Mercurial > repos > boris > phylorelatives
view phylorelatives.py @ 31:75dcc9910f6d draft
Uploaded
author | boris |
---|---|
date | Mon, 10 Jun 2013 00:43:11 -0400 |
parents | 29ca7d28f6c7 |
children | 8d663949bdc8 |
line wrap: on
line source
#!/usr/bin/env python # # Boris Rebolledo-Jaramillo (boris-at-bx.psu.edu) #usage: phylorelatives.py [-h] [-i FASTA] [-b INT] [-p] [-r FASTA] # #Constructs relatedness of a set of sequences based on the pairwise proportion #of different sites. It reports the test sequences relatives, tree plot and #tree in Newick format. One or more test sequences are accepted as long as #their name includes the strict suffix "_minor" or "_test" (i.e. >seq1_minor). #IMPORTANT: Sequences must have the same length! # #optional arguments: # -h, --help show this help message and exit # -i FASTA, --input FASTA # This option can be specified multiple times. Sequences # will be added to "multi-fasta.fa". (e.g. -i major1.fa # -i major2.fa -i minor1.fa) # -b INT, --bootstrap INT # Change number of replicas. 0 to deactivate. (Default: # 1000) # -p, --pairwise Use pairwise deletion of gaps/missing data. (Default: # Complete deletion) # -r FASTA, --root FASTA # Root trees using FASTA sequence as outgroup. (Default: # Display unrooted trees) import sys import argparse import array import dendropy import rpy2.rinterface rpy2.rinterface.set_initoptions(('rpy2','--vanilla','--quiet')) import rpy2.robjects as robjects from rpy2.robjects.packages import importr def ape_read_dna(infasta): """Read multi-fasta into phylo object""" ape = importr('ape') data = ape.read_dna(file=infasta,format="fasta") return data def ape_nj(data,missing_info_option=0): """Return ape nj tree""" ape = importr('ape') dist = ape.dist_dna(data,model="raw",pairwise=missing_info_option) nj_tree = ape.nj(dist) return nj_tree def ape_consensus(tree,tree_function,data,iters=1000): """Return majority rule consensus tree""" ape = importr('ape') tree_func = robjects.r(tree_function) bootstrap = ape.boot_phylo(tree, data, tree_func, B=iters, quiet=1, trees=1) bs_trees = bootstrap.rx('trees')[0] cons_tree = ape.consensus(bs_trees,p=0.5) return cons_tree def dendro_relatives(tree,minor): """Return minor allele sequence relatives in tree""" ape = importr('ape') newick = list(ape.write_tree(tree))[0] t = dendropy.Tree.get_from_string(newick,"newick") minor_leaf = [node for node in t.leaf_nodes() if node.get_node_str() == minor][0] parent = minor_leaf.parent_node relatives = [] while len(relatives) == 0: output = [relative.get_node_str() for relative in parent.leaf_nodes()] relatives = [relative for relative in output if not (relative.endswith('minor') or relative.endswith('test'))] parent = parent.parent_node return output def dendro_plot(tree, root=False ): """Plot tree to file in ascii format""" ape = importr('ape') if root: newick = list(ape.write_tree(ape.root(tree,root)))[0] else: newick = list(ape.write_tree(tree))[0] t = dendropy.Tree.get_from_string(newick,"newick") ascii_tree = t.as_ascii_plot() return ascii_tree def write_nwk(tree): "Write proper Newick string" ape = importr('ape') nwk = list(ape.write_tree(tree))[0] return nwk def main(): # Parse command line options parser = argparse.ArgumentParser(description='Constructs relatedness of a set of sequences based on the pairwise proportion of different sites. It reports the test sequences relatives, tree plots and trees in Newick format. One or more test sequences are accepted as long as their name includes the strict suffix "_minor" or "_test" (i.e. >seq1_minor). IMPORTANT: Sequences must have the same length!', epilog='Boris Rebolledo-Jaramillo (boris-at-bx.psu.edu)') parser.add_argument('-i', '--input', metavar='FASTA', action='append', type=str, help='This option can be specified multiple times. Sequences will be added to "multi-fasta.fa". (e.g. -i major1.fa -i major2.fa -i minor1.fa)') parser.add_argument('-b', '--bootstrap', type=int, metavar='INT',default=1000, help='Change number of replicas. 0 to deactivate. (Default: 1000)') parser.add_argument('-p', '--pairwise', action='store_true', help='Use pairwise deletion of gaps/missing data. (Default: Complete deletion)') parser.add_argument('-r', '--root', type=str, metavar='FASTA', default=False, help='Root trees using FASTA sequence as outgroup. (Default: Display unrooted trees)') parser.add_argument('--relatives-out', type=str, metavar='FILE', default=None, help=argparse.SUPPRESS) parser.add_argument('--newick-out', type=str, metavar='FILE', default=None, help=argparse.SUPPRESS) parser.add_argument('--trees-out', type=str, metavar='FILE', default=None, help=argparse.SUPPRESS) parser.add_argument('-m', '--multi-fasta', type=str, default='multi-fasta.fa', help=argparse.SUPPRESS) args = parser.parse_args('-i multi.fa -b 0 -r RSRS.fasta'.split()) if args.input: for fasta in args.input: try: open(fasta) except: sys.exit("\nERROR: Could not open %s\n" % fasta) try: multi = open(args.multi_fasta, 'w+') except: sys.exit("\nERROR: Could not create %s\n" % args.multi_fasta) for fasta in args.input: for line in list(open(fasta)): multi.write(line.replace('-', '_')) if args.root: try: root = list(open(args.root)) root_id = [line.strip()[1:].replace('-', '_') for line in root if line.strip().startswith(">")][0] for line in root: multi.write(line.replace('-', '_')) except: sys.exit("\nERROR: Could not open %s\n" % args.root) else: root_id = args.root multi.close() try: data = ape_read_dna(args.multi_fasta) except: sys.exit("\nERROR: Check existence or proper format of %s\n" % args.multi_fasta) # Get sequence ids in alignment and identify test sequences fasta_ids = [seqname.strip()[1:] for seqname in list(open(args.multi_fasta)) if seqname.startswith('>')] minor_ids = [seqname for seqname in fasta_ids if seqname.endswith('minor') or seqname.endswith('test')] if len(minor_ids) == 0: sys.exit("\nERROR: No test sequences found. _minor or _test suffixes are required in the sequence name!\n") else: pass if args.pairwise: nj_tree = ape_nj(data,1) nj_func = 'function (xx) nj(dist.dna(xx, model="raw", pair=1))' else: nj_tree = ape_nj(data) nj_func = 'function (xx) nj(dist.dna(xx, model="raw"))' if args.bootstrap == 0: cons_tree = nj_tree elif args.bootstrap !=1000: cons_tree = ape_consensus(nj_tree,nj_func,data,iters=args.bootstrap) else: cons_tree = ape_consensus(nj_tree,nj_func,data) # Generate report, trees, and Newick strings if args.relatives_out is not None: relatives = open(args.relatives_out,'w+') else: relatives = open(args.multi_fasta+'-relatives.tab','w+') if args.newick_out is not None: newick = open(args.newick_out,'w+') else: newick = open(args.multi_fasta+'-newick.txt','w+') if args.trees_out is not None: tree_plot_file = open(args.trees_out, 'w+') else: tree_plot_file = open(args.multi_fasta+'-tree-ascii.txt', 'w+') newick.write('%s\n' % (write_nwk(cons_tree))) relatives.write('#source\tsample\trelatives\n') for node in minor_ids: nj_relatives = [relative for relative in dendro_relatives(cons_tree,node) if relative != node] relatives.write( 'Neighbor_Joining_tree\t%s\t%s\n' % (node,','.join(sorted(nj_relatives))) ) tree_plot_file.write(dendro_plot(cons_tree,root=root_id)) newick.close() relatives.close() if __name__ == "__main__": main()