Dear Symphony developer team,
First of all, thanks a lot for this great package and vignette!
Following the steps of vignette, I got good results with all my data sets except with one for which I am running into following error:
s = irlba(ref_exp_scaled, nv = 20)
Error in irlba(ref_exp_scaled, nv = 20) :
BLAS/LAPACK routine 'DLASCLS' gave error code -4
I noticed that there are many NAs (115200) in the ref_exp_scaled (1:789, 1:19200), which might cause this problem.
Running ref_exp_scaled = singlecellmethods::scaleDataWithStats(ref_exp, vargenes_means_sds$mean, vargenes_means_sds$stddev, 1)
does not give an error and ref_exp, vargenes_means_sds$mean and vargenes_means_sds$stddev do not contain any NAs.
For troubleshooting, I tried to run the contents of the function scaleDataWithStats, but can't find the function scaleRowsWithStats_dgc. My dataset is unfortunately not public. ref_exp is a Large dgCMatrix with 472536288 elements, 387.6 MB.
Do you have any idea what could cause this problem?
Thanks a lot in advance!
Dear Symphony developer team, First of all, thanks a lot for this great package and vignette!
Following the steps of vignette, I got good results with all my data sets except with one for which I am running into following error:
I noticed that there are many NAs (115200) in the ref_exp_scaled (1:789, 1:19200), which might cause this problem. Running ref_exp_scaled = singlecellmethods::scaleDataWithStats(ref_exp, vargenes_means_sds$mean, vargenes_means_sds$stddev, 1) does not give an error and ref_exp, vargenes_means_sds$mean and vargenes_means_sds$stddev do not contain any NAs. For troubleshooting, I tried to run the contents of the function scaleDataWithStats, but can't find the function scaleRowsWithStats_dgc. My dataset is unfortunately not public. ref_exp is a Large dgCMatrix with 472536288 elements, 387.6 MB.
Do you have any idea what could cause this problem? Thanks a lot in advance!