hdWGCNA
hdWGCNA copied to clipboard
smorabit
MetaspotsByGroups : Error in `as.Graph()`
Hi Sam, i get an error while runnning MetaspotsByGroups. It used to work fine but I have recently been getting this error
Error in as.Graph():
! Please provide rownames to the matrix before converting to a Graph
Traceback:
- MetaspotsByGroups(K12B_wgcna, group.by = c("path_anno"), ident.group = "path_anno", . assay = "Spatial", slot = "counts")
- lapply(groupings, function(x) { . seurat_obj[, seurat_obj$metacell_grouping == x] . })
- FUN(X[[i]], ...)
- seurat_obj[, seurat_obj$metacell_grouping == x]
[.Seurat(seurat_obj, , seurat_obj$metacell_grouping == x)- subset.Seurat(x = x, features = i, cells = j, ...)
- suppressWarnings(expr = x[[g]] <- as.Graph(x = x[[g]][cells.g, . cells.g, drop = FALSE]))
- withCallingHandlers(expr, warning = function(w) if (inherits(w, . classes)) tryInvokeRestart("muffleWarning"))
- as.Graph(x = x[[g]][cells.g, cells.g, drop = FALSE])
- as.Graph.Matrix(x = x[[g]][cells.g, cells.g, drop = FALSE])
- abort(message = "Please provide rownames to the matrix before converting to a Graph")
- signal_abort(cnd, .file)
Hi,
Can you please provide the code that you used which produced this error message? Any other relevant information would also help. It would also help me to know if you are able to reproduce this in the same dataset that was used in the tutorial?
Hi @smorabit
I had the same issue appear in my data. I did run the tutorial dataset and did not get this issue, and I did run my data previously and did not encounter this issue, so I'm still tracking down what's different. I'll share my code in case you have any insight, and hopefully one of us can record a solution here.
Upstream processing
library(tidyverse)
library(here)
# library(wgcna)
library(hdWGCNA)
# library(BPCells)
library(Seurat)
LoadSeuratRds <- SeuratObject::LoadSeuratRds
set.seed(44)
options(future.globals.maxSize= 891289600) ## 850*1024^2 = 891289600
data <- SeuratObject::LoadSeuratRds(here('saved_rds/01-obj_generation/joined_obj.Rds'))
data <- SketchData(data,
assay = 'RNA',
ncells=10000)
data
DefaultAssay(data) <- "sketch"
data <- FindVariableFeatures(data)
data <- ScaleData(data)
data <- RunPCA(data)
data <- FindNeighbors(data, dims = 1:50)
# obj <- FindClusters(obj, resolution = 2)
data <- ProjectData(
object = data,
assay = "RNA",
full.reduction = "pca.full",
sketched.assay = "sketch",
sketched.reduction = "pca",
#umap.model = "umap",
dims = 1:30#,
#refdata = list(cluster_full = "seurat_clusters")
)
# now that we have projected the full dataset, switch back to analyzing all cells
DefaultAssay(data) <- "RNA"
data <- harmony::RunHarmony(data,
group.by.vars = "donor_label",
reduction = "pca.full",
reduction.save = 'harmony')
## Finding features by fraction isn't working, so using variable features, but increasing the number to have more data for wgcna
tmp <- rowSums(data@assays$RNA@layers$counts>0)
tmp1 <- tmp / 2036755 ## cell count
tmp2 <- tmp1 > 0.05
feats <- rownames(data)[tmp2]
data <- SetupForWGCNA(
data,
gene_select = "custom", # the gene selection approach
# fraction = 0.1, # fraction of cells that a gene needs to be expressed in order to be included
# group.by = 'class',
features = feats,
wgcna_name = "class" # the name of the hdWGCNA experiment
)
MetacellsByGroups call
data <- MetacellsByGroups(
seurat_obj = data,
group.by = c("class", "donor_label"), # specify the columns in [email protected] to group by
reduction = 'harmony', # select the dimensionality reduction to perform KNN on
dims=1:30,
k = 25, # nearest-neighbors parameter
max_shared = 10, # maximum number of shared cells between two metacells
ident.group = 'class' # set the Idents of the metacell seurat object
)
Error
Error in `as.Graph()`:
! Please provide rownames to the matrix before converting to a Graph
---
Backtrace:
▆
1. └─hdWGCNA::MetacellsByGroups(...)
2. └─base::lapply(...) at smorabit-hdWGCNA-8280ba5/R/metacells.R:394:2
3. └─hdWGCNA (local) FUN(X[[i]], ...)
4. ├─seurat_obj[, seurat_obj$metacell_grouping == x] at smorabit-hdWGCNA-8280ba5/R/metacells.R:394:47
5. └─SeuratObject:::`[.Seurat`(...) at smorabit-hdWGCNA-8280ba5/R/metacells.R:394:47
6. └─SeuratObject:::subset.Seurat(...)
7. ├─base::suppressWarnings(...)
8. │ └─base::withCallingHandlers(...)
9. ├─SeuratObject::as.Graph(x = x[[g]][cells.g, cells.g, drop = FALSE])
10. └─SeuratObject:::as.Graph.Matrix(x = x[[g]][cells.g, cells.g, drop = FALSE])
11. └─rlang::abort(message = "Please provide rownames to the matrix before converting to a Graph")
SessionInfo
R version 4.3.1 (2023-06-16)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 22.04.3 LTS
Matrix products: default
BLAS: /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so; LAPACK version 3.10.0
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=en_US.UTF-8
[4] LC_COLLATE=en_US.UTF-8 LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=en_US.UTF-8 LC_NAME=C LC_ADDRESS=C
[10] LC_TELEPHONE=C LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
time zone: Etc/UTC
tzcode source: system (glibc)
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] BPCells_0.1.0 shinyBS_0.61.1 hdWGCNA_0.3.01 igraph_1.5.1
[5] WGCNA_1.72-1 fastcluster_1.2.3 dynamicTreeCut_1.63-1 ggrepel_0.9.4
[9] Seurat_5.1.0 SeuratObject_5.0.2 sp_2.1-1 harmony_1.1.0
[13] Rcpp_1.0.11 here_1.0.1 lubridate_1.9.3 forcats_1.0.0
[17] stringr_1.5.0 dplyr_1.1.3 purrr_1.0.2 readr_2.1.4
[21] tidyr_1.3.0 tibble_3.2.1 ggplot2_3.4.4 tidyverse_2.0.0
loaded via a namespace (and not attached):
[1] IRanges_2.36.0 R.methodsS3_1.8.2
[3] progress_1.2.2 nnet_7.3-19
[5] poweRlaw_0.70.6 goftest_1.2-3
[7] DT_0.30 Biostrings_2.70.3
[9] vctrs_0.6.4 spatstat.random_3.2-1
[11] digest_0.6.33 png_0.1-8
[13] proxy_0.4-27 deldir_1.0-9
[15] parallelly_1.36.0 renv_1.0.3
[17] MASS_7.3-60 Signac_1.13.0
[19] reshape2_1.4.4 httpuv_1.6.12
[21] foreach_1.5.2 BiocGenerics_0.48.1
[23] withr_2.5.1 xfun_0.40
[25] ellipsis_0.3.2 survival_3.5-5
[27] EnsDb.Hsapiens.v86_2.99.0 memoise_2.0.1
[29] zoo_1.8-12 gtools_3.9.4
[31] pbapply_1.7-2 R.oo_1.25.0
[33] Formula_1.2-5 prettyunits_1.2.0
[35] KEGGREST_1.42.0 promises_1.2.1
[37] httr_1.4.7 restfulr_0.0.15
[39] globals_0.16.2 fitdistrplus_1.1-11
[41] rhdf5filters_1.14.1 rhdf5_2.46.1
[43] rstudioapi_0.15.0 miniUI_0.1.1.1
[45] generics_0.1.3 base64enc_0.1-3
[47] curl_5.1.0 S4Vectors_0.40.2
[49] zlibbioc_1.48.2 polyclip_1.10-6
[51] GenomeInfoDbData_1.2.11 SparseArray_1.2.4
[53] xtable_1.8-4 pracma_2.4.2
[55] doParallel_1.0.17 evaluate_0.22
[57] S4Arrays_1.2.1 BiocFileCache_2.10.2
[59] preprocessCore_1.64.0 hms_1.1.3
[61] GenomicRanges_1.54.1 irlba_2.3.5.1
[63] colorspace_2.1-0 filelock_1.0.2
[65] hdf5r_1.3.8 ROCR_1.0-11
[67] reticulate_1.34.0 spatstat.data_3.0-3
[69] magrittr_2.0.3 lmtest_0.9-40
[71] later_1.3.1 lattice_0.21-8
[73] spatstat.geom_3.2-7 future.apply_1.11.0
[75] scattermore_1.2 XML_3.99-0.14
[77] cowplot_1.1.1 matrixStats_1.0.0
[79] RcppAnnoy_0.0.21 Hmisc_5.1-1
[81] pillar_1.9.0 nlme_3.1-162
[83] iterators_1.0.14 caTools_1.18.2
[85] compiler_4.3.1 RSpectra_0.16-1
[87] stringi_1.7.12 tensor_1.5
[89] SummarizedExperiment_1.32.0 GenomicAlignments_1.38.2
[91] plyr_1.8.9 crayon_1.5.2
[93] abind_1.4-5 BiocIO_1.12.0
[95] googledrive_2.1.1 bit_4.0.5
[97] fastmatch_1.1-4 codetools_0.2-19
[99] SeuratData_0.2.2.9001 plotly_4.10.3
[101] mime_0.12 splines_4.3.1
[103] fastDummies_1.7.3 dbplyr_2.4.0
[105] cellranger_1.1.0 knitr_1.44
[107] blob_1.2.4 utf8_1.2.4
[109] seqLogo_1.68.0 AnnotationFilter_1.26.0
[111] fs_1.6.3 listenv_0.9.0
[113] checkmate_2.3.0 Matrix_1.6-4
[115] tzdb_0.4.0 pkgconfig_2.0.3
[117] tools_4.3.1 cachem_1.0.8
[119] RhpcBLASctl_0.23-42 RSQLite_2.3.2
[121] viridisLite_0.4.2 DBI_1.1.3
[123] impute_1.76.0 fastmap_1.1.1
[125] rmarkdown_2.25 scales_1.2.1
[127] grid_4.3.1 ica_1.0-3
[129] shinydashboard_0.7.2 Rsamtools_2.18.0
[131] patchwork_1.1.3 dotCall64_1.1-0
[133] RANN_2.6.1 rpart_4.1.19
[135] farver_2.1.1 yaml_2.3.7
[137] MatrixGenerics_1.14.0 foreign_0.8-84
[139] rtracklayer_1.62.0 cli_3.6.1
[141] stats4_4.3.1 tester_0.1.7
[143] leiden_0.4.3 lifecycle_1.0.3
[145] uwot_0.1.16 Biobase_2.62.0
[147] presto_1.0.0 backports_1.4.1
[149] BSgenome.Hsapiens.UCSC.hg38_1.4.5 BiocParallel_1.36.0
[151] annotate_1.80.0 timechange_0.2.0
[153] gtable_0.3.4 rjson_0.2.21
[155] ggridges_0.5.4 progressr_0.14.0
[157] parallel_4.3.1 jsonlite_1.8.7
[159] RcppHNSW_0.6.0 TFBSTools_1.40.0
[161] bitops_1.0-7 bit64_4.0.5
[163] Rtsne_0.16 spatstat.utils_3.0-4
[165] CNEr_1.38.0 shinyjs_2.1.0
[167] SeuratDisk_0.0.0.9021 R.utils_2.12.2
[169] lazyeval_0.2.2 shiny_1.7.5.1
[171] Azimuth_0.5.0 htmltools_0.5.6.1
[173] GO.db_3.18.0 sctransform_0.4.1
[175] rappdirs_0.3.3 ensembldb_2.26.0
[177] glue_1.6.2 TFMPvalue_0.0.9
[179] spam_2.10-0 googlesheets4_1.1.1
[181] XVector_0.42.0 RCurl_1.98-1.12
[183] rprojroot_2.0.3 BSgenome_1.70.2
[185] gridExtra_2.3 JASPAR2020_0.99.10
[187] R6_2.5.1 labeling_0.4.3
[189] RcppRoll_0.3.0 GenomicFeatures_1.54.4
[191] cluster_2.1.4 Rhdf5lib_1.24.2
[193] gargle_1.5.2 GenomeInfoDb_1.38.8
[195] DirichletMultinomial_1.44.0 DelayedArray_0.28.0
[197] tidyselect_1.2.0 ProtGenerics_1.34.0
[199] htmlTable_2.4.2 xml2_1.3.5
[201] AnnotationDbi_1.64.1 future_1.33.0
[203] munsell_0.5.0 KernSmooth_2.23-21
[205] data.table_1.14.8 htmlwidgets_1.6.2
[207] RColorBrewer_1.1-3 biomaRt_2.58.2
[209] rlang_1.1.1 spatstat.sparse_3.0-3
[211] spatstat.explore_3.2-5 fansi_1.0.5
From the backtrace it looks like there is a problem with the way that it is trying to subset your data. Unfortunately it is very difficult for me to resolve the issue unless I can reproduce it on my side.
I was following this vignette, so I had the SetupForWGCNA call before the MetacellsByGroups call, but the vignette says
Notably, since we consider hdWGCNA to be a downstream data analysis step, we do not support subsetting the Seurat object after SetupForWGCNA has been run.
If I load the object without running SetupForWGCNA, then try to subset it on some of the values that were causing issues previously, it runs fine. Is the workflow outlined in that vignette still accurate? Should I be running SetupForWGCNA prior to creating metacells?
I see it needs the WGCNAname set up, so I can't switch the order. I'm running it now where I define the metadata combination prior to the SetupForWGCNA call, so maybe establishing the metadata prior to the setup will help subsetting