Chapter 7 Lawlor human pancreas (SMARTer)
7.1 Introduction
This performs an analysis of the Lawlor et al. (2017) dataset, consisting of human pancreas cells from various donors.
7.3 Quality control
library(scater)
stats <- perCellQCMetrics(sce.lawlor,
subsets=list(Mito=which(rowData(sce.lawlor)$SEQNAME=="MT")))
qc <- quickPerCellQC(stats, percent_subsets="subsets_Mito_percent",
batch=sce.lawlor$`islet unos id`)
sce.lawlor <- sce.lawlor[,!qc$discard]colData(unfiltered) <- cbind(colData(unfiltered), stats)
unfiltered$discard <- qc$discard
gridExtra::grid.arrange(
plotColData(unfiltered, x="islet unos id", y="sum", colour_by="discard") +
scale_y_log10() + ggtitle("Total count") +
theme(axis.text.x = element_text(angle = 90)),
plotColData(unfiltered, x="islet unos id", y="detected",
colour_by="discard") + scale_y_log10() + ggtitle("Detected features") +
theme(axis.text.x = element_text(angle = 90)),
plotColData(unfiltered, x="islet unos id", y="subsets_Mito_percent",
colour_by="discard") + ggtitle("Mito percent") +
theme(axis.text.x = element_text(angle = 90)),
ncol=2
)
Figure 7.1: Distribution of each QC metric across cells from each donor of the Lawlor pancreas dataset. Each point represents a cell and is colored according to whether that cell was discarded.
Figure 7.2: Percentage of mitochondrial reads in each cell in the 416B dataset compared to the total count. Each point represents a cell and is colored according to whether that cell was discarded.
## low_lib_size low_n_features high_subsets_Mito_percent
## 9 5 25
## discard
## 347.4 Normalization
library(scran)
set.seed(1000)
clusters <- quickCluster(sce.lawlor)
sce.lawlor <- computeSumFactors(sce.lawlor, clusters=clusters)
sce.lawlor <- logNormCounts(sce.lawlor)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.295 0.781 0.963 1.000 1.182 2.629plot(librarySizeFactors(sce.lawlor), sizeFactors(sce.lawlor), pch=16,
xlab="Library size factors", ylab="Deconvolution factors", log="xy")
Figure 7.3: Relationship between the library size factors and the deconvolution size factors in the Lawlor pancreas dataset.
7.5 Variance modelling
Using age as a proxy for the donor.
dec.lawlor <- modelGeneVar(sce.lawlor, block=sce.lawlor$`islet unos id`)
chosen.genes <- getTopHVGs(dec.lawlor, n=2000)par(mfrow=c(4,2))
blocked.stats <- dec.lawlor$per.block
for (i in colnames(blocked.stats)) {
current <- blocked.stats[[i]]
plot(current$mean, current$total, main=i, pch=16, cex=0.5,
xlab="Mean of log-expression", ylab="Variance of log-expression")
curfit <- metadata(current)
curve(curfit$trend(x), col='dodgerblue', add=TRUE, lwd=2)
}
Figure 7.4: Per-gene variance as a function of the mean for the log-expression values in the Lawlor pancreas dataset. Each point represents a gene (black) with the mean-variance trend (blue) fitted separately for each donor.
7.7 Clustering
snn.gr <- buildSNNGraph(sce.lawlor, use.dimred="PCA")
colLabels(sce.lawlor) <- factor(igraph::cluster_walktrap(snn.gr)$membership)##
## Acinar Alpha Beta Delta Ductal Gamma/PP None/Other Stellate
## 1 1 0 1 13 2 16 2 0
## 2 0 0 75 1 0 0 0 0
## 3 0 161 1 0 0 1 2 0
## 4 0 1 0 1 0 0 5 19
## 5 22 0 0 0 0 0 0 0
## 6 0 0 174 4 1 0 1 0
## 7 0 76 1 0 0 0 0 0
## 8 0 0 0 1 20 0 2 0##
## ACCG268 ACCR015A ACEK420A ACEL337 ACHY057 ACIB065 ACIW009 ACJV399
## 1 8 2 2 4 4 4 9 2
## 2 13 3 2 33 3 2 4 16
## 3 36 23 14 13 14 14 21 30
## 4 7 1 0 1 0 4 9 4
## 5 0 2 13 0 0 0 5 2
## 6 34 10 4 39 7 23 24 39
## 7 33 12 0 5 6 7 4 10
## 8 1 1 2 1 2 1 12 3gridExtra::grid.arrange(
plotTSNE(sce.lawlor, colour_by="label"),
plotTSNE(sce.lawlor, colour_by="islet unos id"),
ncol=2
)
Figure 5.3: Obligatory \(t\)-SNE plots of the Lawlor pancreas dataset. Each point represents a cell that is colored by cluster (left) or batch (right).
Session Info
R version 4.5.0 RC (2025-04-04 r88126)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 24.04.2 LTS
Matrix products: default
BLAS: /home/biocbuild/bbs-3.21-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] BiocSingular_1.24.0 scran_1.36.0
[3] scater_1.36.0 ggplot2_3.5.2
[5] scuttle_1.18.0 ensembldb_2.32.0
[7] AnnotationFilter_1.32.0 GenomicFeatures_1.60.0
[9] AnnotationDbi_1.70.0 AnnotationHub_3.16.0
[11] BiocFileCache_2.16.0 dbplyr_2.5.0
[13] scRNAseq_2.21.1 SingleCellExperiment_1.30.0
[15] SummarizedExperiment_1.38.0 Biobase_2.68.0
[17] GenomicRanges_1.60.0 GenomeInfoDb_1.44.0
[19] IRanges_2.42.0 S4Vectors_0.46.0
[21] BiocGenerics_0.54.0 generics_0.1.3
[23] MatrixGenerics_1.20.0 matrixStats_1.5.0
[25] BiocStyle_2.36.0 rebook_1.18.0
loaded via a namespace (and not attached):
[1] jsonlite_2.0.0 CodeDepends_0.6.6 magrittr_2.0.3
[4] ggbeeswarm_0.7.2 gypsum_1.4.0 farver_2.1.2
[7] rmarkdown_2.29 BiocIO_1.18.0 vctrs_0.6.5
[10] memoise_2.0.1 Rsamtools_2.24.0 RCurl_1.98-1.17
[13] htmltools_0.5.8.1 S4Arrays_1.8.0 curl_6.2.2
[16] BiocNeighbors_2.2.0 Rhdf5lib_1.30.0 SparseArray_1.8.0
[19] rhdf5_2.52.0 sass_0.4.10 alabaster.base_1.8.0
[22] bslib_0.9.0 alabaster.sce_1.8.0 httr2_1.1.2
[25] cachem_1.1.0 GenomicAlignments_1.44.0 igraph_2.1.4
[28] mime_0.13 lifecycle_1.0.4 pkgconfig_2.0.3
[31] rsvd_1.0.5 Matrix_1.7-3 R6_2.6.1
[34] fastmap_1.2.0 GenomeInfoDbData_1.2.14 digest_0.6.37
[37] colorspace_2.1-1 dqrng_0.4.1 irlba_2.3.5.1
[40] ExperimentHub_2.16.0 RSQLite_2.3.9 beachmat_2.24.0
[43] labeling_0.4.3 filelock_1.0.3 httr_1.4.7
[46] abind_1.4-8 compiler_4.5.0 bit64_4.6.0-1
[49] withr_3.0.2 BiocParallel_1.42.0 viridis_0.6.5
[52] DBI_1.2.3 HDF5Array_1.36.0 alabaster.ranges_1.8.0
[55] alabaster.schemas_1.8.0 rappdirs_0.3.3 DelayedArray_0.34.0
[58] bluster_1.18.0 rjson_0.2.23 tools_4.5.0
[61] vipor_0.4.7 beeswarm_0.4.0 glue_1.8.0
[64] h5mread_1.0.0 restfulr_0.0.15 rhdf5filters_1.20.0
[67] grid_4.5.0 Rtsne_0.17 cluster_2.1.8.1
[70] gtable_0.3.6 metapod_1.16.0 ScaledMatrix_1.16.0
[73] XVector_0.48.0 ggrepel_0.9.6 BiocVersion_3.21.1
[76] pillar_1.10.2 limma_3.64.0 dplyr_1.1.4
[79] lattice_0.22-7 rtracklayer_1.68.0 bit_4.6.0
[82] tidyselect_1.2.1 locfit_1.5-9.12 Biostrings_2.76.0
[85] knitr_1.50 gridExtra_2.3 bookdown_0.43
[88] ProtGenerics_1.40.0 edgeR_4.6.0 xfun_0.52
[91] statmod_1.5.0 UCSC.utils_1.4.0 lazyeval_0.2.2
[94] yaml_2.3.10 evaluate_1.0.3 codetools_0.2-20
[97] tibble_3.2.1 alabaster.matrix_1.8.0 BiocManager_1.30.25
[100] graph_1.86.0 cli_3.6.4 munsell_0.5.1
[103] jquerylib_0.1.4 Rcpp_1.0.14 dir.expiry_1.16.0
[106] png_0.1-8 XML_3.99-0.18 parallel_4.5.0
[109] blob_1.2.4 bitops_1.0-9 viridisLite_0.4.2
[112] alabaster.se_1.8.0 scales_1.3.0 purrr_1.0.4
[115] crayon_1.5.3 rlang_1.1.6 cowplot_1.1.3
[118] KEGGREST_1.48.0 References
Lawlor, N., J. George, M. Bolisetty, R. Kursawe, L. Sun, V. Sivakamasundari, I. Kycia, P. Robson, and M. L. Stitzel. 2017. “Single-cell transcriptomes identify human islet cell signatures and reveal cell-type-specific expression changes in type 2 diabetes.” Genome Res. 27 (2): 208–22.