trainNMF error

[Lihyunseo]

Hello! I am a student currently studying single-cell genomics and have been deeply impressed by your excellent model.

I tried following the vignettes/SPOTlight_kidney.Rmd that you posted, but encountered an error.

Here is the code I used:

mod_ls <- trainNMF( x = sce, y = spe, groups = sce$type, mgs = mgs, weight_id = "weight", group_id = "type", gene_id = "gene" )

And here is the error message I received:

Error in trainNMF(x = sce, y = spe, groups = sce$type, mgs = mgs, weight_id = "weight", : ids %in% names(mgs) are not all TRUE In addition: Warning message: In max(table(mgs[[group_id]])) : no non-missing arguments to max; returning -Inf

Could you advise on how to resolve this issue?

I look forward to your response. Thank you!

[Lihyunseo]

I followed only your vignettes/SPOTlight_kidney.Rmd!

[532at]

Hello, did you get any response? I encountered the same error, but I see most of issues are not answered? Perhaps, nobody is no longer working on that software?

[Lihyunseo]

Thank you for the email. Unfortunately, I haven't been able to solve the problem either. 😢 However, there are many other good tools available. How about comparing it with some of them?

-----Original Message----- From: "Adrian @.> To: @.>; Cc: @.>; @.>; Sent: 2024-06-05 (수) 05:13:42 (GMT+09:00) Subject: Re: [MarcElosua/SPOTlight] trainNMF error (Issue #116)

Hello, did you get any response? I encountered the same error, but I see most of issues are not answered? Perhaps, nobody is no longer working on that software? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

[532at]

Yes, I did try CARD and RCTD which worked for me. Currently, trying to work with cell2location and STdeconvolve. :) How about you?

[Lihyunseo]

Thank you for your reply! I'm preparing a presentation for my class comparing the results of STdeconvolve and SPOTlight. For SPOTlight, the previous version works well. You can find it here: https://github.com/MarcElosua/SPOTlight/tree/spotlight-0.1.7/R I've heard that in R, RCTD is good, and in Python, cell2location is recommended. I'm an undergraduate student at Soongsil University in Seoul, Korea. Who are you?

-----Original Message----- From: "Adrian @.> To: @.>; Cc: @.>; @.>; Sent: 2024-06-05 (수) 23:35:48 (GMT+09:00) Subject: Re: [MarcElosua/SPOTlight] trainNMF error (Issue #116)

Yes, I did try CARD and RCTD which worked for me. Currently, trying to work with cell2location and STdeconvolve. :) How about you? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

[532at]

Yes, but you can run cell2location and take cell type proportions to run downstream analysis in R - working on that currently. I am a graduate student (MD) doing my PhD. Currently, I am working in US, but I am from Poland. Happy to connect on LinkedIn (https://www.linkedin.com/in/adrianperdyan/)

[Lihyunseo]

Thank you for your reply I will go into the link you sent me!

-----Original Message----- From: "Adrian @.> To: @.>; Cc: @.>; @.>; Sent: 2024-06-06 (목) 00:03:55 (GMT+09:00) Subject: Re: [MarcElosua/SPOTlight] trainNMF error (Issue #116)

Yes, but you can run cell2location and take cell type proportions to run downstream analysis in R - working on that currently. I am a graduate student (MD) doing my PhD. Currently, I am working in US, but I am from Poland. Happy to connect on LinkedIn (https://www.linkedin.com/in/adrianperdyan/) — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>