Amplicon data for species distribution comparison

[maggimars]

Look at the global distribution of phaeo species based on Tara 18S v9 data

[maggimars]

Based on Sarah's taxonomy and count files, there is no P. pouchetii or P. antarctica in the Tara 18S data set. If it was just pouchetii that dropped out that would be one thing, but the lack of antarctica showing up indicates that there is a problem. I exported all the representative sequences from the Tara database and tried reclassifying with the qiime2 classifier and the updated pr2 database, but that actually didn't annotate any Tara sequences as any phaeo species. (Many phaeo seqs of all species are in the PR2 database, e.g., there are 25 P. antarctica 18S seqs and all but one are more than 1000bp, 16 are > 1400bp, 11 are >1800bp).

Hypothesis - 18S ref sequences don't have the whole v9 region or don't have the v9 region at all

To Do:

• [ ] run in silico PCR with v9 primers on the Phaeocystis reference sequences - is the full v9 sequence represented in references?
• [ ] try to pull v4 seqs out of the Tara raw metaG reads and classify (potential pitfalls: they will be short ... not full v4 region because hiseq rather than miseq sequencing, not the same part of the variable region will be captured in sequencing reads ...)