Forum:Discussion question: Cutoff for SNP phylogeny
0
0
Entering edit mode
9 weeks ago
SushiRoll ▴ 100

Dear all,

I've performed a core SNP analysis on a set of E. coli isolates and now my question is rather phylosophical. Which would you say is the cutoff for calling a sequence/clade different from the other. For example, in a more wet lab approach, while performing a PFGE, I would consider a 90% similarity to be the cutoff. With the case of SNP I believe is much more difficult to set this break-point since (Please, correct me if I'm wrong):

1) SNP phylogeny has a much higher resolution.

2) Due to higher resolution, there are inevitable inherent errors.

3) The amount of SNP that can be found is not constant in any species/isolates (Or it might be but within a significant range)

Any opinions/discussions would be very welcome!

Phylogenetics SNP • 1.3k views
ADD COMMENT
0
Entering edit mode

Since you are inviting discussion I would contend that a 10% difference in sequence would no longer keep the organism E.coli. I have not worked with bacterial phylogenetics at sequence level but that seems like a huge difference.

ADD REPLY
0
Entering edit mode

Hey @GenoMax ! Yes, completely agree, the 10% cutoff I mentioned corresponded to PFGE in which you digest the DNA of your isolate with certain enzymes and then analyse the band pattern of each isolate as a 0/1 matrix. This method was considered the gold standard (is it still?) For typing and would render different pulsetypes. In any case, as I mentioned in my three bullet points and as you state, using a method with a much higher sensibility should need a much lower cut off for distinguishing between types/clusters of isolates within the same species. My question now is, which is this cut off value? I've been through a lot of papers and don't seem to find such a value or even a range.

ADD REPLY

Login before adding your answer.

Traffic: 2422 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6