GregorySchwartz/too-many-cells

error, called at src/Data/Sparse/SpMatrix.hs:793:22 in sparse-linear-algebra-0.3.2-LJ5QWDyWqk8g1L9MLmVBX:Data.Sparse.SpMatrix

Roger-GOAT opened this issue · 5 comments

Hi, I am running a 10X data in Docker in a Linux Webserver. The dir has three cellrange output file like matrix, barcodes and features. I created labels.csv like
vip06@tpm5-desktop:~$ head ./labels.csv
AAACCCAAGAAACACT-1,day14
AAACCCAAGAAACCAT-1,day14
AAACCCAAGAAACCCA-1,day14
AAACCCAAGAAACCCG-1,day14
AAACCCAAGAAACCTG-1,day14
AAACCCAAGAAACGAA-1,day14
AAACCCAAGAAACGTC-1,day14
AAACCCAAGAAACTAC-1,day14
AAACCCAAGAAACTCA-1,day14

It showed error like below:

vip06@tpm5-desktop:~$ docker run -it --rm -v "/home/data/vip06:/home/data/vip06" gregoryschwartz/too-many-cells:2.0.0.0 make-tree --matrix-path /home/data/vip06/scdata/allraw/day14 --labels-file /home/data/vip06/labels.csv --draw-collection "PieRing" --output /home/data/vip06/result/
too-many-cells: matMat : incompatible matrix sizes((31053,6794880),(1671225,1)).............................................] 18%
CallStack (from HasCallStack):
error, called at src/Data/Sparse/SpMatrix.hs:793:22 in sparse-linear-algebra-0.3.2-LJ5QWDyWqk8g1L9MLmVBX:Data.Sparse.SpMatrix

Thanks a lot!

It looks like you did not change the filtering thresholds. It's possible that cells are being filtered out, depending on their value. What if you specify -F?

Thanks for the reply. I check the workshop at https://gregoryschwartz.github.io/too-many-cells/workshop/workshop.html, didn't find what -F means? Do you mean --filter-thresholds? I add --filter-thresholds "(250, 1)" \ and try
vip06@tpm5-desktop:~$ docker run -it --rm -v "/home/data/vip06:/home/data/vip06" gregoryschwartz/too-many-cells:2.0.0.0 make-tree --matrix-path /home/data/vip06/scdata/allraw/day14 --labels-file /home/data/vip06/labels.csv --filter-thresholds"(250, 1)" --output /home/data/vip06/result/
Invalid option --filter-thresholds(250, 1)' it showed no Invalid option --filter-thresholds(250, 1)'

-F is from older versions which I think that docker uses. You can customize the --filter-thresholds argument to make sure no filtering is done (I don't know what your data looks like so I don't know if it's normalized or has some other kind of count data).

Also, you need a space in your argument --filter-thresholds "(250, 1)", but also you should choose different numbers to be lower depending on your data.

Closing due to inactivity.