Concatenating issue when plotting analytic_pearson_residuals layers

[ngvananh2508]

when I ran these code which followed Single-cell best practices:

analytic_pearson = sc.experimental.pp.normalize_pearson_residuals(adata, inplace = False)
adata.layers["analytic_pearson_residuals"] = csr_matrix(analytic_pearson["X"])
p2 = sns.histplot(adata.layers["analytic_pearson_residuals"].sum(1), bins = 100, kde = False, ax = axes[1])

it appears this error: ValueError: No objects to concatenate.

I am running seaborn version 0.12.2 and scanpy version 1.9.3

Can anyone please help me with this issue?

Thank you so much!

[lconan]

I just ran into this error. I think it's on sc.experimental.pp.normalize_pearson_residuals where nan's are generated because of divided by 0 issue, at least in my case, and I don't know how to solve this

[ngvananh2508]

I just ran into this error. I think it's on sc.experimental.pp.normalize_pearson_residuals where nan's are generated because of divided by 0 issue, at least in my case, and I don't know how to solve this

Do you get the kernel death problem when loading adata to robjects on jupyter lab in the feature selection part?

[lconan]

I just ran into this error. I think it's on sc.experimental.pp.normalize_pearson_residuals where nan's are generated because of divided by 0 issue, at least in my case, and I don't know how to solve this

Do you get the kernel death problem when loading adata to robjects on jupyter lab in the feature selection part?

No, I didn't. But I did get: ValueError: Unknown dtype dtype('int64') cannot be converted to ?gRMatrix.

[ngvananh2508]

I just ran into this error. I think it's on sc.experimental.pp.normalize_pearson_residuals where nan's are generated because of divided by 0 issue, at least in my case, and I don't know how to solve this

Do you get the kernel death problem when loading adata to robjects on jupyter lab in the feature selection part?

No, I didn't. But I did get: ValueError: Unknown dtype dtype('int64') cannot be converted to ?gRMatrix.

Are you practicing with single-cell best practice? May I keep contact with you in case if we have bugs, we can assist each other?

[ngvananh2508]

I solved all problems by checking the right file to run.

[lconan]

I solved all problems by checking the right file to run.

I managed to solve this by converting the original count back to dense matrix and pass only .X to R functions.....I can't do this on test samples, because I do this on published dataset of others.

[ngvananh2508]

When I did like you on test sample, there was not any plot for this data.