Since the recent upgrade to 'BiocVersion' 3.16.0, upgrading 'SingleR' to 2.0.0, I was not able to successfully run 'SingleR' with multiple references as I was used to do up to few weeks ago. I went through all the vignettes and manual pages I was able to retrieve on the matter but without solving the problem.

Hereafter I report a slightly modified version of the 'SingleR' example from the Help, by introducing a second reference to mimic the multiple references analysis:

# Mocking up data with log-normalized expression values:
ref1 <- .mockRefData() # 1st reference
ref2 <- .mockRefData() # 2nd reference
test <- .mockTestData(ref1)

ref1 <- scuttle::logNormCounts(ref1)
ref2 <- scuttle::logNormCounts(ref2)
test <- scuttle::logNormCounts(test)

ref <- list(ref1,ref2) # list of multiple references

ref1.labels <- ref1$label # ref1 labels
ref.labels <- list(ref1$label,ref2$label) # multiple references' labels


# Running the classification with different options:
pred.a <- SingleR(test, ref1, labels=ref1.labels) # works OK

pred.b <- SingleR(test, ref, labels=ref.labels) # DOES NOT work, giving the above mentioned error and warning messsages.


Error in results[[i]] : subscript out of bounds
In addition: Warning messages:
1: In combineRecomputedResults(results, test = test, trained = trained,  :
  entries of 'trained' differ in the universe of available markers
2: In combineRecomputedResults(results, test = test, trained = trained,  :
 entries of 'trained' differ in the universe of available markers

By tracing back the error I get:

> traceback()
3: combineRecomputedResults(results, test = test, trained = trained, 
       check.missing = FALSE, quantile = quantile)
2: classifySingleR(test, trained, quantile = quantile, fine.tune = fine.tune, 
       tune.thresh = tune.thresh, prune = prune, check.missing = FALSE, 
       num.threads = num.threads, BPPARAM = BPPARAM)
1: SingleR(test, ref, labels = ref.labels)

After running 'debugonce(SingleR)', within the 'classifySingleR' function chunk, 'results' is empty and as a consequence the final 'combineRecomputedResults' function leads to the error and warning messages (I report only the crucial lines here):

> debugonce(SingleR)
> pred.b <- SingleR(test, ref, labels=ref.labels) # does not work
debugging in: SingleR(test, ref, labels = ref.labels)

...

Browse[2]> s
debugging in: classifySingleR(test, trained, quantile = quantile, fine.tune = fine.tune, 
    tune.thresh = tune.thresh, prune = prune, check.missing = FALSE, 
    num.threads = num.threads, BPPARAM = BPPARAM)

...

Browse[3]> 
debug: results <- lapply(trained, FUN = .classify_internals, test = test, 
    quantile = quantile, fine.tune = fine.tune, tune.thresh = tune.thresh, 
    prune = prune, num.threads = num.threads)

## 'results' is empty here, thus leading to the failure of the next function:

Browse[3]> 
debug: combineRecomputedResults(results, test = test, trained = trained, 
    check.missing = FALSE, quantile = quantile)
Browse[3]> 
Error in results[[i]] : subscript out of bounds
In addition: Warning messages:
1: In combineRecomputedResults(results, test = test, trained = trained,  :
  entries of 'trained' differ in the universe of available markers
2: In combineRecomputedResults(results, test = test, trained = trained,  :
  entries of 'trained' differ in the universe of available markers



> sessionInfo()
R version 4.2.0 (2022-04-22)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.5 LTS

Matrix products: default
BLAS:   /home/carlo/R/R-4.2.0/lib/libRblas.so
LAPACK: /home/carlo/R/R-4.2.0/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=it_IT.UTF-8       
 [4] LC_COLLATE=en_US.UTF-8     LC_MONETARY=it_IT.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=it_IT.UTF-8       LC_NAME=C                  LC_ADDRESS=C              
[10] LC_TELEPHONE=C             LC_MEASUREMENT=it_IT.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
 [1] datasets  stats4    grid      parallel  grDevices compiler  stats     graphics  utils     methods  
[11] base     

other attached packages:
  [1] scAnnotatR_0.99.7           devtools_2.4.5              usethis_2.1.6              
  [4] ggalluvial_0.12.3           CellChat_1.5.0              circlize_0.4.15            
  [7] NMF_0.24.0                  rngtools_1.5.2              pkgmaker_0.32.2            
 [10] registry_0.5-1              parallelly_1.32.1           BiocParallel_1.32.1        
 [13] heatmap3_1.1.9              ComplexHeatmap_2.14.0       superheat_0.1.0            
 [16] MAST_1.24.0                 ggraph_2.1.0                glmGamPoi_1.10.0           
 [19] harmony_0.1.1               Rcpp_1.0.9                  biomaRt_2.54.0             
 [22] patchwork_1.1.2             ggpubr_0.5.0                HGNChelper_0.8.1           
 [25] dbplyr_2.2.1                forcats_0.5.2               stringr_1.4.1              
 [28] dplyr_1.0.10                purrr_0.3.5                 readr_2.1.3                
 [31] tidyr_1.2.1                 tibble_3.1.8                tidyverse_1.3.2            
 [34] mclust_6.0.0                monocle3_1.2.9              ggrastr_1.0.1              
 [37] HDF5Array_1.26.0            rhdf5_2.42.0                Matrix.utils_0.9.8         
 [40] batchelor_1.14.0            lme4_1.1-31                 DelayedMatrixStats_1.20.0  
 [43] DelayedArray_0.24.0         infercnv_1.14.0             leiden_0.4.3               
 [46] scater_1.26.1               scran_1.26.0                scuttle_1.8.0              
 [49] scRNAseq_2.12.0             SingleCellExperiment_1.20.0 celldex_1.8.0              
 [52] SingleR_2.0.0               SummarizedExperiment_1.28.0 GenomicRanges_1.50.1       
 [55] GenomeInfoDb_1.34.3         MatrixGenerics_1.10.0       matrixStats_0.62.0         
 [58] SeuratWrappers_0.3.0        SeuratObject_4.1.3          Seurat_4.2.1               
 [61] edgeR_3.40.0                installr_0.23.4             uwot_0.1.14                
 [64] statmod_1.4.37              AnnotationDbi_1.60.0        IRanges_2.32.0             
 [67] S4Vectors_0.36.0            Polychrome_1.5.1            randomcoloR_1.1.0.1        
 [70] GSA_1.03.2                  tm_0.7-9                    NLP_0.2-1                  
 [73] limma_3.54.0                Rmisc_1.5.1                 plotly_4.10.1              
 [76] Hmisc_4.7-1                 ggplot2_3.4.0               Formula_1.2-4              
 [79] survival_3.4-0              lattice_0.20-45             corrgram_1.14              
 [82] corrplot_0.92               umap_0.2.9.0                clusterProfiler_4.6.0      
 [85] tictoc_1.1                  gridBase_0.4-7              eulerr_6.1.1               
 [88] networkD3_0.4               plotrix_3.8-2               ff_4.0.7                   
 [91] bit_4.0.5                   car_3.1-1                   carData_3.0-5              
 [94] smoothie_1.0-3              vcd_1.4-10                  locfit_1.5-9.6             
 [97] Rtsne_0.16                  tsne_0.1-3.1                sp_1.5-1                   
[100] fields_14.1                 viridis_0.6.2               viridisLite_0.4.1          
[103] spam_2.9-1                  doParallel_1.0.17           iterators_1.0.14           
[106] foreach_1.5.2               beeswarm_0.4.0              tgp_2.4-20                 
[109] beanplot_1.3.1              fpc_2.2-9                   dbscan_1.1-11              
[112] vioplot_0.3.7               zoo_1.8-11                  sm_2.2-5.7.1               
[115] MASS_7.3-58.1               ellipse_0.4.3               cluster_2.1.4              
[118] scales_1.2.1                multtest_2.54.0             Biobase_2.58.0             
[121] BiocGenerics_0.44.0         metap_1.8                   magrittr_2.0.3             
[124] colorspace_2.0-3            igraph_1.3.5                RColorBrewer_1.1-3         
[127] plyr_1.8.8                  Matrix_1.5-3                conflicted_1.1.0           
[130] BiocManager_1.30.19

Is there anything I am missing?

Thanks in advance for any advice, Carlo