I have tried multiple ways to get the fastq.gz files to convert to .qza files using the qiime2 tools import function in Galaxy, and I’m at my wits’ end.
When I try to use the manifest file (which requires “Single End Fastq Manifest Phred33V2” to be selected), it lets me run the tool, but it always returns an error.
When I try to use something other than the manifest file, using “Single Lane Per Sample Single End Fastq Directory Format,” the PRJEB41511_single_end_reads collection (which includes the 158 fastqsanger.gz datasets I need), Galaxy will not let me remove the “import manifest” and “import metadata” section, so I cannot run it.
Yes, complicated! The baseline requirements are that the read data retain their original Illumina names (Qiime2 parses this metadata), to put the reads into a simple flat list dataset collection, then to choose the appropriate qvz archive type for that data.
The archive type will be same as chosen on the command line. This is not well documented for Galaxy separately, so the original Qimme2 tutorials and forum are the usual resources if you are not sure which to use.
All that said, let’s get this to work for you!
We have some help in these two topics. Both focus on paired end read archive choices, but the other parts of the help will be common for your use case, too (sequence names and building the flat list collection).
From what you have now, I think your archive choice is good! To get rid of the manifest, use Copy Datasets to move your collection of reads over to a new history and try there. (And, I’ll review and ticket the issue with that form choice being too “sticky”).
Please give that a try and let us know if that is enough! We can’t see your collection or the element identifiers you are currently using but you could screenshot those next.
