summary.single command produces std 0 values

Commands in mothur
1

Dear mothur-community,

we are still working through the SOP. And now are at the point of calculating different alpha-diversity indices. The last command we used is the “summary.single”-command and looked in the resulting ***.groups.ave-std.summary file.

Receiving the following output:

label group method nseqs coverage sobs invsimpson invsimpson_lci invsimpson_hci shannon shannon_lci snip
0.03 FSM-B ave 843801.000000 0.993619 8526.265000 7.824210 7.781387 7.867506 4.414662 4.409967 snip
0.03 SDM-B ave 843801.000000 0.987543 13480.000000 22.228329 22.112835 22.345036 4.723693 4.722382 snip
0.03 FSM-B std 0.000000 0.000021 15.973565 0.004586 0.004552 0.004620 0.000658 0.000660 snip
0.03 SDM-B std 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 snip

What troubles us is the last line. All the values for the standard deviation are 0. Why?

Thank you very much indeed for your input and help.
Regards,

2

Probably because that sample had 843801 sequences. There wouldn’t be a sd if every subsampling had the same number of reads.

Pat

3

Thanks for your reply, Pat!
But what about FSM-B. There we have also 843801 sequences and there is a sd calculated?

4

If you can send your shared file to mothur.bugs@gmail we can take a look. Please reference the url of this post in the email

Pat

5

When you use the subsample parameter with true or false instead of a sample size, mothur uses the size of your smallest group as the subsample size. In your case the smallest group is SDM-B, so the sample size is set to the total number of sequences in that sample. This will result in all the sequences for SDM-B being sampled each iteration resulting in a standard deviation of zero.