Seurat V5 Compatibility Update (Dev Note/Plan)

[samuel-marsh]

Note from Dev:

Thank you everyone for your support and use of scCustomize, I’m really happy so many people find it useful!

As many of you are aware the Satija lab released a major update to Seurat in the form of Seurat V5 (currently in beta).

The current development plan for Seurat V5 is this:

• Push final pre-V5 update May 1st July 1st from current dev branch and open issue fixes.
• Around then I will split dev branch into develop for hot fixes for current release version of Seurat (4.3.0) and a second branch developV5 to start working on updates to ensure compatibility with both pre and post V5 features.

In meantime if you are using Seurat V5 please be aware some functions may not work as intended.

I would also greatly appreciate it if anyone using Seurat V5 who notices issues with scCustomize function/interaction with Seurat could post on this issue with:

• Function(s) causing errors
• Error messages received
• Version number of Seurat V5 beta being used

Thanks so much!! Sam

[samuel-marsh]

Notes to Self:

• [x] Cells for cell names of current layer vs colnames for entire object. (highlight plot by layer?)
• [x] Add layer param inside plotting functions as shortcut for switching default layer before plotting
• [x] slot vs layer
• [x] Swap LayerData for GetAssayData
• [x] Add check for Assay5 vs. Assay?
• [x] Check scattermore compatibility v1.1
• [x] Check accessor functions
• [ ] Add BPCells read functions
• [ ] Add qs based save function for SaveSeuratBP
• [ ] Add qs based read function equivalent to LoadSeuratRds
• [ ] CellBender Diff functions. Make option for joint vs. layer-to-layer comparison.

[samuel-marsh]

Functions using slot param to be altered:

• [x] FetchMeta
• [x] CheckMatrix_scCustom
• [x] FeaturePlot_scCustom (and FetchData within)
• [x] FeaturePlot_DualAssay
• [x] Check pluck compatibility (switched to LayerData for easiest cross-functionality)

[samuel-marsh]

Potential New Functions

• [ ] Ensembl to ID finally? (likely not this release)
• [ ] Density contour/shading on QC FeatureScatter plots (likely not this release)
• [ ] Move Rename 10X (maybe...) (likely not this release)
• [x] Iterate barcode rank plot (and single barcode rank plot)

[samuel-marsh]

Vignette Updates

• [x] Barcode Rank Plots and Iterative Version
• [x] Update vignette package loading to enable removal of tidyverse in suggests per Hadley Wickham email
• [x] MAD Stats function
• [x] Add_Top_Gene_Pct_Seurat
• [x] Add_Cell_QC_Metrics
• [x] Read10X_Metrics multi support including TCR metrics
• [x] VariableFeaturePlot_scCustom supply custom features
• [x] Updates to default color palettes.
• [x] Updates to violin plot (plot_median and plot_boxplot)
• [x] QC_Histogram plotting
• [x] FeatureScatter_scCustom and deprecation of Split_FeatureScatter
• [x] FeaturePlot_DualAssay errors?
• [x] Gene list used from FindMarkers for Clustered_DotPlot

[samuel-marsh]

Fix/Updates:

• [x] Replace old code with isTRUE
• [ ] Chose cell calling option in dual assay object creation (likely not this release)

[pkhoueiry]

Add_Mito_Ribo_Seurat() is throwing the message:

 Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'x' in selecting a method for function 'colSums': subscript out of bounds

[samuel-marsh]

@pkhoueiry Thanks for posting this!

Can you try one thing for me. Do you get any errors if you run this? (change the pattern if not human data)

OBJ[["percent_mito"]] <- PercentageFeatureSet(object = OBJ, pattern = "^MT-")

Can you also send copy of sessionInfo()

Thanks! Sam

[pkhoueiry]

Indeed this works. This is how I overcame the issue.

sessionInfo()

R version 4.3.1 (2023-06-16)
Platform: aarch64-apple-darwin20 (64-bit)
Running under: macOS Ventura 13.5

Matrix products: default
BLAS:   /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib 
LAPACK: /Library/Frameworks/R.framework/Versions/4.3-arm64/Resources/lib/libRlapack.dylib;  LAPACK version 3.11.0

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

time zone: Europe/Berlin
tzcode source: internal

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

other attached packages:
 [1] Seurat.utils_1.9.4      ggExpress_0.7.1         MarkdownHelpers_0.2.2  
 [4] ggpubr_0.6.0            CodeAndRoll2_2.5.2      Stringendo_0.3.5       
 [7] Azimuth_0.4.6           shinyBS_0.61.1          ggplot2_3.4.3          
[10] scCustomize_1.1.3       doParallel_1.0.17       iterators_1.0.14       
[13] foreach_1.5.2           cowplot_1.1.1           HGNChelper_0.8.1       
[16] Seurat_4.9.9.9059       SeuratObject_4.9.9.9091 sp_2.0-0               
[19] tidyr_1.3.0             dplyr_1.1.2            

loaded via a namespace (and not attached):
  [1] RcppAnnoy_0.0.21         splines_4.3.1            later_1.3.1             
  [4] R.oo_1.25.0              tibble_3.2.1             cellranger_1.1.0        
  [7] polyclip_1.10-4          janitor_2.2.0            fastDummies_1.7.3       
 [10] lifecycle_1.0.3          rstatix_0.7.2            vroom_1.6.3             
 [13] globals_0.16.2           lattice_0.21-8           hdf5r_1.3.8             
 [16] MASS_7.3-60              backports_1.4.1          magrittr_2.0.3          
 [19] openxlsx_4.2.5.2         plotly_4.10.2            sm_2.2-5.7.1            
 [22] httpuv_1.6.11            sctransform_0.3.5        zip_2.3.0               
 [25] spam_2.9-1               sessioninfo_1.2.2        spatstat.sparse_3.0-2   
 [28] reticulate_1.31          pbapply_1.7-2            RColorBrewer_1.1-3      
 [31] lubridate_1.9.2          abind_1.4-5              Rtsne_0.16              
 [34] R.utils_2.12.2           purrr_1.0.2              presto_1.0.0            
 [37] rappdirs_0.3.3           circlize_0.4.15          ggrepel_0.9.3           
 [40] irlba_2.3.5.1            listenv_0.9.0            spatstat.utils_3.0-3    
 [43] gdata_2.19.0             pheatmap_1.0.12          goftest_1.2-3           
 [46] RSpectra_0.16-1          spatstat.random_3.1-5    fitdistrplus_1.1-11     
 [49] parallelly_1.36.0        leiden_0.4.3             codetools_0.2-19        
 [52] DT_0.28                  tidyselect_1.2.0         shape_1.4.6             
 [55] SoupX_1.6.2              matrixStats_1.0.0        spatstat.explore_3.2-1  
 [58] ReadWriter_0.2.9         googledrive_2.1.1        jsonlite_1.8.7          
 [61] vioplot_0.4.0            ellipsis_0.3.2           progressr_0.14.0        
 [64] ggridges_0.5.4           survival_3.5-7           tictoc_1.2              
 [67] tools_4.3.1              ica_1.0-3                Rcpp_1.0.11             
 [70] glue_1.6.2               gridExtra_2.3            MatrixGenerics_1.12.3   
 [73] shinydashboard_0.7.2     withr_2.5.0              fastmap_1.1.1           
 [76] fansi_1.0.4              shinyjs_2.1.0            digest_0.6.33           
 [79] MarkdownReports_4.5.9    timechange_0.2.0         R6_2.5.1                
 [82] mime_0.12                ggprism_1.0.4            colorspace_2.1-0        
 [85] scattermore_1.2          gtools_3.9.4             tensor_1.5              
 [88] spatstat.data_3.0-1      R.methodsS3_1.8.2        googlesheets4_1.1.1     
 [91] utf8_1.2.3               generics_0.1.3           data.table_1.14.8       
 [94] httr_1.4.7               htmlwidgets_1.6.2        uwot_0.1.16             
 [97] pkgconfig_2.0.3          gtable_0.3.3             lmtest_0.9-40           
[100] htmltools_0.5.6          carData_3.0-5            dotCall64_1.0-2         
[103] tidyverse_2.0.0          scales_1.2.1             SeuratData_0.2.2        
[106] png_0.1-8                snakecase_0.11.0         rstudioapi_0.15.0       
[109] tzdb_0.4.0               reshape2_1.4.4           nlme_3.1-163            
[112] ggcorrplot_0.1.4         zoo_1.8-12               GlobalOptions_0.1.2     
[115] stringr_1.5.0            KernSmooth_2.23-22       miniUI_0.1.1.1          
[118] vipor_0.4.5              ggrastr_1.0.2            pillar_1.9.0            
[121] grid_4.3.1               vctrs_0.6.3              RANN_2.6.1              
[124] promises_1.2.1           car_3.1-2                princurve_2.1.6         
[127] xtable_1.8-4             cluster_2.1.4            SeuratDisk_0.0.0.9020   
[130] beeswarm_0.4.0           paletteer_1.5.0          readr_2.1.4             
[133] cli_3.6.1                compiler_4.3.1           rlang_1.1.1             
[136] crayon_1.5.2             future.apply_1.11.0      ggsignif_0.6.4          
[139] rematch2_2.1.2           plyr_1.8.8               forcats_1.0.0           
[142] fs_1.6.3                 ggbeeswarm_0.7.2         stringi_1.7.12          
[145] viridisLite_0.4.2        deldir_1.0-9             munsell_0.5.0           
[148] lazyeval_0.2.2           spatstat.geom_3.2-4      Matrix_1.6-1            
[151] RcppHNSW_0.4.1           hms_1.1.3                patchwork_1.1.3         
[154] sparseMatrixStats_1.12.2 bit64_4.0.5              future_1.33.0           
[157] shiny_1.7.5              ROCR_1.0-11              clipr_0.8.0             
[160] gargle_1.5.2             igraph_1.5.1             broom_1.0.5             
[163] bit_4.0.5     

[samuel-marsh]

ok let thanks! one more thing. After the error could you run traceback() and send that output?

Thanks! Sam

[pkhoueiry]

I don't have the error if I specify the DefaultAssay as RNA for all objects in my list.

project <- lapply(project, function(x){
     DefaultAssay(x) <- "RNA"
     x <- Add_Mito_Ribo_Seurat(seurat_object = x, species = "Human")
     x
})

[samuel-marsh]

ok that's interesting. Was the default assay something other than RNA before this step?

The assay specification to PercentageFeatureSet should get set automatically during the function with this: https://github.com/samuel-marsh/scCustomize/blob/2bb2b138edefee5946c523a5d1b2e418974a5d39/R/Object_Utilities.R#L114C1-L115C59

Thanks again really appreciate you help!

[pkhoueiry]

As I am working with Cite-seq, it can be that the assay was set to HTO during the first time I run the counts. Best and thanks for your help too, P.

[samuel-marsh]

Gotcha!! Ok ya that is likely issue. I'll work on adding check for that in function.

Best, Sam

[samuel-marsh]

layers <- Layers(object, search = layers %||% 'data')