Get.oturep: Too much OTUs

[feiluo]

Dear Pat,

Not sure if it is me doing something wrong or the others but…
With the intention to view OUT table, when I run:

mothur >get.oturep(fasta=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.fasta,count=stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.vsearch.pick.pick.count_table,list=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.opti_mcc.list,label=0.03,sorted=name,method=abundance)
Output file names: stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.opti_mcc.0.03.rep.count_table

Finally,I exported this file and opened it in Excel, there are a total of 487,573 OTUs. Starting from the 9660th OTU, the number of sample abundance sequences begins to show a single digit (less than 10). Do you think it is necessary to remove these and finally build a phylogenetic tree?

Looking forward to your reply. Thank you very much！

[pschloss]

Hi,

No I strongly discourage the removal of “rare” OTUs from analyses since it imposes a bias into the depiction of the community structure of your samples. I’m also not sure what value there is to generating a tree from so many sequences.

pat

[feiluo]

Thanks a lot for your quick reply - much appreciated.

Now that we have the species (OTU) annotation information and its abundance, the next step is to return the species information to the taxonomic system relational tree in the database, with the aim of fully understanding the evolutionary relationships of all the microorganisms in the sequencing samples from the whole classification system. In addition, the abundance of each species can be added to fully demonstrate the species evolutionary relationship and the abundance differences in different samples.

Like this:

Do you think that generating so many OTUs is normal? Will it be an error in the previous analysis step?

Thank you !

[pschloss]

The data are what the data are. Sorry, but I’m not sure what you would expect. I would also never encourage anyone to put pie charts on a tree

[feiluo]

Thank you very much for your comments and suggestions. I plan to re-analyze the data and finally make a beautiful and reasonable tree graph. Thank you again!