search

Danko-Lab / TED

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

1 current sample ID:1 Error in FUN(X[[i]], ...) : argument "seed" is missing, with no default #18

Open vincent1715 opened 2 years ago

vincent1715 commented 2 years ago

Hello I have some problem when running function "learn.embedding.Kcls()" as follows:

tcga.ebd.res.k4 <- learn.embedding.Kcls (ted.res = tcga.ted,

  • K.vec = 4,
  • EM.maxit=50,
  • n.cores =1) current.K= 4 [1] "starting EM cycles #..." 1 current sample ID:1 Error in FUN(X[[i]], ...) : argument "seed" is missing, with no default

I set the argument "n.cores =1" as i am running this on windows which Tom suggested before, the inputs is the example data you provide. can you give me some suggestion? thanks .

Regards,

vincent

tinyi commented 2 years ago

Hi Vincent,

I have now fixed this issue - something I forgot to do after adding the random seed that addresses Tom's request.

In our BayesPrism paper, we provided phi.tum and K by first running NMF on inferred tumor expression. This might provide a more sensible way to choose K and initialize phi.tum.

I have now updated our vignette to include this.

Feel free to let me know if you have any questions.

Best,

Tinyi

On Fri, Apr 15, 2022 at 6:07 AM vincent1715 @.***> wrote:

Hello I have some problem when running function "learn.embedding.Kcls()" as follows:

tcga.ebd.res.k4 <- learn.embedding.Kcls (ted.res = tcga.ted,

-

                                    K.vec = 4,

-

                                    EM.maxit=50,

-

                                    n.cores =1)

current.K= 4 [1] "starting EM cycles #..." 1 current sample ID:1 Error in FUN(X[[i]], ...) : argument "seed" is missing, with no default

I set the argument "n.cores =1" as i am running this on windows which Tom suggested before, the inputs is the example data you provide. can you give me some suggestion? thanks .

Regards,

vincent

— Reply to this email directly, view it on GitHub https://github.com/Danko-Lab/TED/issues/18, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB4NHSYNUDZB4ZE366ZZCDDVFE5VFANCNFSM5TQDFA2Q . You are receiving this because you are subscribed to this thread.Message ID: @.***>

vincent1715 commented 2 years ago

Hi Tinyi, very appreicate for your advice and corresponding help, and sorry for replying so late as i was testing the code you provide, and its really need so long time to test. The modification you've made is feasible to solve my problem,but i would continually do more test with other data, and i might ask for more help if any other problem happen. thanks for your timely help.

Best,

Vincent