Chapter 6 Muraro human pancreas (CEL-seq)
6.1 Introduction
This performs an analysis of the Muraro et al. (2016) CEL-seq dataset, consisting of human pancreas cells from various donors.
6.2 Data loading
Converting back to Ensembl identifiers.
library(AnnotationHub)
edb <- AnnotationHub()[["AH73881"]]
gene.symb <- sub("__chr.*$", "", rownames(sce.muraro))
gene.ids <- mapIds(edb, keys=gene.symb,
keytype="SYMBOL", column="GENEID")
# Removing duplicated genes or genes without Ensembl IDs.
keep <- !is.na(gene.ids) & !duplicated(gene.ids)
sce.muraro <- sce.muraro[keep,]
rownames(sce.muraro) <- gene.ids[keep]6.3 Quality control
This dataset lacks mitochondrial genes so we will do without. For the one batch that seems to have a high proportion of low-quality cells, we compute an appropriate filter threshold using a shared median and MAD from the other batches (Figure 6.1).
library(scater)
stats <- perCellQCMetrics(sce.muraro)
qc <- quickPerCellQC(stats, percent_subsets="altexps_ERCC_percent",
batch=sce.muraro$donor, subset=sce.muraro$donor!="D28")
sce.muraro <- sce.muraro[,!qc$discard]colData(unfiltered) <- cbind(colData(unfiltered), stats)
unfiltered$discard <- qc$discard
gridExtra::grid.arrange(
plotColData(unfiltered, x="donor", y="sum", colour_by="discard") +
scale_y_log10() + ggtitle("Total count"),
plotColData(unfiltered, x="donor", y="detected", colour_by="discard") +
scale_y_log10() + ggtitle("Detected features"),
plotColData(unfiltered, x="donor", y="altexps_ERCC_percent",
colour_by="discard") + ggtitle("ERCC percent"),
ncol=2
)
Figure 6.1: Distribution of each QC metric across cells from each donor in the Muraro pancreas dataset. Each point represents a cell and is colored according to whether that cell was discarded.
We have a look at the causes of removal:
## low_lib_size low_n_features high_altexps_ERCC_percent
## 663 700 738
## discard
## 7736.4 Normalization
library(scran)
set.seed(1000)
clusters <- quickCluster(sce.muraro)
sce.muraro <- computeSumFactors(sce.muraro, clusters=clusters)
sce.muraro <- logNormCounts(sce.muraro)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0878 0.5411 0.8208 1.0000 1.2108 13.9869plot(librarySizeFactors(sce.muraro), sizeFactors(sce.muraro), pch=16,
xlab="Library size factors", ylab="Deconvolution factors", log="xy")
Figure 6.2: Relationship between the library size factors and the deconvolution size factors in the Muraro pancreas dataset.
6.5 Variance modelling
We block on a combined plate and donor factor.
block <- paste0(sce.muraro$plate, "_", sce.muraro$donor)
dec.muraro <- modelGeneVarWithSpikes(sce.muraro, "ERCC", block=block)
top.muraro <- getTopHVGs(dec.muraro, prop=0.1)par(mfrow=c(8,4))
blocked.stats <- dec.muraro$per.block
for (i in colnames(blocked.stats)) {
current <- blocked.stats[[i]]
plot(current$mean, current$total, main=i, pch=16, cex=0.5,
xlab="Mean of log-expression", ylab="Variance of log-expression")
curfit <- metadata(current)
points(curfit$mean, curfit$var, col="red", pch=16)
curve(curfit$trend(x), col='dodgerblue', add=TRUE, lwd=2)
}
Figure 6.3: Per-gene variance as a function of the mean for the log-expression values in the Muraro pancreas dataset. Each point represents a gene (black) with the mean-variance trend (blue) fitted to the spike-in transcripts (red) separately for each donor.
6.6 Data integration
library(batchelor)
set.seed(1001010)
merged.muraro <- fastMNN(sce.muraro, subset.row=top.muraro,
batch=sce.muraro$donor)We use the proportion of variance lost as a diagnostic measure:
## D28 D29 D30 D31
## [1,] 0.060847 0.024121 0.000000 0.00000
## [2,] 0.002646 0.003018 0.062421 0.00000
## [3,] 0.003449 0.002641 0.002598 0.081626.8 Clustering
snn.gr <- buildSNNGraph(merged.muraro, use.dimred="corrected")
colLabels(merged.muraro) <- factor(igraph::cluster_walktrap(snn.gr)$membership)tab <- table(Cluster=colLabels(merged.muraro), CellType=sce.muraro$label)
library(pheatmap)
pheatmap(log10(tab+10), color=viridis::viridis(100))
Figure 6.4: Heatmap of the frequency of cells from each cell type label in each cluster.
## Donor
## Cluster D28 D29 D30 D31
## 1 104 6 57 112
## 2 59 21 77 97
## 3 12 75 64 43
## 4 28 149 126 120
## 5 87 261 277 214
## 6 21 7 54 26
## 7 1 6 6 37
## 8 6 6 5 2
## 9 11 68 5 30
## 10 4 2 5 8gridExtra::grid.arrange(
plotTSNE(merged.muraro, colour_by="label"),
plotTSNE(merged.muraro, colour_by="batch"),
ncol=2
)
Figure 6.5: Obligatory \(t\)-SNE plots of the Muraro pancreas dataset. Each point represents a cell that is colored by cluster (left) or batch (right).
Session Info
R version 4.6.0 RC (2026-04-17 r89917)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 24.04.4 LTS
Matrix products: default
BLAS: /home/biocbuild/bbs-3.24-bioc/R/lib/libRblas.so
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.12.0 LAPACK version 3.12.0
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_GB 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: America/New_York
tzcode source: system (glibc)
attached base packages:
[1] stats4 stats graphics grDevices utils datasets methods
[8] base
other attached packages:
[1] pheatmap_1.0.13 batchelor_1.29.0
[3] scran_1.41.0 scater_1.41.1
[5] ggplot2_4.0.3 scuttle_1.23.0
[7] ensembldb_2.37.0 AnnotationFilter_1.37.0
[9] GenomicFeatures_1.65.0 AnnotationDbi_1.75.0
[11] AnnotationHub_4.3.0 BiocFileCache_3.3.0
[13] dbplyr_2.5.2 scRNAseq_2.27.0
[15] SingleCellExperiment_1.35.0 SummarizedExperiment_1.43.0
[17] Biobase_2.73.1 GenomicRanges_1.65.0
[19] Seqinfo_1.3.0 IRanges_2.47.0
[21] S4Vectors_0.51.1 BiocGenerics_0.59.0
[23] generics_0.1.4 MatrixGenerics_1.25.0
[25] matrixStats_1.5.0 BiocStyle_2.41.0
[27] rebook_1.23.0
loaded via a namespace (and not attached):
[1] RColorBrewer_1.1-3 jsonlite_2.0.0
[3] CodeDepends_0.6.7 magrittr_2.0.5
[5] ggbeeswarm_0.7.3 gypsum_1.9.0
[7] farver_2.1.2 rmarkdown_2.31
[9] BiocIO_1.23.3 vctrs_0.7.3
[11] DelayedMatrixStats_1.35.0 memoise_2.0.1
[13] Rsamtools_2.29.0 RCurl_1.98-1.18
[15] htmltools_0.5.9 S4Arrays_1.13.0
[17] BiocBaseUtils_1.15.0 curl_7.1.0
[19] BiocNeighbors_2.7.0 Rhdf5lib_2.1.0
[21] SparseArray_1.13.2 rhdf5_2.57.0
[23] sass_0.4.10 alabaster.base_1.13.0
[25] bslib_0.10.0 alabaster.sce_1.13.0
[27] httr2_1.2.2 cachem_1.1.0
[29] ResidualMatrix_1.23.0 GenomicAlignments_1.49.0
[31] igraph_2.3.1 lifecycle_1.0.5
[33] pkgconfig_2.0.3 rsvd_1.0.5
[35] Matrix_1.7-5 R6_2.6.1
[37] fastmap_1.2.0 digest_0.6.39
[39] dqrng_0.4.1 irlba_2.3.7
[41] ExperimentHub_3.3.0 RSQLite_2.4.6
[43] beachmat_2.29.0 labeling_0.4.3
[45] filelock_1.0.3 httr_1.4.8
[47] abind_1.4-8 compiler_4.6.0
[49] bit64_4.8.0 withr_3.0.2
[51] S7_0.2.2 BiocParallel_1.47.0
[53] viridis_0.6.5 DBI_1.3.0
[55] HDF5Array_1.41.0 alabaster.ranges_1.13.0
[57] alabaster.schemas_1.13.0 rappdirs_0.3.4
[59] DelayedArray_0.39.1 bluster_1.23.0
[61] rjson_0.2.23 tools_4.6.0
[63] vipor_0.4.7 otel_0.2.0
[65] beeswarm_0.4.0 glue_1.8.1
[67] h5mread_1.5.0 restfulr_0.0.16
[69] rhdf5filters_1.25.0 grid_4.6.0
[71] Rtsne_0.17 cluster_2.1.8.2
[73] gtable_0.3.6 metapod_1.21.0
[75] BiocSingular_1.29.0 ScaledMatrix_1.21.0
[77] XVector_0.53.0 ggrepel_0.9.8
[79] BiocVersion_3.24.0 pillar_1.11.1
[81] limma_3.69.0 dplyr_1.2.1
[83] lattice_0.22-9 rtracklayer_1.73.0
[85] bit_4.6.0 tidyselect_1.2.1
[87] locfit_1.5-9.12 Biostrings_2.81.1
[89] knitr_1.51 gridExtra_2.3
[91] bookdown_0.46 ProtGenerics_1.45.0
[93] edgeR_4.11.0 xfun_0.57
[95] statmod_1.5.1 UCSC.utils_1.9.0
[97] lazyeval_0.2.3 yaml_2.3.12
[99] evaluate_1.0.5 codetools_0.2-20
[101] cigarillo_1.3.0 tibble_3.3.1
[103] alabaster.matrix_1.13.0 BiocManager_1.30.27
[105] graph_1.91.0 cli_3.6.6
[107] jquerylib_0.1.4 dichromat_2.0-0.1
[109] Rcpp_1.1.1-1.1 GenomeInfoDb_1.49.0
[111] dir.expiry_1.21.0 png_0.1-9
[113] XML_3.99-0.23 parallel_4.6.0
[115] blob_1.3.0 sparseMatrixStats_1.25.0
[117] bitops_1.0-9 viridisLite_0.4.3
[119] alabaster.se_1.13.0 scales_1.4.0
[121] purrr_1.2.2 crayon_1.5.3
[123] rlang_1.2.0 cowplot_1.2.0
[125] KEGGREST_1.53.0 References
Muraro, M. J., G. Dharmadhikari, D. Grun, N. Groen, T. Dielen, E. Jansen, L. van Gurp, et al. 2016. “A Single-Cell Transcriptome Atlas of the Human Pancreas.” Cell Syst 3 (4): 385–94.