Open xiangpin opened 10 months ago
Thanks for the work, I found when the input is Seurat, we can obtain the features.loadings, but if the input is SingleCellExperiment, this is missing
Seurat
features.loadings
SingleCellExperiment
> pbmc.sce class: SingleCellExperiment dim: 9015 2000 metadata(0): assays(2): counts logcounts rownames(9015): LINC00115 NOC2L ... S100B PRMT2 rowData names(0): colnames(2000): ATCACTTGCTCGAA-1 CCGGAGACTGTGAC-1 ... GCATGATGGGCGAA-1 ATACCACTCGGTAT-1 colData names(5): orig.ident nCount_RNA nFeature_RNA stim ident reducedDimNames(2): PCA HARMONY mainExpName: RNA altExpNames(0): > pbmc.sce |> reducedDim('HARMONY') |> str() num [1:2000, 1:50] 109.5 364.4 64.1 102.3 -206.6 ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:2000] "ATCACTTGCTCGAA-1" "CCGGAGACTGTGAC-1" "CAAGCCCTGTTAGC-1" "GAGGTACTAACGGG-1" ... ..$ : chr [1:50] "HARMONY_1" "HARMONY_2" "HARMONY_3" "HARMONY_4" ... > pbmc.sce |> reducedDim('PCA') |> str() num [1:2000, 1:50] 110.2 367 64.2 103.2 -141.2 ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:2000] "ATCACTTGCTCGAA-1" "CCGGAGACTGTGAC-1" "CAAGCCCTGTTAGC-1" "GAGGTACTAACGGG-1" ... ..$ : chr [1:50] "PC1" "PC2" "PC3" "PC4" ... - attr(*, "varExplained")= num [1:50] 28583 18766 3558 3110 1537 ... - attr(*, "percentVar")= num [1:50] 47.2 30.99 5.88 5.14 2.54 ... - attr(*, "rotation")= num [1:500, 1:50] -0.8427 0.1212 -0.4521 -0.104 -0.0562 ... ..- attr(*, "dimnames")=List of 2 .. ..$ : chr [1:500] "FTH1" "HBB" "FTL" "ISG15" ... .. ..$ : chr [1:50] "PC1" "PC2" "PC3" "PC4" ... > # this is Seurat object as input > pbmc@reductions$harmony@feature.loadings[1:10, 1:3] harmony_1 harmony_2 harmony_3 LINC00115 -1089.81768 422.0957 95.493779 NOC2L -646.07807 3688.8628 471.784832 HES4 -9200.61102 -2606.4641 199.833325 ISG15 -14507.01536 -4591.7228 -175.591820 TNFRSF18 -730.27738 3033.6111 246.001729 TNFRSF4 -2164.10675 3107.5163 430.330156 SDF4 -4078.34267 1318.9736 -516.130637 B3GALT6 36.91902 1457.2042 9.419754 UBE2J2 -2115.69325 1161.7211 91.321004 ACAP3 110.42502 264.2238 118.171979
I want to know How to calculate the feature.loadings? And what is the meaning of feature.loadings in harmony?
feature.loadings
harmony
Thanks for the work, I found when the input is
Seurat
, we can obtain thefeatures.loadings
, but if the input isSingleCellExperiment
, this is missingI want to know How to calculate the
feature.loadings
? And what is the meaning offeature.loadings
inharmony
?