tree.shared with certain samples

Hi all,

I am trying to run tree.shared command using only some of the samples in the final.an.shared file but I am getting some errors. I removed some samples from the final.an.shared, as we have a control population with high number of samples, which make the tree visualization a bit difficult to understand. Among the 340 samples in total, 136 are control population. Does anybody have any ideas?


mothur > tree.shared(shared=donors.final.an.shared, calc=thetayc-jclass, subsample=3000)
Using 1 processors. You have not provided enough valid groups. I cannot run the command.

It looks like mothur thinks you have less than 2 samples. Have you modified the shared file? Could you post part of the shared file so we can see if its a formatting problem?

Hi Pat,

I removed some samples from the shared file, as I don’t want them in the dendrogram you get after tree.shared. But I still have 208 samples left in the file.
It looks like that, but with quite a few more OTU columns

label Group numOtus Otu00001 Otu00002 Otu00003 Otu00004 Otu00005 Otu00006 Otu00007 Otu00008 Otu00009 Otu00010 Otu00011 Otu00012 Otu00013 Otu00014 Otu00015 Otu00016 Otu00017
0.015 D10_28dpi 21170 271 329 27 908 9 215 53 0 550 3 431 0 46 24 0 1 118
0.015 D10_3dpi 21170 224 603 16 202 31 163 297 0 328 13 584 0 11 87 0 5 50
0.015 D10_6dpi 21170 98 39 0 44 16 33 29 0 9 4 2708 0 0 13 0 0 4
0.015 D10_B 21170 626 305 7 283 25 77 144 9 15 22 704 0 2 56 7 27 27
0.015 D11_28dpi 21170 482 2691 2 518 822 1110 18 38 436 583 23 1 19 88 71 343 8

Instead of removing them manually, which could result in some OTUs having abundance=0 (mothur is pretty picky about that), could you use the remove.groups command? http://www.mothur.org/wiki/Remove.groups

Thanks a lot westcott,

I got it using remove.groups.