Dear List

I obtained a non-monotonic Voom plot with voomByGroup on Xenium data. Does this compromise the reliability of the results, and if so is there anything I can do about it? Here's the plot

Here is my code up until an including the plot and sessionInfo.

library(Seurat)
library(limma)
library(edgeR)

system("mkdir results/")
system("mkdir results/global/")

source("voomByGroup.R")

targets<-readTargets("data/targets.txt")

sprsmat.tmp=Read10X_h5(paste0("data/",targets[1,1]))

counts= rowSums(sprsmat.tmp)

for (i in 2:nrow(targets)){
    region.tmp.sprsmat=Read10X_h5(paste0("data/",targets[i,1]))
    counts.tmp=rowSums(region.tmp.sprsmat)
    counts=cbind(counts,counts.tmp)
    }

colnames(counts)<-targets$Name
flevels<-unique(targets$Target)
f<-factor(targets$Target,levels=flevels)
y<-DGEList(counts=counts, genes=rownames(counts))
isexpr<-rowSums(y$counts>=10) >= 3
hasannot<-rowSums(is.na(y$genes))==0
y<-y[isexpr & hasannot,]
y$samples$lib.size<- colSums(y$counts)
y$samples$lib.size
y<-calcNormFactors(y)
des<-model.matrix(~0+f)
colnames(des)<-flevels
png("results/global/voomByGrouplot.png")
v=voomByGroup(y, design=des, group=targets$Target,plot="combine")
dev.off()



> sessionInfo()
R version 4.5.1 (2025-06-13)
Platform: x86_64-apple-darwin20
Running under: macOS Sonoma 14.6.1

Matrix products: default
BLAS:   /Library/Frameworks/R.framework/Versions/4.5-x86_64/Resources/lib/libRblas.0.dylib 
LAPACK: /Library/Frameworks/R.framework/Versions/4.5-x86_64/Resources/lib/libRlapack.dylib;  LAPACK version 3.12.1

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: America/New_York
tzcode source: internal

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

other attached packages:
[1] edgeR_4.6.3        limma_3.64.1       Seurat_5.3.0       SeuratObject_5.1.0
[5] sp_2.2-0          

loaded via a namespace (and not attached):
  [1] deldir_2.0-4           pbapply_1.7-2          gridExtra_2.3         
  [4] rlang_1.1.6            magrittr_2.0.3         RcppAnnoy_0.0.22      
  [7] matrixStats_1.5.0      ggridges_0.5.6         compiler_4.5.1        
 [10] spatstat.geom_3.4-1    png_0.1-8              vctrs_0.6.5           
 [13] reshape2_1.4.4         hdf5r_1.3.12           stringr_1.5.1         
 [16] pkgconfig_2.0.3        fastmap_1.2.0          promises_1.3.3        
 [19] bit_4.6.0              purrr_1.1.0            jsonlite_2.0.0        
 [22] goftest_1.2-3          later_1.4.2            spatstat.utils_3.1-4  
 [25] irlba_2.3.5.1          parallel_4.5.1         cluster_2.1.8.1       
 [28] R6_2.6.1               ica_1.0-3              stringi_1.8.7         
 [31] RColorBrewer_1.1-3     spatstat.data_3.1-6    reticulate_1.42.0     
 [34] parallelly_1.45.0      spatstat.univar_3.1-4  lmtest_0.9-40         
 [37] scattermore_1.2        Rcpp_1.1.0             tensor_1.5.1          
 [40] future.apply_1.20.0    zoo_1.8-14             sctransform_0.4.2     
 [43] httpuv_1.6.16          Matrix_1.7-3           splines_4.5.1         
 [46] igraph_2.1.4           tidyselect_1.2.1       dichromat_2.0-0.1     
 [49] abind_1.4-8            codetools_0.2-20       spatstat.random_3.4-1 
 [52] miniUI_0.1.2           spatstat.explore_3.4-3 listenv_0.9.1         
 [55] lattice_0.22-7         tibble_3.3.0           plyr_1.8.9            
 [58] shiny_1.11.1           ROCR_1.0-11            Rtsne_0.17            
 [61] future_1.58.0          fastDummies_1.7.5      survival_3.8-3        
 [64] polyclip_1.10-7        fitdistrplus_1.2-4     pillar_1.11.0         
 [67] KernSmooth_2.23-26     plotly_4.11.0          generics_0.1.4        
 [70] RcppHNSW_0.6.0         ggplot2_3.5.2          scales_1.4.0          
 [73] globals_0.18.0         xtable_1.8-4           glue_1.8.0            
 [76] lazyeval_0.2.2         tools_4.5.1            data.table_1.17.8     
 [79] RSpectra_0.16-2        locfit_1.5-9.12        RANN_2.6.2            
 [82] dotCall64_1.2          cowplot_1.2.0          grid_4.5.1            
 [85] tidyr_1.3.1            colorspace_2.1-1       nlme_3.1-168          
 [88] patchwork_1.3.1        cli_3.6.5              spatstat.sparse_3.1-0 
 [91] spam_2.11-1            viridisLite_0.4.2      dplyr_1.1.4           
 [94] uwot_0.2.3             gtable_0.3.6           digest_0.6.37         
 [97] progressr_0.15.1       ggrepel_0.9.6          htmlwidgets_1.6.4     
[100] farver_2.1.2           htmltools_0.5.8.1      lifecycle_1.0.4       
[103] httr_1.4.7             statmod_1.5.0          mime_0.13             
[106] bit64_4.6.0-1          MASS_7.3-65           
>

I would appreciate any guidance you might be able to offer.

Thanks and best wishes,

Rich

Richard Friedman

Herbert Irving Comprehensive Cancer Center

Columbia University Irving Medical Center

It doesn't look like a problem to me, because the increase in the trend at low expression levels is very slight. If you don't like it though, increase the span parameter.