Converts a tree in newick format into a FastTree-compatible constraint alignment. It allows for multifurcated nodes.

Newick2FastTreeConstraints.py

#!/usr/bin/env python
import sys
from collections import defaultdict
from ete2 import Tree
try:
    t = Tree(sys.argv[1])
except IndexError:
    print >>sys.stderr, 'you need to provide a newick tree file as first argument\n\n'
    print >>sys.stderr, 'Usage: Newick2FastTreeConstraints.py tree.nw > constraints.fa'
    sys.exit(1)

n2content = t.get_cached_content()
all_leaves = sorted(n2content[t])
alg = defaultdict(list)
for n in t.traverse("postorder"):
    if len(n.children) > 1:
        ones = n2content[n]
        for leaf in all_leaves:
            if leaf in ones:
                alg[leaf].append("1")
            else:
                alg[leaf].append("0")

print >>sys.stderr, "Number of leafs:", len(alg)
print >>sys.stderr, "Number of constraints:", set([len(v) for v in alg.values()])
for name, values in alg.iteritems():
    print ">%s\n%s" %(name.name, ''.join(values))