search

Danko-Lab / TED

a fully Bayesian approach to deconvolve tumor microenvironment
60 stars 10 forks source link

Issues with snowfall - parallel execution #23

Closed albertop210 closed 2 years ago

albertop210 commented 2 years ago

Hi authors,

thanks for your excellent work and very useful deconvolution tool! I tried to use your package (I updated it just yesterday after looking at other issues), following the tutorial analysis in the vignette; everything works well until I launch run.Ted function. I paste here the output code:

[1] "removing non-numeric genes..."
[1] "removing outlier genes..."
Number of outlier genes filtered= 6
[1] "aligning reference and mixture"
[1] "run first sampling"
Start run... This may take a while 
R Version:  R version 4.1.2 (2021-11-01) 

snowfall 1.84-6.1 initialized (using snow 0.4-4): parallel execution on 8 CPUs.

After this, the CPU is starting working, but with a strange "pattern" looking at the system monitor, then it remains like that (without showing any other output for hours); I'm wondering what does snowfall do inside run.Ted function and if there could be a compatibility problem with the parallelization function, the R version or the system. I'm working on Ubuntu 18.04.6 LTS, AMD Opteron 6344 18 CPU.

Thanks in advance for your kind reply, Best,

Alberto

tinyi commented 2 years ago

Hi Alberto,

It indeed will take a while (especially if using only 8 cores). If you are only going to test the data. you may subset your bulk data using X <- X[1:20,1:5000] to speed up computation.

We are preparing a new version which will show an estimated run time for each step.

Best,

Tinyi

On Fri, May 20, 2022 at 4:27 AM Alberto Peloso @.***> wrote:

Hi authors,

thanks for your excellent work and very useful deconvolution tool! I tried to use your package (I updated it just yesterday after looking at other issues), following the tutorial analysis in the vignette; everything works well until I launch run.Ted function. I paste here the output code:

[1] "removing non-numeric genes..." [1] "removing outlier genes..." Number of outlier genes filtered= 6 [1] "aligning reference and mixture" [1] "run first sampling" Start run... This may take a while R Version: R version 4.1.2 (2021-11-01)

snowfall 1.84-6.1 initialized (using snow 0.4-4): parallel execution on 8 CPUs.

After this, the CPU is starting working, but with a strange "pattern" looking at the system monitor, then it remains like that (without showing any other output for hours); I'm wondering what does snowfall do inside run.Ted function and if there could be a compatibility problem with the parallelization function, the R version or the system. I'm working on Ubuntu 18.04.6 LTS, AMD Opteron 6344 18 CPU.

Thanks in advance for your kind reply, Best,

Alberto

— Reply to this email directly, view it on GitHub https://github.com/Danko-Lab/TED/issues/23, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB4NHS7RP3U55KTILJTREHDVK5EHZANCNFSM5WOWCBQQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

albertop210 commented 2 years ago

Thanks a lot for your reply, I will try to subset the X matrix as you suggested, just to give a try with your tutorial data, than go on with my lab's data. For sure an estimated run time printed on screen would be very useful, so I will wait for next updates of the package. Best,

Alberto