over
groups={}
groups.setdefault(key(item), []).append(item)

and
def keyfunc(it): return it.name[it.name.find(’|') + 1 : it.name.find(’/')]

over
keyfunc = lambda it: it.name[it.name.find(’|') + 1 : it.name.find(’/')]

I will post your comment in a regular post in order to format it. Thanks a lot for your comment.

Cheers

defaultdict is not really required here, the same could be done by initializing data to a normal empty dict (i.e. data={}) and using: data.setdefault(ident, []).append(…)

Generally, while this is nice for practicing your programming skills, itertools.groupby can do most of this work for you. Here’s how (not tested) – notice that this version is shorter and clearer:

def merge_seqs(data1, data2):
from itertools import chain, groupby

format = “%s-%s->%d\n%s%s”
flist = []
keyfunc = lambda it: it.name[it.name.find('|') + 1 : it.name.find('/')]
for item, g in groupby(sorted(chain(data1, data2), key=keyfunc), keyfunc):
values = list(g)
if len(values) == 2:
jname, jseq = values[0].name, values[0].sequence
kname, kseq = values[1].name, values[1].sequence
flist.append(format % (jname, kname, len(jseq), jseq, kseq) )

return flist

The only “ugliness” here is that groupby requires that the items in the iterator it works on be sorted in advance by the same key function. For this reason I usually just use a dict with setdefault, as I mentioned above, but in this case using groupby tidies up the code considerably.

IMO the best solution here is to use this simple groupby replacement:

def groupby_unsorted(iterator, key=lambda x:x):
“”"Like itertools.groupby, but doesn’t require that the given iterator be
sorted in advance, and doesn’t work with infinite iterators.

“”"
groups={}
for item in iterator:
groups.setdefault(key(item), []).append(item)
return groups.iteritems()

This way you can use the the second, more concise, version of the function above, without needed to first sort the data, by replacing the ugly:
groupby(sorted(chain(data1, data2), key=keyfunc), keyfunc)

with:
groupby_unsorted(chain(data1, data2), keyfunc)

Finally, I just want to say that this was a nice post, and I hope to see more like it