Error in `palette()` when both `blend = TRUE, split.by` are used in `FeaturePlot`

[egeulgen]

Hey Seurat team!

Desciption of issue

I'm running FeaturePlot with two features (one from metadata, one expression) using both blend = TRUE, split.by = "<split_var> and I'm getting the below error:

Error in `palette()`:
! Insufficient values in manual scale. 1 needed but only 0 provided.
Run `rlang::last_trace()` to see where the error occurred.

Proposed Solution

When I debugged the error, I found out it happens when one of the features per split provided to BlendExpression is constant, min-max normalization within BlendExpression returns NaN, yielding unusable output data.

I attempted to fix this in #8486 by making the normalization in BlendExpression return 0 if the given feature is constant, which has resolved the issue locally. However, I wanted to describe the bug in more detail, in case there might be a better solution.

Reproducible Example

Here's a reproducible example using pbmc_small:

library(Seurat)

set.seed(123)

test_metadata <- data.frame(
    test_var1 = runif(ncol(pbmc_small)), 
    test_var2 = ifelse(pbmc_small$groups == "g1", 0.5, runif(1)), 
    row.names = colnames(pbmc_small)
)

pbmc_small <- AddMetaData(pbmc_small, test_metadata)
FeaturePlot(pbmc_small, features = c("test_var1", "CD79B"), blend = TRUE, split.by = "groups")
# works OK and produces the plot

FeaturePlot(pbmc_small, features = c("test_var2", "CD79B"), blend = TRUE, split.by = "groups")
# raises:
# Error in `palette()`:
#     ! Insufficient values in manual scale. 1 needed but only 0 provided.
# Run `rlang::last_trace()` to see where the error occurred.

sessionInfo:

R version 4.3.2 (2023-10-31)
Platform: x86_64-apple-darwin20 (64-bit)
Running under: macOS Sonoma 14.1.1

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-x86_64/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/London
tzcode source: internal

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

other attached packages:
[1] Seurat_5.0.1       SeuratObject_5.0.1 sp_2.1-2          

loaded via a namespace (and not attached):
  [1] deldir_2.0-2           pbapply_1.7-2          gridExtra_2.3          rlang_1.1.2            magrittr_2.0.3         RcppAnnoy_0.0.21      
  [7] spatstat.geom_3.2-7    matrixStats_1.1.0      ggridges_0.5.4         compiler_4.3.2         png_0.1-8              vctrs_0.6.5           
 [13] reshape2_1.4.4         stringr_1.5.1          pkgconfig_2.0.3        fastmap_1.1.1          ellipsis_0.3.2         labeling_0.4.3        
 [19] utf8_1.2.4             promises_1.2.1         purrr_1.0.2            jsonlite_1.8.8         goftest_1.2-3          later_1.3.2           
 [25] spatstat.utils_3.0-4   irlba_2.3.5.1          parallel_4.3.2         cluster_2.1.4          R6_2.5.1               ica_1.0-3             
 [31] stringi_1.8.2          RColorBrewer_1.1-3     spatstat.data_3.0-3    reticulate_1.34.0      parallelly_1.36.0      lmtest_0.9-40         
 [37] scattermore_1.2        Rcpp_1.0.11            tensor_1.5             future.apply_1.11.0    zoo_1.8-12             sctransform_0.4.1     
 [43] httpuv_1.6.13          Matrix_1.6-3           splines_4.3.2          igraph_1.5.1           tidyselect_1.2.0       abind_1.4-5           
 [49] rstudioapi_0.15.0      spatstat.random_3.2-2  codetools_0.2-19       miniUI_0.1.1.1         spatstat.explore_3.2-5 listenv_0.9.0         
 [55] lattice_0.21-9         tibble_3.2.1           plyr_1.8.9             withr_2.5.2            shiny_1.8.0            ROCR_1.0-11           
 [61] Rtsne_0.17             future_1.33.0          fastDummies_1.7.3      survival_3.5-7         polyclip_1.10-6        fitdistrplus_1.1-11   
 [67] pillar_1.9.0           BiocManager_1.30.22    KernSmooth_2.23-22     renv_1.0.3             plotly_4.10.3          generics_0.1.3        
 [73] RcppHNSW_0.5.0         ggplot2_3.4.4          munsell_0.5.0          scales_1.3.0           globals_0.16.2         xtable_1.8-4          
 [79] glue_1.6.2             lazyeval_0.2.2         tools_4.3.2            data.table_1.14.10     RSpectra_0.16-1        RANN_2.6.1            
 [85] leiden_0.4.3.1         dotCall64_1.1-1        cowplot_1.1.1          grid_4.3.2             tidyr_1.3.0            colorspace_2.1-0      
 [91] nlme_3.1-163           patchwork_1.1.3        cli_3.6.1              spatstat.sparse_3.0-3  spam_2.10-0            fansi_1.0.6           
 [97] viridisLite_0.4.2      dplyr_1.1.4            uwot_0.1.16            gtable_0.3.4           digest_0.6.33          progressr_0.14.0      
[103] ggrepel_0.9.4          farver_2.1.1           htmlwidgets_1.6.4      htmltools_0.5.7        lifecycle_1.0.4        httr_1.4.7            
[109] mime_0.12              MASS_7.3-60 

[zskylarli]

Thanks for reporting this! Glad you identified the issue and the fix is working - we're a little out of bandwidth at the moment to develop a more robust fix, but if you encounter more plotting issues in the future, please continue to let us know!

[DANIBOMON]

Data_Fenotipico_Ecoli <- as.character(Data_Fenotipico_Ecoli, stringsAsFactors = FALSE)
p <- strsplit(Data_Fenotipico_Ecoli, "", fixed = TRUE)

df <- do.call(rbind, lapply(p, (x) c(x, rep(NA, max(lengths(p)) - length(x))))) %>%
as.data.frame() %>%
cbind.data.frame(id = Data_Fenotipico_Ecoli) %>%
mutate(Id = factor(id, rev(id))) %>%
pivot_longer(V1:V64)

ggplot(df, aes(name, Id, fill = value)) +
geom_tile(col = "white", size = 2) +
coord_equal() +
scale_fill_manual(values = c("#8B2323", "#8B7355"),
breaks = c("R", "S"),
na.value = "#8B814C") +
scale_x_discrete(name = "Antibiotic",
position = "top",
breaks = 1:64) +
labs(fill = "Resistance",
y = "ID",
title = "Resistencia a antibióticos") +
theme_minimal(base_size = 64) +
theme(text = element_text(color = "#838B83"))