Phylotype command

Commands in mothur
1

Hi,

I am using Costello stool analysis to practice and I am up to phylotype-based analysis.

When I type

phylotype(taxonomy=stool.final.taxonomy, name=stool.final.names) command in mothur, it crashes.

I am using Windows and mothur v.1.21.1

Your help will be appreciated.

2

Hmmm… sounds odd. Can you email your files to mothur.bugs@gmail.com?

3

Hi I also have the same problem with Mothur crashing after I run the phylotype command. I am running Windows also. I see this is quite an old forum post, was this issue ever resolved? I am thinking there is probably a problem with the files I have generated before this point.

Many thanks in advance!

4

No, they never followed up with us, unfortunately

I also suspect it’s a problem with your older files, but what version of mothur are you running? You might try going back a few steps and re running the commands and see if the issue resolves itself.

5

Hi! I am also a beginner and I am trying OTUs and Phylotype approaches to see want comes from my data with both approaches. With OTUs I used 0.03 cutoff that if I got it right, will at the end classify my sequences in OTUs identified to the sub-genus level.
With Phylotype, I don’t fully understand how to select the label. The help reads:

The labels count up from the leaves to the root of the phylotree. There may only be a couple of level in your phylotree that you are interested in seeing. You can use the label option to specific the levels you would like. If you want the data for the lines labeled unique, 1, 3 and 5 you would enter:

So, how do I know to which taxonomic level correspond each number? using label=1 as in the SOP454 tutorial, the sequences will be classified to the genus level? and to which levels correspond the other numbers? I don’t know how to count the leaves when some lineages have sub-class or sub-order and others don’t.
And the numbers correspond to the same levels no matter if we use RDP taxonomy or SILVA taxonomy?

With my Sanger dataset, after filtering I ended up with 795 sequences from 8 samples (antarctic soil). Using the OTU approach with 0.03 cutoff I got 545 OTUs, with many unclassified even at the phylum level. Using Phylogeny approach using label=1 I got only 154 OTUs with only une unclassified. What does all this mean exactly and what should I do not to make mistakes? Is that possible to run the OTU approach but to the genus level not to have so many unidentified or should I take this as coming from wrong sequences that could not be detected as wrong during the pipeline?

Thank you very much!!!

6

1 = Genus
2 = Family
3 = Order
4 = Class
5 = Phylum
6 = Kingdom

At this point this mapping to the Linnean levels only works with the greengenes and RDP taxonomies.

Pat

7

Dear Dr. Schloss and other mothur experts,

May I consult you several questions regarding the label in phylotype command:

You said the label 1-6 corresponds genus to kingdom levels. To be honest, I did not get your statement very clearly and my results from “phylotype” command give me another impression (I am using SILVA database):

  1. When I used the label “8”, all the reads can be binned into phylum level. These results made sense in my mind. That’s what I want when we set the label at phylum level. For example:

Taxonomy
Bacteria(100) Proteobacteria(100)
Bacteria(100) Firmicutes(100)
Bacteria(100) Bacteroidetes(100)
Bacteria(100) unclassified(100)
Bacteria(100) Spirochaetes(100)
Bacteria(100) Synergistetes(100)

  1. However, when I use the label “5” correspond “phylum” as you mentioned, many reads were binned into different groups, even several these groups belong to same phylum (e.g. Firmicutes). It does not make sense. For example:

Taxonomy
Bacteria(100) Proteobacteria(100) Gammaproteobacteria(100)
Bacteria(100) Bacteroidetes(100) Bacteroidia(100)
Bacteria(100) Proteobacteria(100) unclassified(100)
Bacteria(100) Firmicutes(100) Clostridia(100)
Bacteria(100) Firmicutes(100) Clostridia(100)
Bacteria(100) Bacteroidetes(100) Bacteroidia(100)
Bacteria(100) Firmicutes(100) Clostridia(100)

Therefore, I am thinking maybe for SILVA database/tax, label 8=phylum, 7=class, 6=order, 5=family, 4=genus…If I am wrong, could you tell me why?

  1. In the example folder “Taxonomy Files” and the MiSeq SOP for the “phylotype” command, you used silva-based taxonomy; however, in this mothur forum, you stated that “at this point this mapping to the Linnean levels only works with the greengenes and RDP taxonomies”. To be honest, that makes me confused. Could you tell me whether we can use this “phylotype” based on silva taxonomy?

Many thanks for your time and help! All the best!

Yours sincerely,
Nathan

8

The levels I posted were for the RDP.

9

Dear Dr. Schloss,

Actually I am using SILVA and the results still make sense if I set the label option as:
8=phylum
7=class
6=order
5=family
4=genus

I checked the output by eyes and hands, they did make sense.

How do you think? Do you agree? Or I made mistake? Many thanks!

Yours,
Nathan

10

That won’t work because there are sub levels inserted in the old SILVA reference taxonomy. In essence there are levels 6.5 and 5.5, etc. Really, just use the new SILVA taxonomy or the RDP or greengenes taxonomies.

pat