Beta_significance.py using OTU table summarized by category

46 views
Skip to first unread message

Amy Fisher

unread,
Apr 21, 2011, 9:19:16 AM4/21/11
to Qiime Forum
Hello,

I am hoping you can help me resolve an issue I am having trying to run
weighted Unifrac on an OTU table summarized by category. I have done
this successfully in the past, but recently am getting an error that
reads:

Traceback (most recent call last):
File "/software/qiime-1.2.1-release/bin/beta_significance.py", line
151, in <module>
main()
File "/software/qiime-1.2.1-release/bin/beta_significance.py", line
126 in main
parse_otu_table (otu_table_lines)
File "/software/python-2.6.4-release/lib/python2.6/site-packages/qiime/
parse.py", line 453, in parse_otu_table
otu_table.append(array(map(count_map_f, fields[1:-1])))
ValueError: invalid literal for int() with base10: '0.0'

Thank you,
Amy Fisher

Greg Caporaso

unread,
Apr 22, 2011, 7:27:45 AM4/22/11
to qiime...@googlegroups.com
Hi Amy,
Can you post the exact command that you're running?

Greg

justink

unread,
Apr 22, 2011, 4:27:04 PM4/22/11
to Qiime Forum
I suspect the issue is that you have floating point numbers in the otu
table you're using as input to beta_significance.py (beta_diversity.py
will behave similarly). The unifrac and weighted unifrac metrics
expect a table of integer counts, which should be obtainable by
omitting -n when you call summarize_otu_by_cat.py.

Give me a holler if that doesn't do the trick, or if you have
fractional counts in your original otu table for some reason.

Amy Fisher

unread,
Apr 26, 2011, 8:40:18 AM4/26/11
to Qiime Forum
Greg,

I was running the 'beta_significance.py' command and the post from
justink hit the nail on the head... I was using a normalized table.

I am re-running and omitting the -n flag, which worked great on the
test data that I ran.

Thanks,
Amy

Amy Fisher

unread,
Apr 26, 2011, 8:40:49 AM4/26/11
to Qiime Forum
Thanks! This was the problem.

Amy
Reply all
Reply to author
Forward
0 new messages