Introduction to scAnnotatR
Introduction
scAnnotatR is an R package for cell type prediction on single cell RNA-sequencing data. Currently, this package supports data in the forms of a Seurat object or a SingleCellExperiment object.
More information about Seurat object can be found here: https://satijalab.org/seurat/ More information about SingleCellExperiment object can be found here: https://osca.bioconductor.org/
scAnnotatR provides 2 main features:
- A set of pretrained and robust classifiers for basic immune cells. See the section below.
- A user-friendly and fully customizable framework to train new classification models. These models can then be easily saved and reused in the future. Details usage of this framework is explained in vignettes 2 and 3.
Installation
The scAnnotatR package can be directly installed from
Bioconductor:
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
if (!require(scAnnotatR))
BiocManager::install("scAnnotatR")For more information, see https://bioconductor.org/install/.
Included models
The scAnnotatR package comes with several pre-trained
models to classify cell types.
# load scAnnotatR into working space
library(scAnnotatR)
#> Loading required package: Seurat
#> Loading required package: SeuratObject
#> Loading required package: sp
#> 'SeuratObject' was built under R 4.4.0 but the current version is
#> 4.4.1; it is recomended that you reinstall 'SeuratObject' as the ABI
#> for R may have changed
#>
#> Attaching package: 'SeuratObject'
#> The following objects are masked from 'package:base':
#>
#> intersect, t
#> Loading required package: SingleCellExperiment
#> Loading required package: SummarizedExperiment
#> Loading required package: MatrixGenerics
#> Loading required package: matrixStats
#>
#> Attaching package: 'MatrixGenerics'
#> The following objects are masked from 'package:matrixStats':
#>
#> colAlls, colAnyNAs, colAnys, colAvgsPerRowSet, colCollapse,
#> colCounts, colCummaxs, colCummins, colCumprods, colCumsums,
#> colDiffs, colIQRDiffs, colIQRs, colLogSumExps, colMadDiffs,
#> colMads, colMaxs, colMeans2, colMedians, colMins, colOrderStats,
#> colProds, colQuantiles, colRanges, colRanks, colSdDiffs, colSds,
#> colSums2, colTabulates, colVarDiffs, colVars, colWeightedMads,
#> colWeightedMeans, colWeightedMedians, colWeightedSds,
#> colWeightedVars, rowAlls, rowAnyNAs, rowAnys, rowAvgsPerColSet,
#> rowCollapse, rowCounts, rowCummaxs, rowCummins, rowCumprods,
#> rowCumsums, rowDiffs, rowIQRDiffs, rowIQRs, rowLogSumExps,
#> rowMadDiffs, rowMads, rowMaxs, rowMeans2, rowMedians, rowMins,
#> rowOrderStats, rowProds, rowQuantiles, rowRanges, rowRanks,
#> rowSdDiffs, rowSds, rowSums2, rowTabulates, rowVarDiffs, rowVars,
#> rowWeightedMads, rowWeightedMeans, rowWeightedMedians,
#> rowWeightedSds, rowWeightedVars
#> Loading required package: GenomicRanges
#> Loading required package: stats4
#> Loading required package: BiocGenerics
#>
#> Attaching package: 'BiocGenerics'
#> The following object is masked from 'package:SeuratObject':
#>
#> intersect
#> The following objects are masked from 'package:stats':
#>
#> IQR, mad, sd, var, xtabs
#> The following objects are masked from 'package:base':
#>
#> Filter, Find, Map, Position, Reduce, anyDuplicated, aperm, append,
#> as.data.frame, basename, cbind, colnames, dirname, do.call,
#> duplicated, eval, evalq, get, grep, grepl, intersect, is.unsorted,
#> lapply, mapply, match, mget, order, paste, pmax, pmax.int, pmin,
#> pmin.int, rank, rbind, rownames, sapply, setdiff, table, tapply,
#> union, unique, unsplit, which.max, which.min
#> Loading required package: S4Vectors
#>
#> Attaching package: 'S4Vectors'
#> The following object is masked from 'package:utils':
#>
#> findMatches
#> The following objects are masked from 'package:base':
#>
#> I, expand.grid, unname
#> Loading required package: IRanges
#>
#> Attaching package: 'IRanges'
#> The following object is masked from 'package:sp':
#>
#> %over%
#> Loading required package: GenomeInfoDb
#> Loading required package: Biobase
#> Welcome to Bioconductor
#>
#> Vignettes contain introductory material; view with
#> 'browseVignettes()'. To cite Bioconductor, see
#> 'citation("Biobase")', and for packages 'citation("pkgname")'.
#>
#> Attaching package: 'Biobase'
#> The following object is masked from 'package:MatrixGenerics':
#>
#> rowMedians
#> The following objects are masked from 'package:matrixStats':
#>
#> anyMissing, rowMedians
#>
#> Attaching package: 'SummarizedExperiment'
#> The following object is masked from 'package:Seurat':
#>
#> Assays
#> The following object is masked from 'package:SeuratObject':
#>
#> Assays
#> Warning: replacing previous import 'ape::where' by 'dplyr::where' when loading
#> 'scAnnotatR'The models are stored in the default_models object:
default_models <- load_models("default")
#> downloading 1 resources
#> retrieving 1 resource
#> loading from cache
names(default_models)
#> [1] "B cells" "Plasma cells" "NK"
#> [4] "CD16 NK" "CD56 NK" "T cells"
#> [7] "CD4 T cells" "CD8 T cells" "Treg"
#> [10] "NKT" "ILC" "Monocytes"
#> [13] "CD14 Mono" "CD16 Mono" "DC"
#> [16] "pDC" "Endothelial cells" "LEC"
#> [19] "VEC" "Platelets" "RBC"
#> [22] "Melanocyte" "Schwann cells" "Pericytes"
#> [25] "Mast cells" "Keratinocytes" "alpha"
#> [28] "beta" "delta" "gamma"
#> [31] "acinar" "ductal" "Fibroblasts"The default_models object is named a list of
classifiers. Each classifier is an instance of the
scAnnotatR S4 class. For example:
default_models[['B cells']]
#> An object of class scAnnotatR for B cells
#> * 31 marker genes applied: CD38, CD79B, CD74, CD84, RASGRP2, TCF3, SP140, MEF2C, DERL3, CD37, CD79A, POU2AF1, MVK, CD83, BACH2, LY86, CD86, SDC1, CR2, LRMP, VPREB3, IL2RA, BLK, IRF8, FLI1, MS4A1, CD14, MZB1, PTEN, CD19, MME
#> * Predicting probability threshold: 0.5
#> * No parent modelBasic pipeline to identify cell types in a scRNA-seq dataset using scAnnotatR
Preparing the data
To identify cell types available in a dataset, we need to load the dataset as Seurat or SingleCellExperiment object.
For this vignette, we use a small sample datasets that is available
as a Seurat object as part of the package.
# load the example dataset
data("tirosh_mel80_example")
tirosh_mel80_example
#> An object of class Seurat
#> 91 features across 480 samples within 1 assay
#> Active assay: RNA (91 features, 0 variable features)
#> 2 layers present: counts, data
#> 1 dimensional reduction calculated: umapThe example dataset already contains the clustering results as part of the metadata. This is not necessary for the classification process.
head(tirosh_mel80_example[[]])
#> orig.ident nCount_RNA nFeature_RNA percent.mt
#> Cy80_II_CD45_B07_S883_comb SeuratProject 42.46011 8 0
#> Cy80_II_CD45_C09_S897_comb SeuratProject 74.35907 14 0
#> Cy80_II_CD45_H07_S955_comb SeuratProject 42.45392 8 0
#> Cy80_II_CD45_H09_S957_comb SeuratProject 63.47043 12 0
#> Cy80_II_CD45_B11_S887_comb SeuratProject 47.26798 9 0
#> Cy80_II_CD45_D11_S911_comb SeuratProject 69.12167 13 0
#> RNA_snn_res.0.8 seurat_clusters RNA_snn_res.0.5
#> Cy80_II_CD45_B07_S883_comb 4 4 2
#> Cy80_II_CD45_C09_S897_comb 4 4 2
#> Cy80_II_CD45_H07_S955_comb 4 4 2
#> Cy80_II_CD45_H09_S957_comb 4 4 2
#> Cy80_II_CD45_B11_S887_comb 4 4 2
#> Cy80_II_CD45_D11_S911_comb 1 1 1Cell classification
To launch cell type identification, we simply call the
classify_cells function. A detailed description of all
parameters can be found through the function’s help page
?classify_cells.
Here we use only 3 classifiers for B cells, T cells and NK cells to
reduce computational cost of this vignette. If users want to use all
pretrained classifiers on their dataset, cell_types = 'all'
can be used.
seurat.obj <- classify_cells(classify_obj = tirosh_mel80_example,
assay = 'RNA', slot = 'counts',
cell_types = c('B cells', 'NK', 'T cells'),
path_to_models = 'default')
#> loading from cacheParameters
- The option cell_types = ‘all’ tells the function to
use all available cell classification models. Alternatively, we can
limit the identifiable cell types:
- by specifying:
cell_types = c('B cells', 'T cells') - or by indicating the applicable classifier using the
classifiers option:
classifiers = c(default_models[['B cells']], default_models[['T cells']])
- by specifying:
- The option path_to_models = ‘default’ is to automatically use the package-integrated pretrained models (without loading the models into the current working space). This option can be used to load a local database instead. For more details see the vignettes on training your own classifiers.
Result interpretation
The classify_cells function returns the input object but
with additional columns in the metadata table.
# display the additional metadata fields
seurat.obj[[]][c(50:60), c(8:ncol(seurat.obj[[]]))]
#> B_cells_p B_cells_class NK_p
#> cy80.Cd45.pos.PD1.pos.B09.S45.comb 0.007754246 no 0.4881285
#> cy80.Cd45.pos.Pd1.neg.S366.H06.S366.comb 0.999385770 yes 0.4440553
#> cy80.Cd45.pos.Pd1.neg.S202.A10.S202.comb 0.998317662 yes 0.4416114
#> cy80.Cd45.pos.Pd1.neg.S201.A09.S201.comb 0.997774856 yes 0.4398997
#> cy80.Cd45.pos.Pd1.neg.S221.B05.S221.comb 0.998874031 yes 0.4541005
#> cy80.Cd45.pos.PD1.pos.A03.S15.comb 0.999944282 yes 0.4511450
#> cy80.Cd45.pos.PD1.pos.B11.S47.comb 0.015978230 no 0.4841041
#> cy80.Cd45.pos.PD1.pos.S189.H09.S189.comb 0.099311534 no 0.4858084
#> cy80.Cd45.pos.PD1.pos.A05.S17.comb 0.055754074 no 0.4924746
#> cy80.Cd45.pos.PD1.pos.C02.S62.comb 0.048558881 no 0.5002238
#> cy80.Cd45.pos.PD1.pos.D12.S96.comb 0.996979702 yes 0.4994867
#> NK_class T_cells_p T_cells_class
#> cy80.Cd45.pos.PD1.pos.B09.S45.comb no 0.94205232 yes
#> cy80.Cd45.pos.Pd1.neg.S366.H06.S366.comb no 0.11269306 no
#> cy80.Cd45.pos.Pd1.neg.S202.A10.S202.comb no 0.09834696 no
#> cy80.Cd45.pos.Pd1.neg.S201.A09.S201.comb no 0.22256938 no
#> cy80.Cd45.pos.Pd1.neg.S221.B05.S221.comb no 0.12903487 no
#> cy80.Cd45.pos.PD1.pos.A03.S15.comb no 0.27242536 no
#> cy80.Cd45.pos.PD1.pos.B11.S47.comb no 0.94929624 yes
#> cy80.Cd45.pos.PD1.pos.S189.H09.S189.comb no 0.93390248 yes
#> cy80.Cd45.pos.PD1.pos.A05.S17.comb no 0.98161289 yes
#> cy80.Cd45.pos.PD1.pos.C02.S62.comb yes 0.96436674 yes
#> cy80.Cd45.pos.PD1.pos.D12.S96.comb no 0.94848597 yes
#> predicted_cell_type
#> cy80.Cd45.pos.PD1.pos.B09.S45.comb T cells
#> cy80.Cd45.pos.Pd1.neg.S366.H06.S366.comb B cells
#> cy80.Cd45.pos.Pd1.neg.S202.A10.S202.comb B cells
#> cy80.Cd45.pos.Pd1.neg.S201.A09.S201.comb B cells
#> cy80.Cd45.pos.Pd1.neg.S221.B05.S221.comb B cells
#> cy80.Cd45.pos.PD1.pos.A03.S15.comb B cells
#> cy80.Cd45.pos.PD1.pos.B11.S47.comb T cells
#> cy80.Cd45.pos.PD1.pos.S189.H09.S189.comb T cells
#> cy80.Cd45.pos.PD1.pos.A05.S17.comb T cells
#> cy80.Cd45.pos.PD1.pos.C02.S62.comb NK/T cells
#> cy80.Cd45.pos.PD1.pos.D12.S96.comb B cells/T cells
#> most_probable_cell_type
#> cy80.Cd45.pos.PD1.pos.B09.S45.comb T cells
#> cy80.Cd45.pos.Pd1.neg.S366.H06.S366.comb B cells
#> cy80.Cd45.pos.Pd1.neg.S202.A10.S202.comb B cells
#> cy80.Cd45.pos.Pd1.neg.S201.A09.S201.comb B cells
#> cy80.Cd45.pos.Pd1.neg.S221.B05.S221.comb B cells
#> cy80.Cd45.pos.PD1.pos.A03.S15.comb B cells
#> cy80.Cd45.pos.PD1.pos.B11.S47.comb T cells
#> cy80.Cd45.pos.PD1.pos.S189.H09.S189.comb T cells
#> cy80.Cd45.pos.PD1.pos.A05.S17.comb T cells
#> cy80.Cd45.pos.PD1.pos.C02.S62.comb T cells
#> cy80.Cd45.pos.PD1.pos.D12.S96.comb B cellsNew columns are:
predicted_cell_type: The predicted cell type, also containing any ambiguous assignments. In these cases, the possible cell types are separated by a “/”
most_probable_cell_type: contains the most probably cell type ignoring any ambiguous assignments.
columns with syntax
[celltype]_p: probability of a cell to belong to a cell type. Unknown cell types are marked as NAs.
Result visualization
The predicted cell types can now simply be visualized using the
matching plotting functions. In this example, we use Seurat’s
DimPlot function:
With the current number of cell classifiers, we identify cells belonging to 2 cell types (B cells and T cells) and to 2 subtypes of T cells (CD4+ T cells and CD8+ T cells). The other cells (red points) are not among the cell types that can be classified by the predefined classifiers. Hence, they have an empty label.
For a certain cell type, users can also view the prediction probability. Here we show an example of B cell prediction probability:
Cells predicted to be B cells with higher probability have darker color, while the lighter color shows lower or even zero probability of a cell to be B cells. For B cell classifier, the threshold for prediction probability is currently at 0.5, which means cells having prediction probability at 0.5 or above will be predicted as B cells.
The automatic cell identification by scAnnotatR matches the traditional cell assignment, ie. the approach based on cell canonical marker expression. Taking a simple example, we use CD19 and CD20 (MS4A1) to identify B cells:
We see that the marker expression of B cells exactly overlaps the B cell prediction made by scAnnotatR.
Session Info
sessionInfo()
#> R version 4.4.1 (2024-06-14)
#> Platform: x86_64-pc-linux-gnu
#> Running under: Ubuntu 24.04 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-r0.3.26.so; LAPACK version 3.12.0
#>
#> locale:
#> [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
#> [3] LC_TIME=en_US.UTF-8 LC_COLLATE=C
#> [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
#> [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
#> [9] LC_ADDRESS=C LC_TELEPHONE=C
#> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
#>
#> time zone: Etc/UTC
#> tzcode source: system (glibc)
#>
#> attached base packages:
#> [1] stats4 stats graphics grDevices utils datasets methods
#> [8] base
#>
#> other attached packages:
#> [1] scAnnotatR_1.11.0 SingleCellExperiment_1.27.2
#> [3] SummarizedExperiment_1.35.1 Biobase_2.65.1
#> [5] GenomicRanges_1.57.1 GenomeInfoDb_1.41.1
#> [7] IRanges_2.39.2 S4Vectors_0.43.2
#> [9] BiocGenerics_0.51.0 MatrixGenerics_1.17.0
#> [11] matrixStats_1.3.0 Seurat_5.1.0
#> [13] SeuratObject_5.0.2 sp_2.1-4
#> [15] rmarkdown_2.28
#>
#> loaded via a namespace (and not attached):
#> [1] RcppAnnoy_0.0.22 splines_4.4.1 later_1.3.2
#> [4] filelock_1.0.3 tibble_3.2.1 polyclip_1.10-7
#> [7] hardhat_1.4.0 pROC_1.18.5 rpart_4.1.23
#> [10] fastDummies_1.7.4 lifecycle_1.0.4 globals_0.16.3
#> [13] lattice_0.22-6 MASS_7.3-61 magrittr_2.0.3
#> [16] plotly_4.10.4 sass_0.4.9 jquerylib_0.1.4
#> [19] yaml_2.3.10 httpuv_1.6.15 sctransform_0.4.1
#> [22] spam_2.10-0 spatstat.sparse_3.1-0 reticulate_1.38.0
#> [25] cowplot_1.1.3 pbapply_1.7-2 DBI_1.2.3
#> [28] buildtools_1.0.0 RColorBrewer_1.1-3 lubridate_1.9.3
#> [31] abind_1.4-5 zlibbioc_1.51.1 Rtsne_0.17
#> [34] purrr_1.0.2 nnet_7.3-19 rappdirs_0.3.3
#> [37] ipred_0.9-15 lava_1.8.0 GenomeInfoDbData_1.2.12
#> [40] data.tree_1.1.0 ggrepel_0.9.5 irlba_2.3.5.1
#> [43] listenv_0.9.1 spatstat.utils_3.1-0 maketools_1.3.0
#> [46] goftest_1.2-3 RSpectra_0.16-2 spatstat.random_3.3-1
#> [49] fitdistrplus_1.2-1 parallelly_1.38.0 leiden_0.4.3.1
#> [52] codetools_0.2-20 DelayedArray_0.31.11 tidyselect_1.2.1
#> [55] farver_2.1.2 UCSC.utils_1.1.0 BiocFileCache_2.13.0
#> [58] spatstat.explore_3.3-2 jsonlite_1.8.8 caret_6.0-94
#> [61] e1071_1.7-14 progressr_0.14.0 iterators_1.0.14
#> [64] ggridges_0.5.6 survival_3.7-0 foreach_1.5.2
#> [67] tools_4.4.1 ica_1.0-3 Rcpp_1.0.13
#> [70] glue_1.7.0 prodlim_2024.06.25 gridExtra_2.3
#> [73] SparseArray_1.5.31 xfun_0.47 dplyr_1.1.4
#> [76] withr_3.0.1 BiocManager_1.30.25 fastmap_1.2.0
#> [79] fansi_1.0.6 digest_0.6.37 timechange_0.3.0
#> [82] R6_2.5.1 mime_0.12 colorspace_2.1-1
#> [85] scattermore_1.2 tensor_1.5 spatstat.data_3.1-2
#> [88] RSQLite_2.3.7 utf8_1.2.4 tidyr_1.3.1
#> [91] generics_0.1.3 recipes_1.1.0 data.table_1.16.0
#> [94] class_7.3-22 httr_1.4.7 htmlwidgets_1.6.4
#> [97] S4Arrays_1.5.7 ModelMetrics_1.2.2.2 uwot_0.2.2
#> [100] pkgconfig_2.0.3 gtable_0.3.5 timeDate_4032.109
#> [103] blob_1.2.4 lmtest_0.9-40 XVector_0.45.0
#> [106] sys_3.4.2 htmltools_0.5.8.1 dotCall64_1.1-1
#> [109] scales_1.3.0 png_0.1-8 gower_1.0.1
#> [112] spatstat.univar_3.0-0 knitr_1.48 reshape2_1.4.4
#> [115] nlme_3.1-166 curl_5.2.2 proxy_0.4-27
#> [118] cachem_1.1.0 zoo_1.8-12 stringr_1.5.1
#> [121] BiocVersion_3.20.0 KernSmooth_2.23-24 parallel_4.4.1
#> [124] miniUI_0.1.1.1 AnnotationDbi_1.67.0 pillar_1.9.0
#> [127] grid_4.4.1 vctrs_0.6.5 RANN_2.6.2
#> [130] promises_1.3.0 dbplyr_2.5.0 xtable_1.8-4
#> [133] cluster_2.1.6 evaluate_0.24.0 cli_3.6.3
#> [136] compiler_4.4.1 rlang_1.1.4 crayon_1.5.3
#> [139] future.apply_1.11.2 labeling_0.4.3 plyr_1.8.9
#> [142] stringi_1.8.4 viridisLite_0.4.2 deldir_2.0-4
#> [145] munsell_0.5.1 Biostrings_2.73.1 lazyeval_0.2.2
#> [148] spatstat.geom_3.3-2 Matrix_1.7-0 RcppHNSW_0.6.0
#> [151] patchwork_1.2.0 bit64_4.0.5 future_1.34.0
#> [154] ggplot2_3.5.1 KEGGREST_1.45.1 shiny_1.9.1
#> [157] highr_0.11 AnnotationHub_3.13.3 kernlab_0.9-33
#> [160] ROCR_1.0-11 igraph_2.0.3 memoise_2.0.1
#> [163] bslib_0.8.0 bit_4.0.5 ape_5.8