Parallel run about spotlight HOT 2 CLOSED

Thanks ! I have one Seurat with 4 slides.

nmf support DoParallel. can you activate it in the future ?

.pbackend | specification of the foreach parallel backend to register and/or use when running in parallel mode. See options p and P in argument .options for how to enable this mode. Note that any backend that is internally registered is cleaned-up on exit, so that the calling foreach environment should not be affected by a call to nmf – except when .pbackend=NULL.

https://rdrr.io/cran/NMF/man/nmf.html

from spotlight.

Hi @BenjaminDEMAILLE ,

Thank you for your interest in SPOTlight, currently this has not been implemented yet.

If you want to optimize time what I would suggest is training the model once with trainNMF.R and using its output to pass it to runDeconvolution.R for each visium slide of interest. Make sure the entire gene set used is maintained in the trainNMF step!

Hope this helps

from spotlight.