I am looking at some sequences collected before and after a perturbation and hoping to quantify and identify the shared and unique OTUs across the perturbation. So I used the venn() command to look at a venn diagram and get the number of shared OTUs from the pre and post samples. I then was curious to see how “abundant” these OTUs were in the pre or post datasets. It seemed that get.sharedotu() was the best bet. However, it seems the results from venn() and get.sharedotu() are not agreeing. Is this expected? Are the two functions doing something different? My expectation was that if venn() reported 50 shared OTUs, then there would be at the very least 100 sequences (1 from each group in each of 50 OTUs) in the output file from get.sharedotu(). However, I am getting much less than my expected number of sequences.
Thanks,
Stuart
Stuart,
Try the Heatmap command, it will give you a .svg file with an indication of how abundant this OTUs are. You may need to change the scale to adjust to your dataset. It worked fine for me.
Cheers,
Sorry for the delay, someone else simultaneously asked the same question by email and I think we forgot to reply to you by forum. Here was our reply…
The difference occurs because the venn command reports otus that are shared by the two groups, whereas the get.sharedseqs command reports otus that are unique to the two groups. For example if otu 1 contains sequences from groups a,b,c and d. The venn diagram would report it as shared between a and b, but the get.sharedseqs command would not report it because it also contains sequences from groups c and d. To make the command more flexible we have removed the groups parameter and in it’s place added unique and shared. If you enter your groups under the unique parameter mothur will report otus that contain only sequences from the groups you enter. If you enter your groups under the shared parameter mothur will report otus that contain sequences from the groups you enter as well as sequences from other groups. If you do not enter any groups mothur reports the otus that are unique to all the groups in your groupfile. Here are a few example:
–OTUS unique to all the groups in your groupfile
mothur > get.sharedseqs(list=dwlibs.fn.list, group=dwlibs.good.groups, label=0.02)
0.02 4
–OTUS unique to groups DWDNA6 DWRNA6 (to be included in v 1.8)
mothur > get.sharedseqs(list=dwlibs.fn.list, group=dwlibs.good.groups, label=0.02, unique=DWDNA6-DWRNA6)
0.02 1
–OTUS unique to groups DWDNA6 DWRNA6 (to be included in v 1.8)
–OTUS shared by groups DWDNA6 DWRNA6
mothur > get.sharedseqs(list=dwlibs.fn.list, group=dwlibs.good.groups, label=0.02, shared=DWDNA6-DWRNA6)
0.02 11