Question: Galaxy failed to unzip gz files after uploading
2
gravatar for jamwest101
8 weeks ago by
jamwest10120
Canada
jamwest10120 wrote:

Hi All: Compressed files used to be decompressed in galaxy after uploading. Recently, the compressed files (gz files) are failed to be decompressed after uploaded into Galaxy. Is it a common problem or just happened to me? Do I need to unzip files before uploading or manually decompress files using some galaxy tools? Thanks!

unzip galaxy gz chip-seq • 165 views
ADD COMMENTlink modified 8 weeks ago • written 8 weeks ago by jamwest10120
1

It happened to me just now.

ADD REPLYlink written 8 weeks ago by atcglj10

This has been reported a few times. The Main server at http://usegalaxy.org is currently undergoing updates. More feedback soon. Jen, Galaxy team

ADD REPLYlink written 8 weeks ago by Jennifer Hillman Jackson21k

Thank you! Problem solved.

ADD REPLYlink written 8 weeks ago by jamwest10120

Admin change: marked answer as accepted. Next time you can do this yourself to help others when an answer resolves a problem. Thanks! Jen, Galaxy team

ADD REPLYlink written 8 weeks ago by Jennifer Hillman Jackson21k
2
gravatar for Martin Čech
8 weeks ago by
Martin Čech ♦♦ 3.7k
United States
Martin Čech ♦♦ 3.7k wrote:

This is now fixed on usegalaxy.org.

They way it works now is that .gz files do not uncompress on upload but rather on tool run. You can now select the fastq.gz files as inputs for tools that expect fastq and Galaxy will uncompress them before sending to the tool. See example below.

enter image description here

ADD COMMENTlink written 8 weeks ago by Martin Čech ♦♦ 3.7k

I would love to be able to do this on a local instance. Is this feature only available on the main public server or will it be in the next release as well?

ADD REPLYlink written 8 weeks ago by jdv20
1

We will release Galaxy 17.01 (which contains this feature) in the upcoming weeks. If you don't want to wait you can clone it already from our repo (branch release_17.01) but beware that there might be more bugs than usual since we are still testing it.

ADD REPLYlink written 8 weeks ago by Martin Čech ♦♦ 3.7k

I can wait, but this is great - it is a long-desired feature to be able to handle compressed read data transparently. Is this a fundamental datatype (i.e. can it be set as an output format for tools that can natively write compressed fastq or which we can pipe through gzip when history size is more important than speed of execution)?

ADD REPLYlink written 8 weeks ago by jdv20

Correct, these are new datatypes (e.g. fastq.gz, fastqsanger.gz). It works in a backward compatible manner and allows for future tools to be adapted to work with .gz archives efficiently.

ADD REPLYlink written 8 weeks ago by Martin Čech ♦♦ 3.7k

Hi,

I was hoping that what you described was that case. But when I uploaded my fastq.gz files and run Bowtie with them as input (this is a part of a workflow), Bowtie failed to run and this is the error message:

Error aligning sequence. Error: reads file does not look like a FASTQ file terminate called after throwing an instance of 'int' /bin/sh: line 1: 20345 Aborted (core dumped) bowtie -q -p 6 -S -n 3 -e 70 -l 28 --maxbts 800 -y -k 1 -m 1 --best --strata --phred33-quals /galaxy/data/galGal4/bowtie_index/galGal4 /galaxy-repl/main/files/018/641/dataset_18641959.dat > /galaxy-repl/main/files/018/642/dataset_18642032.dat

ADD REPLYlink written 7 weeks ago by Bao Ho50
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 76 users visited in the last hour