Error at 'error in evaluating the argument 'x' in selecting a method for function 't': non-conformable arrays'

[shivUSF]

Thanks for making this available, I am trying to annotate 79 sales bone marrow samples against the azimuth 'bonemarrowref' This is my Seurat object: An object of class Seurat 36714 features across 502620 samples within 2 assays Active assay: RNA (36601 features, 2000 variable features) 3 layers present: scale.data, data, counts 1 other assay present: ADT

The code I am using is - megeredObj <- RunAzimuth(megeredObj, reference = "bonemarrowref")

This is the error it gives every time - Warning: Overwriting miscellanous data for modelWarning: Adding a dimensional reduction (refUMAP) without the associated assay being presentWarning: Adding a dimensional reduction (refUMAP) without the associated assay being presentdetected inputs from HUMAN with id type Gene.name reference rownames detected HUMAN with id type Gene.name Normalizing query using reference SCT model Warning: 28 features of the features specified were not present in both the reference query assays. Continuing with remaining 2972 features.Projecting cell embeddings Error in h(simpleError(msg, call)) : error in evaluating the argument 'x' in selecting a method for function 't': non-conformable arrays

I guess the warnings can be ignored but post Projecting cell embeddings Its gives this simple error. I went through the issue, didn't find any one else having similar issue.

[shivUSF]

This is the trace back of where the error starts -

15. h(simpleError(msg, call)) 14. .handleSimpleError(function (cond) .Internal(C_tryCatchHelper(addr, 1L, cond)), "non-conformable arrays", base::quote(y - mu)) 13. pearson_residual(y, mu, model_pars[genes_bin, "theta"], min_var = min_var) 12. get_residuals(vst_out = vst_out, umi = umi, residual_type = "pearson", min_variance = min_var, verbosity = as.numeric(x = verbose) *2) 11. FetchResiduals_reference(object = query, reference.SCT.model = reference.SCT.model, features = features, nCount_UMI = nCount_UMI) 10. ProjectCellEmbeddings.default(query = LayerData(object = query, layer = layers.set[i]), reference = reference, reference.assay = reference.assay, reduction = reduction, dims = dims, scale = scale, normalization.method = normalization.method, verbose = verbose, features = features, nCount_UMI = nCount_UMI[Cells(x = query, ...

Hope this helps.

[hvbakel]

Can you try collapsing layers before running Azimuth on your object?

megeredObj = JoinLayers(megeredObj)

[shivUSF]

I did join all the layers and it has the SCT object as well in the object.

An object of class Seurat 63752 features across 502620 samples within 3 assays Active assay: SCT (27038 features, 3000 variable features) 3 layers present: counts, data, scale.data 2 other assays present: RNA, ADT 3 dimensional reductions calculated: pca, inte_harmony, umap

is there some column name causing the problem - This is the colnames in metadata -

"capID","nUMI","nGene", "nCount_ADT", "nFeature_ADT", "log10GenesPerUMI", "mitoRatio" , "mitoFr", "riboRatio", "riboFr", "log10ADTPerUMI", "cellID", "input", "outliers", "Sample", "Disease", "Diagnosis", "MRD", "Race", "Age" , "Isotype" , "VitalStatus" , "nCount_RNA" , "nFeature_RNA" , "DF_pANN" , "DF_classifications" "nCount_SCT" , "nFeature_SCT" , "SCT_snn_res.1.1" "seurat_clusters"

[immunemd]

I'm dealing with the same problem.

[aminards]

I am getting the same error. Was a resolution ever found for this error?