Re: [Chipster-users] Chipster v4 and the Import Tool (microarray expression data)

[Oliver H. <o....@dk...>]

Hi Eija,

no, no, I think we are on a side track here.
The problem is not about lumi.
You brought lumi up as a tool for data imported as a whole table. For 
this to work,
the table needs to be in a specific format as described in lumis 
documentation.

I just said, that I tried with lumi (for our specific data table, which 
is not in the right
format) and I got an error. That's totally fine. It should not work with 
our specific data
table. We can bring it into the right format, to have a workaround. Lumi 
is fine!

A more general way to use tables is having a split tool. That's where I 
am still on.

And all this around, distracting from the real problem I try to solve, 
are those
issues with containers, chipster not restarting, and so on. Actually, my 
opinion
is, the whole container architecture is a mess. Not only the specific 
chipster
architecture, but containers in general. Well, I have to admit, I am 
still not an
expert when it comes to containers/docker/...whatever, my motivation to 
become
an expert on containers is literally zero.

So, let me role us all back in time: I just want to have the 
documentation of
my new split tool visible in our chipster installation (other thread 
with Petri), but
I am failing miserably. There is no current issue with lumi.

Best,

Oli

Am 29.11.2022 um 14:57 schrieb chi...@cs...:
> Hi Oliver,
>
> Sorry to hear that running the tool "Illumina - lumi pipeline" brought 
> Chipster in to a bad state! I just tested lumi on our server and it 
> produced normalized data and phenodata without any error message. 
> While our technical guys can hopefully help you to sort out the server 
> situation, I would like to ask more about the actual lumi job you were 
> trying to run.
>
> You wrote that the imported file does not comply to the needed format 
> like column names. I'm puzzled by this, because when you click the 
> green "Add file" button it brings the file in exactly as it is. For 
> the lumi pipeline the input file is the BeadStudio / GenomeStudio 
> output file, so no conversion is needed. I tested with a file from 
> BeadStudio version 1.5.0.34, but if you send me a newer file I can 
> test with that too.
>
> Best,
> Eija--
>
> Oliver Heil
> Microarray Core Facility
> Bioinformatics
>
> German Cancer Research Center (DKFZ)
> Foundation under Public Law
> Im Neuenheimer Feld 580
> 69120 Heidelberg
> Germany
>
> o....@dk... <mailto:o....@dk...>
> Support: www.dkfz.de/gpcf/phpBB3/index.php 
> <https://www.dkfz.de/gpcf/phpBB3/index.php>
> www.dkfz.de <http://www.dkfz.de>
>
> Research for a Life without Cancer
>
> Management Board: Prof. Dr. Michael Baumann, Ursula Weyrich
> VAT-ID No.: DE143293537
>