Run error in our own Seurat object

[l-magnificence]

When I run in our own Seurat object, it occurs error

infos1 <- Scissor(bulk_dataset, tumor, phenotype, alpha = 0.05, 
                  family = "cox", Save_file = './processed/Scissor_survival.RData')
Error in array(x, c(length(x), 1L), if (!is.null(names(x))) list(names(x),  : 
  'data' must be of a vector type, was 'NULL'

But when we use Seurat_preprocessing before, it run success

tumor<- Seurat_preprocessing(seurat@assays$SCT@counts, verbose = F)
infos1 <- Scissor(bulk_dataset, tumor, phenotype, alpha = 0.05, 
                  family = "cox", Save_file = './processed/Scissor_survival.RData')

Is there any bug in applying Scissor function to our own seurat object?

[sunduanchen]

For your own seurat object, Scissor requires the normalized data and a constructed nearest neighbor network (slot name: data and RNA_snn).

You can check whether your data has these Scissor inputs. If our code still cannot run, you can send me your seurat data to my email: sunduanchen@gmail.com and I will debug it soon.

Thanks, Duanchen

[l-magnificence]

Yes, my data have normalized data and a constructed nearest neighbor network. I think the problem may be the process of seurat. In your Seurat_preprocessing function, the object only do the LogNormalize function, so the downstream analysis in Scissor only accept LogNormalized seurat. But if one use SCTransform to normalize the data and FindNeighbors in slot SCT, there may occur error.

[sunduanchen]

Thank you very much for your feedback. I will make an update to allow users to choose the desired slot.

[Eddylyf]

I have the same issue!

[hyjforesight]

agree with @l-magnificence . I changed from SCTransform back to the LogNormalized workflow. this error was solved.

[Miooooooo]

My input data is integrated and my sc_dataset@graphs list have two dataframe named "integrated_nn" and "integrated_snn" so I got the same error. I can rename the fies to sc_dataset@graphs$RNA_snn to deal with this error. But I confuse about sample `s batch when using Scissor, better remove or not ?

[AClab-sgarcia]

@sunduanchen Has the correction been made to be able to use SCT?

[hyjforesight]

@Miooooooo is correct. Merge2.combined@graphs$RNA_snn=Merge2.combined@graphs$integrated_snn plays the trick.