Running Qiime2 on Demultiplexed fast files

[jecohen]

Hi, I hope all's well!
I'm new to Qiime2 and I'm a bit unsure of how to proceed with processing my data. I currently have a folder full of demultiplexed and unzipped fastq files (both forward and reversed reads, shown below).

However, they are not in .qzv file format and Qiime2 will not work on them. I've been trying to run Qiime2 scripts with the following error message:

Is it possible to use these fastq files for Qiime2 or do I need to assemble a new database? If so, how can I convert them into .qzv files to run on the software. I realize this is kind of a basic question, so thanks for your time & help!

• Jake

[jwdebelius (Justine Debelius)]

Hi @jecohen,

Welcome to the :qiime2: forum! You need to import your data. My best recommendation is to use the manifest format. It may require gzipping, so possibly test a file and see.

Best,
Justine

[jecohen]

Thanks! I created a test file in the manifest format and attempted to upload it using the instructions on the page given above. I followed the upload instructions and successfully created the output files for both the forward and reverse reads.

However, this doesn't upload the .tsv file from my computer. Am I on the right track? I'm assuming that the .tsv file will allow Qiime2 to align the Sample IDs with the fastq files and let me process the data. However, when I replace "--type 'SampleData..." with "Qiime2Test" (my manifest file format) I get a message saying "an unexpected error has occured: Semantic type Qiime2Test does not have a compatible directory format".
Thanks again!

[jecohen]

I've been working on this, and have managed to upload a test dataset (as indicated above). However, when I try and input it into QIIME I'm given the following error:

I'm unsure if my fastq.gz files are the incorrect file type or if it's the manifest file. There seems to be some issue reading that given the output "file path on line 1 and column could not be found..."
Thanks again!

[jwdebelius (Justine Debelius)]

Hi @jecohen,

It looks like you figured out your semantic type and format issue. (Usually, the semantic type is fixed; you can to change the --input or --output. Looking at the error message, which is telling you the reverse file doesn't exist. Make sure the files are where you say they are (I'm a big fan of pwd to get the file path. One difference I can see is that the manifest is looking for gzipped files and the input directory you posted has unzipped files.

Best,
Justine

[jecohen]

Awesome, thanks! I was able to get the qza files created once I fixed the directory path.

One more quick question- do you know how to format code to use vsearch for quality filtering and denoising? I'm trying to figure that out, but no luck so far. My code
gives the following response.
I'm not sure what file type they're looking for/what they want with --I-table.
Thanks again!

[jwdebelius (Justine Debelius)]

HI @jecohen,

I can't see your command, so I'd know specifically. In general, I think the best way to navigate QIIME 2 (as a new or experienced user) is often to find one of the tutorials that lines up with your goals. In your case, you might find the OTU clustering tutorial helpful for information of the types of files and how to format your commands. The help docs are also always thtere (just enter --help after the command). You can also check your semantic type at view.qiime2.org if you're not sure what kind of artifact you have compared to the kind of artifact you need.

Best,
Justine

[jecohen]

Awesome, thanks, I think I should be able to get it!