Concatenating Nexus Files (For Mrbayes, Etc)

Question

[Andre Elias]

Hi, I've been trying to concatenate multiple genes into a single nexus file to be used in MrBayes.

Apparently, Biopython has this script in their website, but it's not working for me (ref: http://biopython.org/wiki/Concatenate_nexus):

from Bio.Nexus import Nexus
# the combine function takes a list [(name, nexus instance)...], if we provide the
# file handles in a list we can use a list comprehension to such a list easily
handles = [open('btCOI.nex', 'r'), open('btCOII.nex', 'r'), open('btITS.nex', 'r')]   
nexi =  [handle.name, Nexus.Nexus(handle)) for handle in file_list]

combined = Nexus.combine(nexi)
combined.write_nexus_data(filename='btCOMBINED.nex')

It keeps spitting me this error:

nexi =  [handle.name, Nexus.Nexus(handle)) for handle in file_list]
NameError: name 'file_list' is not defined

If it worked it would be great, because it would automatically create partitions for each gene, etc, which is exactly what I need. Any help in this direction is welcome. :-)

[David W]

I think Frédéric's answer is right - I actually wrote the example on the Biopython wiki, so sorry for letting the bug slip in.

[Frédéric Mahé]

The BioPython Wiki helped me so many times! Thank you for work.

[Peter]

Yep - the encoding thing was reported (but not on the mailing list or bug tracker) and fixed ready for our next release: https://github.com/biopython/biopython/commit/f152fa2c2a7f71adaaf51b746cfd0a4308ea2edb - Sorry about that.

[Peter]

David, could this be simplified a little by giving Nexus filenames instead of handles?

[Andre Elias]

Sounds great, but now I'm hitting this error message: File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/Bio/Nexus/Nexus.py", line 1275, in write_nexus_data fh=open(filename,'w',encoding="utf-8") TypeError: 'encoding' is an invalid keyword argument for this function PS: Thanks for all the hard work in the wiki, it's very useful indeed! :-)

[Andre Elias]

I changed, as suggested by david w, the last line to combined.write_nexus_data(open("combined.nex", "w")) and it works now. Thank you all for the help! :D

[David W]

@Peter - yep, the example was written when SeqIO took handles, so I guess that was just the style. Nexus.combine() takes either handles or filenames

[David W]

@Andre, thanks for alerting us to these bugs, I gather the 'new' error you found came from file handles changing in python 2/3. I'll update the wiki with the workaround, but the problem is already fixed in the github version of Biopython - so you will be able to use a filename in the next release

Answer 1

[Frédéric Mahé]

I guess that if you replace file_list by handles, your python example should work just fine.

[Peter]

David has fixed the wiki example now.

Answer 2

[Zev.Kronenberg]

You might find mesquite useful. Pretty easy to run and I believe it does what you want.

Look under:

Fused Matrix Export (NEXUS):

http://mesquiteproject.org/mesquite_folder/docs/mesquite/molecular/molecular.html

[Zev.Kronenberg]

It is damn useful for phylogenetics. Happy tree building!

[Andre Elias]

I believe it does that (concatenating, etc), but it uses a GUI that makes its use impractical for really large data sets (imagine concatenating a hundred alignment files, one by one) and it's also not very speedy (I'm not sure but I've heard it requires a lot of memory too). A script (perl/python/shell) could do it much faster and also could be integrated in a workflow pipeline. Your suggestion is good for people who intend to concatenate smaller datasets and would like to use a GUI for that. :-)

Answer 3

[arunprasanna83]

Hello, I have it working fine for the example. When I used it for my data with ~600 .nex files, it sure did generate the output. But the problem is, few taxon names were duplicated with copy tag. For example, ideally I should have got X Y Z

but I got

X Y Z X.copy ??? X.copy1 ??? Y.copy ??? MA??

I did ensure that, all my files do not have duplicate copies of alignment. Thanks in Advance. Arun