Hi QIIME Team,
Thanks for robust support and tool! I've been using VSEARCH with denovo clustering and chimera checking until now. However, I'm interested to use picrust, bugbase and these rely on greengenes database.
I ran clustering using gg_13_8 database and seems
bugbase needs ids as is from greengenes.
I followed following procedure to get OTU table using VSEARCH and GG:
vsearch --derep_full full_tagclean.fasta --output full_derep.fasta --log=vsearch_log --sizeout --minuniquesize 2 2>derep_log.txt
#dereplicate seqs.fna
vsearch -cluster_fast full_derep.fasta -id 0.97 --sizein --sizeout --relabel OTU_ --centroids otus.fna 2> cluster_log.txt
#cluster at 97%
vsearch --uchime_ref otus.fna --db 97_otus.fasta --nonchimeras otus_checked.fna --sizein --chimeras chimeras.fasta --borderline border_line.fasta --xsize 2> chimera_log.txt
#Pick OTUs at 97% threshold and use gg's 97% otu FASTA as reference db
vsearch -usearch_global full_tagclean.fasta -db otus_checked.fna -strand plus -id 0.97 -uc otu_table_mapping.uc --xsize 2> usearch_global_log.txt
#get .uc file
Later, process using robert edagr's python script to get OTU table.
1- Bugbase does't accept OTU ids with OTU_XXX and such.
Is there a way I could retain database's IDs anywhere in these steps? I know this should be VSEARCH's question but don't know if this would be a appropriate query on their end. I'm using tools and guidance from QIIME developers and external collaborators so thought to ask here. Sorry. :(
2- Or, I use QIIME's scripts and have it do all of these themselves?
Looking forward to hear from you.
VSESARCH version: vsearch v2.0.3_osx_x86_64
Thanks,
Sanjeev