Trying to cluster at 99%, getting 97%

[Jocelyn Sietsma Penington]

Hello,

I wanted to cluster my sequences at 99%. I've attached parameters99.txt, the parameters file I used.

The commands I used were : 

  pick_de_novo_otus.py -i $DATADIR/labelled_seqs/seqs_rc.fna -o $WORKDIR/denovo_otus \

  -p $SCRIPTDIR/denovo99parameters.txt -aO$(nproc) -f 

(denovoparameters99.txt is the same as parameters99.txt)

and 

  pick_open_reference_otus.py -i $DATADIR/labelled_seqs/seqs.fna -r $REFSEQ -o $SSODIR -aO$(nproc) \

  --min_otu_size=3 --suppress_align_and_tree -p $SCRIPTDIR/parameters99.txt

In each case the clustering was actually done at 97%

This is shown in the log files, which both include:-

  UclustOtuPicker parameters:

  Application:uclust

  Similarity:0.97

and by inspecting the clusters.

What am I doing wrong?  Do I need to add UclustOtuPicker to the parameters file? I didn't see it listed in the possible parameters in the overall log file or when I used option -w to check commands

Thanks,

Jocelyn

[Jose]

Hi Jocelyn,

in the parameters file you need to specify the name of the script without the ".py", so for instance you should have:

pick_otus:similarity 0.99

and so on. See here for more details on the format of parameters files:

http://qiime.org/documentation/qiime_parameters_files.html

Jose