SetDatExpr error - not finding group_name despite it existing

[jmuribes]

Hi! Thank you so much for creating hdWGCNA, such a powerful tool.

I'm currently having an issue where after creating the Metacells, when I try to set up the gene expression matrix focusing on a level of the grouping variable I get an error that the group_name is not found. This issue happens with some group_names but not all, so I'm not sure what could be causing the issue. I double checked that the levels of the grouping variable does in fact exist.

Thank you!

Steps to reproduce

My Seurat object includes the variables "orig.ident" for the samples and "clusters_hdwgcna" for the clusters to be tested. After setting up for the WGCNA, I create the metacells using these two variables (orig.ident and clusters_hgwgcna). This runs without any issues. But when I set up an expression matrix focusing on one of the levels of my cluster_hdwgcna variable, I get an error that it cannot be found. This does not happen to all levels, some run without a problem. I double checked that the level of cluster I'm focusing on does exist in the "cluster_hdwgcna" in the meta.data.

# Set up the Seurat object for hdWGCNA:
> all.data.combined.neurons.filt.hdwgcna <- SetupForWGCNA(
  all.data.combined.neurons.filt.hdwgcna,
  gene_select= "fraction",
  fraction= 0.05,
  wgcna_name = "srgap2"
)

# Get metacells:
> all.data.combined.neurons.filt.hdwgcna <- MetacellsByGroups(
  seurat_obj = all.data.combined.neurons.filt.hdwgcna,
  group.by= c("orig.ident", "clusters_hdwgcna"),
  reduction = 'pca',
  k = 25,
  max_shared=10,
  ident.group = 'clusters_hdwgcna'
)

# Normalize metacells:
> all.data.combined.neurons.filt.hdwgcna <- NormalizeMetacells(all.data.combined.neurons.filt.hdwgcna)

# Set expression matrix for hdWGCNA:
> all.data.combined.neurons.filt.hdwgcna <- SetDatExpr(all.data.combined.neurons.filt.hdwgcna,
                                             group_name= "ONL",
                                             group.by= "clusters_hdwgcna",
                                             assay= "RNA",
                                             slot= "data")

Error in SetDatExpr(all.data.combined.neurons.filt.hdwgcna, group_name = "ONL",  :
  Some groups in group_name are not found in the seurat_obj: ONL

# The level "ONL" does exist in "clusters_hdwgcna"
> levels(all.data.combined.neurons.filt.hdwgcna$clusters_hdwgcna)
 [1] "NB"    "NGlut" "FB"    "MB"    "MHB"   "HB"    "RB"    "GCL"   "INL"  
[10] "ONL"   "RPE"   "NCe"   "NP"    "Ot"    "LL"    "PhA" 

R session info
R version 4.2.2 (2022-10-31)
Platform: x86_64-conda-linux-gnu (64-bit)
Running under: Ubuntu 20.04 LTS

Matrix products: default
BLAS/LAPACK: /share/dennislab/programs/dennis-miniconda/envs/r_scirnaseq/lib/libopenblasp-r0.3.21.so

locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=en_US.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] stats4 stats graphics grDevices utils datasets methods
[8] base

other attached packages:
[1] DescTools_0.99.48 enrichR_3.2
[3] ggrepel_0.9.3 UpSetR_1.4.0
[5] slingshot_2.6.0 TrajectoryUtils_1.6.0
[7] princurve_2.1.6 RColorBrewer_1.1-3
[9] tradeSeq_1.12.0 WGCNA_1.72-1
[11] fastcluster_1.2.3 dynamicTreeCut_1.63-1
[13] hdWGCNA_0.2.18 dendextend_1.17.1
[15] anndata_0.7.5.6 sceasy_0.0.7
[17] reticulate_1.28 clustree_0.5.0
[19] ggraph_2.1.0 scater_1.26.1
[21] SC3_1.26.2 MAST_1.24.1
[23] cluster_2.1.4 intrinsicDimension_1.2.0
[25] yaImpute_1.0-33 SingleR_2.0.0
[27] scDblFinder_1.12.0 ggridges_0.5.4
[29] gghalves_0.1.4 ggforce_0.4.1
[31] viridis_0.6.3 viridisLite_0.4.2
[33] scran_1.26.2 scuttle_1.8.4
[35] SingleCellExperiment_1.20.1 SummarizedExperiment_1.28.0
[37] Biobase_2.58.0 GenomicRanges_1.50.2
[39] GenomeInfoDb_1.34.9 IRanges_2.32.0
[41] S4Vectors_0.36.2 BiocGenerics_0.44.0
[43] MatrixGenerics_1.10.0 matrixStats_0.63.0
[45] reshape2_1.4.4 dunn.test_1.3.5
[47] speckle_0.0.3 magrittr_2.0.3
[49] data.table_1.14.8 cowplot_1.1.1
[51] lubridate_1.9.2 forcats_1.0.0
[53] stringr_1.5.0 purrr_1.0.1
[55] readr_2.1.4 tidyr_1.3.0
[57] tibble_3.2.1 tidyverse_2.0.0
[59] patchwork_1.1.2 ggplot2_3.4.2
[61] Matrix_1.5-4 dplyr_1.1.2
[63] SeuratObject_4.1.3 Seurat_4.3.0

loaded via a namespace (and not attached):
[1] rsvd_1.0.5 Hmisc_5.1-0
[3] ica_1.0-3 class_7.3-22
[5] Rsamtools_2.14.0 foreach_1.5.2
[7] lmtest_0.9-40 crayon_1.5.2
[9] MASS_7.3-60 WriteXLS_6.4.0
[11] nlme_3.1-162 backports_1.4.1
[13] qlcMatrix_0.9.7 impute_1.72.3
[15] rlang_1.1.1 readxl_1.4.2
[17] XVector_0.38.0 ROCR_1.0-11
[19] irlba_2.3.5.1 limma_3.54.2
[21] xgboost_1.7.5.1 BiocParallel_1.32.6
[23] rjson_0.2.21 bit64_4.0.5
[25] glue_1.6.2 harmony_0.1.1
[27] pheatmap_1.0.12 rngtools_1.5.2
[29] sctransform_0.3.5 parallel_4.2.2
[31] vipor_0.4.5 spatstat.sparse_3.0-1
[33] AnnotationDbi_1.60.2 spatstat.geom_3.2-1
[35] tidyselect_1.2.0 fitdistrplus_1.1-11
[37] XML_3.99-0.14 zoo_1.8-12
[39] GenomicAlignments_1.34.1 org.Mm.eg.db_3.16.0
[41] xtable_1.8-4 evaluate_0.21
[43] cli_3.6.1 zlibbioc_1.44.0
[45] rstudioapi_0.14 doRNG_1.8.6
[47] miniUI_0.1.1.1 sp_1.6-0
[49] rpart_4.1.19 shiny_1.7.4
[51] BiocSingular_1.14.0 xfun_0.39
[53] tidygraph_1.2.3 KEGGREST_1.38.0
[55] expm_0.999-7 listenv_0.9.0
[57] Biostrings_2.66.0 png_0.1-8
[59] future_1.32.0 withr_2.5.0
[61] slam_0.1-50 bitops_1.0-7
[63] cellranger_1.1.0 plyr_1.8.8
[65] sparsesvd_0.2-2 pcaPP_2.0-3
[67] e1071_1.7-13 dqrng_0.3.0
[69] pillar_1.9.0 cachem_1.0.8
[71] tester_0.1.7 DelayedMatrixStats_1.20.0
[73] vctrs_0.6.2 ellipsis_0.3.2
[75] generics_0.1.3 tools_4.2.2
[77] foreign_0.8-84 beeswarm_0.4.0
[79] munsell_0.5.0 tweenr_2.0.2
[81] proxy_0.4-27 DelayedArray_0.24.0
[83] fastmap_1.1.1 compiler_4.2.2
[85] abind_1.4-5 httpuv_1.6.10
[87] rtracklayer_1.58.0 plotly_4.10.1
[89] GenomeInfoDbData_1.2.9 gridExtra_2.3
[91] edgeR_3.40.2 lattice_0.21-8
[93] deldir_1.0-9 utf8_1.2.3
[95] later_1.3.1 jsonlite_1.8.4
[97] scales_1.2.1 docopt_0.7.1
[99] gld_2.6.6 ScaledMatrix_1.6.0
[101] pbapply_1.7-0 sparseMatrixStats_1.10.0
[103] lazyeval_0.2.2 promises_1.2.0.1
[105] doParallel_1.0.17 goftest_1.2-3
[107] spatstat.utils_3.0-3 checkmate_2.2.0
[109] rmarkdown_2.21 textshaping_0.3.6
[111] statmod_1.5.0 Rtsne_0.16
[113] uwot_0.1.14 igraph_1.4.3
[115] survival_3.5-5 yaml_2.3.7
[117] systemfonts_1.0.4 htmltools_0.5.5
[119] memoise_2.0.1 BiocIO_1.8.0
[121] locfit_1.5-9.7 graphlayouts_1.0.0
[123] digest_0.6.31 rrcov_1.7-2
[125] assertthat_0.2.1 mime_0.12
[127] RSQLite_2.3.1 future.apply_1.11.0
[129] Exact_3.2 blob_1.2.4
[131] preprocessCore_1.60.2 ragg_1.2.4
[133] splines_4.2.2 Formula_1.2-5
[135] labeling_0.4.2 RCurl_1.98-1.12
[137] hms_1.1.3 colorspace_2.1-0
[139] base64enc_0.1-3 ggbeeswarm_0.7.2
[141] ggrastr_1.0.1 nnet_7.3-19
[143] Rcpp_1.0.10 RANN_2.6.1
[145] mvtnorm_1.1-3 fansi_1.0.4
[147] tzdb_0.3.0 parallelly_1.36.0
[149] R6_2.5.1 grid_4.2.2
[151] lifecycle_1.0.3 rootSolve_1.8.2.3
[153] bluster_1.8.0 curl_4.3.3
[155] leiden_0.4.3 robustbase_0.95-1
[157] RcppAnnoy_0.0.20 org.Hs.eg.db_3.16.0
[159] iterators_1.0.14 spatstat.explore_3.2-1
[161] htmlwidgets_1.6.2 beachmat_2.14.2
[163] polyclip_1.10-4 timechange_0.2.0
[165] mgcv_1.8-42 globals_0.16.2
[167] lmom_2.9 htmlTable_2.4.1
[169] spatstat.random_3.1-5 progressr_0.13.0
[171] codetools_0.2-19 FNN_1.1.3.2
[173] GO.db_3.16.0 metapod_1.6.0
[175] gtable_0.3.3 DBI_1.1.3
[177] tensor_1.5 httr_1.4.6
[179] KernSmooth_2.23-21 stringi_1.7.12
[181] farver_2.1.1 boot_1.3-28.1
[183] BiocNeighbors_1.16.0 restfulr_0.0.15
[185] scattermore_1.0 DEoptimR_1.0-13
[187] bit_4.0.5 spatstat.data_3.0-1
[189] pkgconfig_2.0.3 knitr_1.43

[jmuribes]

Hi!

I actually think I found what the issue is. In the metacells Seurat object (created inside seurat_obj@misc$wgcna$wgcna_metacell_obj), I realized my clustering variable (clusters_hdwgcna) was not there. I manually added it as a column in the metadata using the current active identity of the metacells and now is working fine.

# Add a variable 'cluster_hdwgcna' to the metacells Seurat object:
all.data.combined.neurons.filt.hdwgcna@misc$srgap2$wgcna_metacell_obj$cluster_hdwgcna <- all.data.combined.neurons.filt.hdwgcna@misc$srgap2$wgcna_metacell_obj@active.ident