I encountered a problem when reading a loom file whose col_attrs include columns that are matrices. In R, data frames can have matrices as columns, so I think those should be able to be managed, or moved to reducedDims. Here's what the col_attrs look like when I debug .importLoom_DataFrame:
Browse[1]> str(df)
List of 27
$ 10x_cluster : chr [1:271679(1d)] "2_H27a" "2_H27a" "2_H27a" "2_H27a" ...
$ CellID : chr [1:271679(1d)] "aaaabljk-1_dev_HE27a" "aaaadhll-1_dev_HE27a" "aaaaiade-1_dev_HE27a" "aaaaodnc-1_dev_HE27a" ...
$ CellTypes : chr [1:271679(1d)] "? Face mesenchyme - anterior" "? Face mesenchyme - anterior" "? Face mesenchyme - anterior" "? Face mesenchyme - anterior" ...
$ Clusters : int [1:271679(1d)] 0 0 0 0 0 0 0 40 0 0 ...
$ NonNeural_ClusterText : chr [1:271679(1d)] "Anterior chondrogenic mesenchyme" "Anterior chondrogenic mesenchyme" "Anterior chondrogenic mesenchyme" "Anterior chondrogenic mesenchyme" ...
$ TP_35_Recombined3Clean2_C_FBPC: num [1:35, 1:271679] 0.052657 0.023825 0.001418 0.017986 0.000611 ...
$ UMAP : num [1:2, 1:271679] 8.38 3.38 8.7 3.95 8.79 ...
$ X : num [1:271679(1d)] 3879 3948 3969 3921 3949 ...
$ X_umap : num [1:271679(1d)] 8.38 8.7 8.79 8.4 8.28 ...
$ Y : num [1:271679(1d)] 32028 32014 32049 32080 32105 ...
$ Y_umap : num [1:271679(1d)] -3.38 -3.95 -3.8 -3.59 -3.5 ...
Here's the culprit where the matrices messed up:
|
df[] <- lapply(df, as.vector) |
Here's the error I get: Error in DataFrame(df) : cannot coerce class "list" to a DataFrame
I encountered a problem when reading a
loomfile whosecol_attrsinclude columns that are matrices. In R, data frames can have matrices as columns, so I think those should be able to be managed, or moved toreducedDims. Here's what thecol_attrslook like when I debug.importLoom_DataFrame:Here's the culprit where the matrices messed up:
LoomExperiment/R/import-method.R
Line 43 in dce7132
Here's the error I get:
Error in DataFrame(df) : cannot coerce class "list" to a DataFrame