All Commands used in Basespace QIIME Preprocessing

[Maxime D.]

Dear QIIME forum/support,

I searched this forum to find the answers to my questions but I didn't find it. So I allow myself to open a new question.

I'm using Basespace QIIME Preprocessing app to convert uploaded FASTQ Illumina files to OTU table and FNA file. 

After that step, I download QIIME Preprocessing outputs (OTU table) to continue 16S metagenomics analyses on my side.

Is it possible to have the exact QIIME commands used in QIIME Preprocessing app to know exactly what happens during this process?

I guess commands are used with their default parameters.

QIIME Preprocessing app performs reads 1 and 2 assembly? I see that FNA file from QIIME Preprocessing output seems to contain only 300 pb reads (I used paired reads 2x 300pb).

Is there a chimera check? quality filtering? 

Sorry for all that questions.

Thanks a lot for your help,

Best regards,

Maxime

[Colin Brislawn]

Hello Maxime,

Thanks for getting in touch with us. These are good questions and I'll address each one of them.

Is it possible to have the exact QIIME commands used in QIIME Preprocessing app to know exactly what happens during this process?
I guess commands are used with their default parameters.

As far as I know, the qiime basespace app is on github hub. Here is the 'upstream' code:

https://github.com/biocore/basespace-qiime/blob/master/scripts/upstream.py 

QIIME Preprocessing app performs reads 1 and 2 assembly? I see that FNA file from QIIME Preprocessing output seems to contain only 300 pb reads (I used paired reads 2x 300pb).

I don't think reads are joined, because I don't see the script multiple_join_paired_ends.py.

http://qiime.org/scripts/multiple_join_paired_ends.html 

Is there a chimera check? quality filtering? 

The split_libraries script also does basic quality filtering. 

 https://github.com/biocore/basespace-qiime/blob/master/scripts/upstream.py#L47 

 

Because closed-ref OTU picking is used, and the referance database is presumed to be chimeric free, no chimera checking is performed. 

You ask all the right questions about the basespace application and highlight the limitations of the basespace app. This app was made to be a quick overview of your samples. Using these qiime scripts directly is much better because you can customize your workflow and add important steps, like multiple_join_paired_ends.py which will increase quality. To use these other scripts, you could try installing qiime: 

http://qiime.org/install/install.html 

Keep in touch,

Colin

[Maxime D.]

Dear Colin,

Thanks for your time and answers.

Indeed in upstream.py QIIME Preprocessing code I only see commands "multiple_split_libraries_fastq.py", "pick_closed_reference_otus.py" and "biom summarize-table".

I'll try to use "multiple_join_paired_ends.py" on my QIIME virtual machine to merge paired reads. 

Thanks again, you answered all my questions. 

Kind regards,

Maxime