Meryl: wrong outputs with compressed FASTQ files

[gallardoalba]

A user reported an error while running Meryl (original post). The error seems to be related to compressed FASTQ files, however, according to the developer, Meryl should be able to process such files. When using compressed files, Meryl generates count files, but those seem to be corrupted.

• Test small dataset (decompressed): history link
• Test small dataset (compressed): history link

The decompressed test worked fine.

[bernt-matthias]

Hi @gallardoalba the datasets in the first two histories are not accessible.

[gallardoalba]

Hi @gallardoalba the datasets in the first two histories are not accessible.

Hi, thanks. When I try to share the history, it triggers this error, but not sure what is the reason.

[bernt-matthias]

@bgruening can you help here?

[bgruening]

The fastq dataset is only shared with @gallardoalba. So he has not permissions to share it further.

[gallardoalba]

Thank!

[gallardoalba]

I fixed the links @bernt-matthias; I tested the problem by running Meryl in the command line, and both generated correct outputs.

[bernt-matthias]

So I can close the issue?

[gallardoalba]

So I can close the issue?

Nop, it still doesn't work in Galaxy.

[bernt-matthias]

Can you post the command line that you used? In my experiments the command line for count does not work:

meryl count k=19  /tmp/tmp5is6p6g8/files/000/dataset_1.dat  output read-db.meryl

Problem might be the .dat extension. But this should be easy to fix.

[gallardoalba]

This is the command I used:

meryl count Galaxy4-\[Illumina.fastq.gz\].fastqsanger.gz k=19 output tmp

[bernt-matthias]

@gallardoalba this has been fixed in #3876. Latest version (1.3+galaxy4) is already on eu.