Link for fungal ITS database

[Bioinfonext]

Hi,

I need to use vsearch for chimera detection of fungal ITS sequences, and further, I need to process data using qiime1..

do I need to download different database for vsearch and qiime.?

Thanks

Yogesh

[TonyWalters]

Hello Yogesh,

I'd download the UNITE ITS database for QIIME, e.g. the latest version: https://files.plutof.ut.ee/doi/AE/CB/AECB427AE56E4883A30BE88A7AAAAAAFEC082EF456AF37EE2E81C85228C49658.zip

This link is from the UNITE website (make sure to cite them): https://unite.ut.ee/repository.php

Unless you do processing with the ITSx software (http://microbiology.se/software/itsx/) to trim the SSU/LSU regions of your reads, I would use the reference files in the /developer/ folder of the UNITE database.

-Tony

[Bioinfonext]

Hello Tony,

Thanks. I already processed data using  ITSX command:

ITSx -i seqs.fna -o seqs-its-cleaned.fna -p /opt/apps/ITSx/1.0.11/ITSx_db/HMMs -t F --preserve T --save_regions SSU,ITS1,5.8S,ITS2,LSU --minlen 150 --detailed_results

Now, I need to use vsearch for chimera detection of fungal ITS sequences, and further, I need to process data using qiime1.

do I need to download different database for vsearch and qiime.? do you think should I use reference files in the /developer/ folder of the UNITE database for both vsearch and qiime. does it also contain taxonomy file?

Kind Regards

Yogesh

[TonyWalters]

I would use the database that I linked for UNITE above. If you used ITSx, you don't have to use the /developer/ folder, you can use the reference fasta file that's in the base folder. These *are* the reference files you would use with QIIME (for any method with pick_otus.py, whether it's the default uclust or if you're using vsearch as a plugin for the -m usearch61 method).

[Bioinfonext]

Hi Tony,

thanks for your all help. 

Actually I am following this protocol for ITS fungal analysis.

They are suggesting two different database for chimeric detection and further qiime analysis.

here is the link:

http://geoffreyzahn.com/getting-started-with-qiime-for-fungal-its-cleaning-its-reads-and-picking-otus/

and what is the difference between developer and references database of fungal ITS.

Kind Regards

Yogesh

[TonyWalters]

So I think (unless I'm missing something) that the "two" databases are just separate files for a single database-one is the taxonomy mapping file, and the other is the matched fasta sequence file. These lines point to the two different files within the database:

assign_taxonomy:id_to_taxonomy_fp ~/ITS_Reference_Databases/sh_taxonomy_qiime_ver7_dynamic_s_01.08.2015.txt
assign_taxonomy:reference_seqs_fp ~/ITS_Reference_Databases/sh_refs_qiime_ver7_dynamic_s_01.08.2015.fasta

Inside the UNITE database, there are multiple options-I usually go with the file with the 'dynamically' picked reads, since that is supposed to improve taxonomic assignments.

The /developer/ files have the overhangs from the SSU/LSU in them. In the base folder, these overhangs are trimmed, e.g. by ITSx. It is a problem when you have reads with overhangs that do not match the reference database because the reference database is trimmed, causing them not to cluster or to have bad taxonomic assignments. Since you processed with ITSx, you don't need to use the /developer/ files.