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.
## 0.088 0.541 0.821 1.000 1.211 13.987plot(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 4.4.0 beta (2024-04-15 r86425)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 22.04.4 LTS
Matrix products: default
BLAS: /home/biocbuild/bbs-3.19-bioc/R/lib/libRblas.so
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.10.0
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_1.0.12 batchelor_1.20.0
[3] scran_1.32.0 scater_1.32.0
[5] ggplot2_3.5.1 scuttle_1.14.0
[7] ensembldb_2.28.0 AnnotationFilter_1.28.0
[9] GenomicFeatures_1.56.0 AnnotationDbi_1.66.0
[11] AnnotationHub_3.12.0 BiocFileCache_2.12.0
[13] dbplyr_2.5.0 scRNAseq_2.18.0
[15] SingleCellExperiment_1.26.0 SummarizedExperiment_1.34.0
[17] Biobase_2.64.0 GenomicRanges_1.56.0
[19] GenomeInfoDb_1.40.0 IRanges_2.38.0
[21] S4Vectors_0.42.0 BiocGenerics_0.50.0
[23] MatrixGenerics_1.16.0 matrixStats_1.3.0
[25] BiocStyle_2.32.0 rebook_1.14.0
loaded via a namespace (and not attached):
[1] BiocIO_1.14.0 bitops_1.0-7
[3] filelock_1.0.3 tibble_3.2.1
[5] CodeDepends_0.6.6 graph_1.82.0
[7] XML_3.99-0.16.1 lifecycle_1.0.4
[9] httr2_1.0.1 edgeR_4.2.0
[11] lattice_0.22-6 alabaster.base_1.4.0
[13] magrittr_2.0.3 limma_3.60.0
[15] sass_0.4.9 rmarkdown_2.26
[17] jquerylib_0.1.4 yaml_2.3.8
[19] metapod_1.12.0 cowplot_1.1.3
[21] DBI_1.2.2 RColorBrewer_1.1-3
[23] ResidualMatrix_1.14.0 abind_1.4-5
[25] zlibbioc_1.50.0 Rtsne_0.17
[27] purrr_1.0.2 RCurl_1.98-1.14
[29] rappdirs_0.3.3 GenomeInfoDbData_1.2.12
[31] ggrepel_0.9.5 irlba_2.3.5.1
[33] alabaster.sce_1.4.0 dqrng_0.3.2
[35] DelayedMatrixStats_1.26.0 codetools_0.2-20
[37] DelayedArray_0.30.0 tidyselect_1.2.1
[39] UCSC.utils_1.0.0 farver_2.1.1
[41] ScaledMatrix_1.12.0 viridis_0.6.5
[43] GenomicAlignments_1.40.0 jsonlite_1.8.8
[45] BiocNeighbors_1.22.0 tools_4.4.0
[47] Rcpp_1.0.12 glue_1.7.0
[49] gridExtra_2.3 SparseArray_1.4.0
[51] xfun_0.43 dplyr_1.1.4
[53] HDF5Array_1.32.0 gypsum_1.0.0
[55] withr_3.0.0 BiocManager_1.30.22
[57] fastmap_1.1.1 rhdf5filters_1.16.0
[59] bluster_1.14.0 fansi_1.0.6
[61] digest_0.6.35 rsvd_1.0.5
[63] R6_2.5.1 mime_0.12
[65] colorspace_2.1-0 RSQLite_2.3.6
[67] paws.storage_0.5.0 utf8_1.2.4
[69] generics_0.1.3 rtracklayer_1.64.0
[71] httr_1.4.7 S4Arrays_1.4.0
[73] pkgconfig_2.0.3 gtable_0.3.5
[75] blob_1.2.4 XVector_0.44.0
[77] htmltools_0.5.8.1 bookdown_0.39
[79] ProtGenerics_1.36.0 scales_1.3.0
[81] alabaster.matrix_1.4.0 png_0.1-8
[83] knitr_1.46 rjson_0.2.21
[85] curl_5.2.1 cachem_1.0.8
[87] rhdf5_2.48.0 BiocVersion_3.19.1
[89] parallel_4.4.0 vipor_0.4.7
[91] restfulr_0.0.15 pillar_1.9.0
[93] grid_4.4.0 alabaster.schemas_1.4.0
[95] vctrs_0.6.5 BiocSingular_1.20.0
[97] beachmat_2.20.0 cluster_2.1.6
[99] beeswarm_0.4.0 evaluate_0.23
[101] cli_3.6.2 locfit_1.5-9.9
[103] compiler_4.4.0 Rsamtools_2.20.0
[105] rlang_1.1.3 crayon_1.5.2
[107] paws.common_0.7.2 labeling_0.4.3
[109] ggbeeswarm_0.7.2 alabaster.se_1.4.0
[111] viridisLite_0.4.2 BiocParallel_1.38.0
[113] munsell_0.5.1 Biostrings_2.72.0
[115] lazyeval_0.2.2 Matrix_1.7-0
[117] dir.expiry_1.12.0 ExperimentHub_2.12.0
[119] sparseMatrixStats_1.16.0 bit64_4.0.5
[121] Rhdf5lib_1.26.0 KEGGREST_1.44.0
[123] statmod_1.5.0 alabaster.ranges_1.4.0
[125] highr_0.10 igraph_2.0.3
[127] memoise_2.0.1 bslib_0.7.0
[129] bit_4.0.5 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.