Map with KMA using the parameter "-t_db" is not clear!

usegalaxy.eu support
, ,
1

I’m trying to use KMA and am confused with the parameter “-t-db”.

Assuming I have the ref database. how can I plug that into the tool only by specifying the index? should I upload my database onto galaxy and then use the index parameter of kma?

can anyone who have used kma in galaxy please help me I’m quit confused.

thanks in advanced,

Ahmad Mahagna

2

Hi @Smkingsize

Which exact tool are you using (full tool name/version from the top of the tool form) and at which public Galaxy server (URL from the top of the browser window)?

Big picture, using your own reference genome with tools in Galaxy is very simple and involves just the fasta file plus possibly a reference annotation file.

The first few topics in this search cover how these are used → Search results for 'custom genome build' - Galaxy Community Help

Please give that a review and let us know if it actually helps! I’ll watch for the tool/server details if not. :slight_smile:

1 Like
3

Hey @jennaj ,

I am on the European server, trying to use the tool: Map with KMA (Galaxy Version 1.5.1+galaxy0).

which requires a FASTQ and a kma index as input. I’m not sure how the kma index should work. I have a fasta file with my refrence sequences and I did locally used the subtool kma_index (since it’s not on galaxy) to extract the database (.comp.b, .lenght.b, .seq.b, .name). now I don’t know weather I should upload those to galaxy and then specify the index. or what should i do :confused:

Thanks for your support.

Edited:

I found those Topics to be helpful:

Adding a custom database/build (dbkey)

How to use Custom Reference Genomes?

so I’ve created a custom database, But KMA still outputs the error:

parameter kma_index requires a value but no legal value was defined

1 Like
4

Hello @Smkingsize

Thanks for clarifying!

The tool is only hosted at the UseGalaxy.eu server and appears to not be usable at this time. The way the tool is currently written requires that indexes are hosted by the server administrators.

It seem it would be better to:

  1. Create an indexing tool, then allow selection of that user-generated index from the history.
  2. Or, to allow the input of the file to index and run that process first at runtime, followed by the the full tool mapping step.

I may be misunderstanding something. Given that the tool is only hosted at the EU server, I’ve decided to make a request for EU team specifically to get more clarity on the expected usage (instead of creating tool development request against the IUC, but that could be next). Maybe it is hosted for development purposes only for now.

This week is really busy for them! But they will see our discussion after the events this week complete (European Galaxy Days (EGD) - Galaxy Community Hub).

Thanks for asking about the tool and we can follow up later on. Feel free to @ me here in a week or so if this doesn’t get addressed by then. :slight_smile:

1 Like
5

Hi @Smkingsize,

we, as admins of Galaxy Europe, have the possibility to install kma indexes on the server that you will be able to select then from the tool interface.

Which genomes are you interested in? If these are not standard ones, can you provide a link to the reference sequences? Ideally you could also post the indexing command you used locally so that we can see if you used any special settings.

Also, I’m interested in kma for some planned own work and would very much like to work properly in Galaxy. So if you find more current limitations or bugs with this tool, please keep reporting them!

Thanks,

Wolfgang

1 Like
6

Hey @wm75 ,

I require the index of the standard viruses database if it’s installed. Here is a link to the reference sequences history: Galaxy History with the reference seqences file

I didn’t use any special parameters when I’ve indexed the FASTA file locally.

In my opinion, it would be very helpful to wrap the kma index and allow kma to take a fasta file as input from history, index it and use it. KMA index. or optionally to post somewhere the databases which are installed on the server with their indexes so that users can look them up and use them inside kma.

I will keep reporting regarding kma tool.

Thank you,

Ahmad Mahagna

1 Like
7

Thanks @Smkingsize,

changes to the tool interface will take a bit more time (like maybe a week from now).

In the meantime, any genome added to the existing dropdown menu would be available for all users (and not count towards anyones quota). For this reason, sources should be clearly defined. So when you’re saying “the standard viruses database” what’s the original source of that?

Cheers,

Wolfgang

8

Hey @wm75,

thanks for the quick reply!

Since I’m trying to get the tool running first to understand it. I would like to try “NT DB restricted to Viruses“ since it’s locally installed.

I’m new to the topic so gaining understatement and getting the tool running is the first priority and then I will work on choosing maybe better suited databases.

I also don’t quit get what existing dropdown menu are you refering to.

regards,

Ahmad Mahagna

9

a) What do you mean by locally installed? What we need for installation on Galaxy Europe is a public link to the original ref data (your previous link to a Galaxy history is not enough, unfortunately, but it needs to be a well-known, possibly citeable resource).

b) Drop-down menu was misleading maybe. What I meant is the “kma_index” select param that currently says “No options available”. This will turn into a drop-down menu once there are options available. Sorry for that confusion.

10

Thanks for the clarification!

here is a link to where I got the reference sequences from: ref sequences

and that’s the description of what is included in the database: GitHub - NLKaiser/VirDetector

regards,

Ahmad

1 Like
11

Ok, your requested kma index is now available on Galaxy Europe :slight_smile:

1 Like
12

HI Wolfgang, I have interest in the addition of BBsplit from BBtools, so that I can use multiple references and will competitively align to the different references. I have several closely related species that I’m having difficulty separating out. I have 3 plasmodium species that are very similar, as well as a Naegleria Fowleri.

From what I have read, this seems to be the best tool for this??