Forum:GalaxyProject, why and why not?

Question

[support]

Guys,

The galaxy project is most popular and used across. Can you please share your thoughts on,

1. Why you are using galaxy project?

2. Why you are not opting for it?

   I really need to understand the good and limitation of galaxy project. Thanks.

Thanks

[Devon Ryan]

I'm reopening this, mostly because the Galaxy project has changed so much in the last couple years that the reasons for using it have expanded.

[Michael]

Then it needs to be forum, because it is mainly a opinion question.

[Devon Ryan]

Indeed, good catch!

[GouthamAtla]

Why Should I Use Galaxy ?

[Michael]

Hello support!

Questions similar to yours can already be found at:

• Why Should I Use Galaxy ?

We have closed your question to allow us to keep similar content in the same thread.

If you disagree with this please tell us why in a reply below. We'll be happy to talk about it.

Cheers!

[support]

HI Michael, This not only about why but also I'm not using galaxy. Thanks.

[Ram]

Erm, sorry, what do you mean to say?

[harold.smith.tarheel]

I think it means 'what reasons do you have for not using Galaxy?'

[Ram]

Ah, I see. Sorry, my language parser exploded.

Answer 1

[Devon Ryan]

I run two Galaxy instances, so my list of reasons for using Galaxy will be rather more extensive. However, below are the reasons that one might not want to use Galaxy:

1. Workflows don't support pipes, so you end up creating a LOT more files than if you scripted something (of course, the target audience for Galaxy doesn't normally know how to use the command line...so this is a rather moot point).
2. Half the software changes require a server restart (granted, if you're then running the actual jobs on a cluster then this doesn't matter so much, but I swear everytime I do this I get an email from someone complaining).
3. I at least personally seem prone to finding odd and unreproducible bugs that result in me needing to manually modify the database. I'm pretty sure I'm the only person this happens to, though.
4. It hates compressed fastq and unsorted BAM files.

Reasons to use Galaxy:

1. It already does what we need.
2. There are already wrappers for almost anything we want.
3. If there isn't a wrapper already, they're simple to write.
4. Installation is really really really easy (docker pull ...)
5. Integrates with our cluster nicely.
6. Integrates with RStudio/ipython/jupyter and now some of the iobio stuff.
7. I only spend a few minutes each week doing admin stuff.
8. Our wetlab folks can easily be taught how to do a lot of things in it and are comfortable using it.
9. It enables nice reproducibility.

1 and 8 are the biggest selling points for me.

[John]

1. It's free

[John]

Ooh, a bug in how Biostars treats "10. something"

[Istvan Albert]

I think it is not a bug, more like it is the Markdown standard or shall we say the way it is not specified and open to interpretation.

Technically the markdown marker for a numbered list is the 1. you can just list them as such 1. 1. has the same effect as 1. 2.

In addition in the original spec one cannot start the counter at a higher number but they can in the common-mark spec.

The only solution right now would be patching the markdown2 library (FWIW I had to patch mozilla bleach as well based on uses cases seen in Biostar)

[Istvan Albert]

Also this made me do a little more research, looks like there is a common mark module for python as well.

[Devon Ryan]

Good find, can you let Istvan know?

[Ram]

1. down for what? #testing

Answer 2

[san.san]

In my experience people will use Galaxy because of it's user friendly format and unwillingness to use the command line. I don't use it since I like to have all my files, etc. in one place and I'm using an HPC cluster at work and doing analysis on both the cluster and Galaxy means dealing with having my stuff in different places.

Hope this helps!