Is this phylogenetic tree normal

Hi QIIME2 developer,

I have found an interesting issue which needs your help. I basically followed Moving picture tutorial pipeline to process my 16s data covering V4 region. I did quality control by DADA2 and classified taxa by Bayesian classifier. Here is a small list of taxon I am interested and their taxonomic table and phylogenetic tree(I relabelled the tip name by their phylum ID).

image

Looks these taxa are classified pretty solidly. But why some taxa from same phylum(e.g., Firmcuites) were clustered into different branches together with taxa from other phylum? Is that normal from evolutionary perspective or due to some possible issues in processing?

Thanks in advance.

Dong

Hi @lindd,
How was this tree built? Is it really a phylogenetic tree, based on the trimmed V4 sequences? Or something else?

That definitely doesn't sound normal. You are working with V4 sequences — which aren't really going to replicate the same phylogenetic distances you'd expect from the full 16S — but I still wouldn't expect the firmicutes to scatter across the tree like this.

I would do the following if I were in your shoes: :athletic_shoe:

  1. hand check the features in that tree to make sure that the features/sequences/taxa are all correct.
  2. use NCBI BLAST to double-check the closest matches to these sequences (remember to exclude "uncultured" sequences). The classifier is getting high-confidence hits for most of these, but you can never be too sure.

Let us know how it goes! I am very curious to get to the bottom of this.

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.