Import sequence into Qiime2

I am new to using Qiime2 on a Windows 10 Windows subsystem for linux Ubuntu system. I installed Qiime2 using the conda environment and activated it via: conda activate qiime2-2020.11. My sequences have been stripped of primer and barcodes by the Novogene sequencing facility and I have both reverse and forward sequences for each sequence installed in a sub directory (B61_1.fq.gz, B61_2.fq.gz) . After reading the most recent import guidelines I decided to import using a manifest file system. I assume that my sequences are Phred33V2. I have spent several days searching the forum for helpful information, much of which was helpful in promoting an overall understanding of importing. I chose to simplify the procedure to see if it would work on importing just one forward and reverse sequence. If that was successful, then I would be on the right track and could proceed with further imports. However, I still have a problem. I get the following error message:

(1/1) Invalid value for '--input-path': Path '/home/michaelanderson74078/practice/pe-b61-manifest.txt' is not readable.

Here is my manifest file entitled: pe-b61-manifest.txt

sample-id forward-absolute-filepath reverse-absolute-filepath

B61 /home/michaelanderson74078/practice/sequences/B61_1.fq.gz /home/michaelanderson74078/practice/sequences/B61_2.fq.gz

#my import command is as below:

qiime tools import \
--type 'SampleData[PairedEndSequencesWithQuality]' \
--input-path /home/michaelanderson74078/practice/pe-b61-manifest.txt \
--output-path /home/michaelanderson74078/practice/paired-end-demux.qza \
--input-format PairedEndFastqManifestPhred33V2

The sequences are found in a subfolder /home/michaelanderson74078/sequences/ B61_1.fq.gz or …// B61_2.fq.gz

Also when I try to open the manifest file via the nano text editor in the WSL window I get an error message:

Error reading pe-b61-manifest.txt: Permission denied.

Obviously the WSL Ubuntu system is not able to read the manifest file. The manifest file was created by Notepad++ text editor and saved as a normal txt file.

What am I missing? Am I using the right file type for the manifest file? Should I save it as a csv file type?

The import system works when I use the downloaded unzipped sequence data following all the directions. I get a legitimate qza file, but not with my sequences.

I am stuck.

Welcome to the forum, @mander, and thanks for a solid first post. You gave us lots of good information to work with! I added some markdown to your post to make the code blocks easier to read. If you want to see how I did that, just ‘edit’ the post.

This appears to be a permissions issue, specific to your manifest.txt. I’m not sure why the permissions are broken for that file, but it’s likely related to WSL, or how files created in Windows are permissioned when opened in WSL. (These are just educated guesses - I don’t spend much time with Windows.)

Try navigating to your practice directory from the WSL command line, and entering the command ls -lah. This will show you all the files there in list view: image

The rs, ws and xs on the left side indicate ‘read’, ‘write’, and ‘execute’ permissions for each file. If you don’t have r permissions, the command chmod +r myfile.thing should add them for you. If that doesn’t do the job, share a screenshot of the ls -lah results for your file, and whatever error message you run into.

Best,
Chris :baby_chick:

Thanks for the solution. It was the permissions which were easily adjusted based on your input. You were very helpful and your expertise is greatly appreciated. It turned out that not only the manifest file was lacking permissions but also the sequence files as well. Once all that was straightened out I was able to run the program and generated a .qza file. Now I can repeat the same with all the other sequences.

2 Likes