemporaryFile()
    temp_link = "%s.%s" % (temp_file.name, 'fps')
    temp_file.close()
    os.symlink(args.input_path, temp_link)

    arena = chemfp.load_fingerprints( temp_link )
    distances  = distance_matrix( arena, args.tanimoto_threshold )
    linkage = hcluster.linkage( distances, method="single", metric="euclidean" )

    hcluster.dendrogram(linkage, labels=arena.ids)

    pylab.savefig( args.output_path, format=args.oformat )

22:6c496b524b41

import os
import chemfp
import scipy.cluster.hierarchy as hcluster
import pylab
import numpy

def distance_matrix(arena, tanimoto_threshold = 0.0):
    n = len(arena)
    # Start off a similarity matrix with 1.0s along the diagonal
    try:

        for target_index, target_score in row:
            similarities[row_index, target_index] = target_score

    # Return the distance matrix using the similarity matrix
    return 1.0 - similarities


if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="""NxN clustering for fps files.
For more details please see the chemfp documentation:

    parser.add_argument('-p', '--processors', type=int, 
        default=4)

    args = parser.parse_args()

    targets = chemfp.open( args.input_path, format='fps' )
    arena = chemfp.load_fingerprints( targets )
    distances  = distance_matrix( arena, args.tanimoto_threshold )
    linkage = hcluster.linkage( distances, method="single", metric="euclidean" )

    hcluster.dendrogram(linkage, labels=arena.ids)

    pylab.savefig( args.output_path, format=args.oformat )