Hello Asun,

3- one read can produce more than one high quality segment

2- in your case, the counts are related to the number of blast hits that meet the criteria you put in the mega blast application. There are many blast hits per high quality segment.

1-after fetch taxonomic representation first run "find lowest diagnostic rank' and then run the summarize taxonomy and draw phylogeny tools. This should work.

Regards,

Judith van Bleijswijk

............................................................................. Hi, I am trying to analyse my eukaryotic metagenome data using yours workflow for windshield splatter analysis. But I find several problems:

1- it is related to "draw phylogeny", this tool fails always for most of my samples, reporting the error: incorrect tree structure. Tree string position 341

2- It is related with the "summarize taxonomy" step. I this case I obtain results but the number of counts are too high compared with the original reads. My samples are small, around 8000 reads and the counts obtained are more that 200000 for some phyla. I guess this is not ok, as far as I understand from your paper than counts are equivalent to reads, am I right?

3- another strange thing is the fact that the number of sequences increase after run the step for select the high quality segments. Any reason for this???

I will really appreciate you help. Thanks in advance Asun