cannot convert to .loom from-cellranger

[Eneritz]

Hi, I am trying to convert 10X data to loom file. I already did it with some previous version, now with the new version I cannot.

• I do not get any error, but the output folder is empty
• Did the command line usage change? ( I was doing the conversion through command line)
• as input I use the input for ( -i ../outs/ ) and out folder ( -o ../file.loom )
• my data is in GRCh38

loom v0.22.26

thanks E

[slinnarsson]

Hi

The cellranger output changes slightly with each version, so maybe your folder is a bit different. The Python code should work with cellranger 1.1 and 1.2, not sure about later versions.

Can you try to do the conversion from code, rather than use the command? You would do this:

import loompy loompy.create_from_cellranger(in_folder, out_file)

Because we have not been keeping the command-line tool up to date.

/Sten

-- Sten Linnarsson, PhD Professor of Molecular Systems Biology Karolinska Institutet Unit of Molecular Neurobiology Department of Medical Biochemistry and Biophysics Scheeles väg 1, 171 77 Stockholm, Sweden +46 8 52 48 75 77tel://+46%208%2052%2048%2075%2077 (office) +46 70 399 32 06tel://+46%2070%20399%2032%2006 (mobile)

On 27 April 2017 at 14:18:53, Eneritz (notifications@github.commailto:notifications@github.com) wrote:

Hi, I am trying to convert 10X data to loom file. I already did it with some previous version, now with the new version I cannot.

• I do not get any error, but the output folder is empty
• Did the command line usage change? ( I was doing the conversion through command line)
• as input I use the input for ( -i ../outs/ ) and out folder ( -o ../file.loom )
• my data is in GRCh38

loom v0.22.26

thanks E

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/linnarsson-lab/Loom/issues/100, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AKKag4y25SGTn0V6cw3P12E-YSBM8FkLks5r0IeegaJpZM4NKIrB.

[JobLeonard]

I also have to check if the cellranger code still creates the same time of metadata annotations as the other definitions - I may not have updated those.

Although I suspect this would only affect the loom data browser, not loompy usage within Python.

[JobLeonard]

Ok, should be fine. I got confused with CEF for a second.

[Eneritz]

Hi, Ok, I will run from the code.

• I am using outputs from cell ranger v 1.3.1

• I got an error, because of the genome assembly. This I have edited directly in the code, but probably you can choose the assembly when running _loompy.create_fromcellranger .

• The outputs from cell ranger 1.3.1 changed a little the folder structure. I guess, that is the reason why I get some errors,

this is your code kmeans = np.loadtxt(os.path.join(folder, "analysis", "kmeans", "10_clusters", "clusters.csv"), usecols=(1, ), delimiter=',', skiprows=1)

the actual path in v 1.3.1 outs/analysis/clustering/kmeans_10_clusters/clusters.csv

Anyway, I could run it with some data that I have from previous versions of cell ranger. Editing some lines from the code I can also run it for v 1.3.1

Thanks E

[pl-ki]

This should be solved in code now. Use updated loompy package or velocyto on commandline.