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.

7.2 Data loading

library(scRNAseq)
sce.lawlor <- LawlorPancreasData()
library(AnnotationHub)
edb <- AnnotationHub()[["AH73881"]]
anno <- select(edb, keys=rownames(sce.lawlor), keytype="GENEID", 
    columns=c("SYMBOL", "SEQNAME"))
rowData(sce.lawlor) <- anno[match(rownames(sce.lawlor), anno[,1]),-1]

7.3 Quality control

unfiltered <- sce.lawlor
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
)
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 7.1: 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.

plotColData(unfiltered, x="sum", y="subsets_Mito_percent",
    colour_by="discard") + scale_x_log10()
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.

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.

colSums(as.matrix(qc))
##              low_lib_size            low_n_features high_subsets_Mito_percent 
##                         9                         5                        25 
##                   discard 
##                        34

7.4 Normalization

library(scran)
set.seed(1000)
clusters <- quickCluster(sce.lawlor)
sce.lawlor <- computeSumFactors(sce.lawlor, clusters=clusters)
sce.lawlor <- logNormCounts(sce.lawlor)
summary(sizeFactors(sce.lawlor))
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   0.295   0.781   0.963   1.000   1.182   2.629
plot(librarySizeFactors(sce.lawlor), sizeFactors(sce.lawlor), pch=16,
    xlab="Library size factors", ylab="Deconvolution factors", log="xy")
Relationship between the library size factors and the deconvolution size factors in the Lawlor pancreas dataset.

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)
}
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.

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.

7.6 Dimensionality reduction

library(BiocSingular)
set.seed(101011001)
sce.lawlor <- runPCA(sce.lawlor, subset_row=chosen.genes, ncomponents=25)
sce.lawlor <- runTSNE(sce.lawlor, dimred="PCA")

7.7 Clustering

snn.gr <- buildSNNGraph(sce.lawlor, use.dimred="PCA")
colLabels(sce.lawlor) <- factor(igraph::cluster_walktrap(snn.gr)$membership)
table(colLabels(sce.lawlor), sce.lawlor$`cell type`)
##    
##     Acinar Alpha Beta Delta Ductal Gamma/PP None/Other Stellate
##   1      1     0    0    13      2       16          2        0
##   2      0     1   76     1      0        0          0        0
##   3      0   161    1     0      0        1          2        0
##   4      0     1    0     1      0        0          5       19
##   5      0     0  175     4      1        0          1        0
##   6     22     0    0     0      0        0          0        0
##   7      0    75    0     0      0        0          0        0
##   8      0     0    0     1     20        0          2        0
table(colLabels(sce.lawlor), sce.lawlor$`islet unos id`)
##    
##     ACCG268 ACCR015A ACEK420A ACEL337 ACHY057 ACIB065 ACIW009 ACJV399
##   1       8        2        2       4       4       4       9       1
##   2      14        3        2      33       3       2       4      17
##   3      36       23       14      13      14      14      21      30
##   4       7        1        0       1       0       4       9       4
##   5      34       10        4      39       7      23      24      40
##   6       0        2       13       0       0       0       5       2
##   7      32       12        0       5       6       7       4       9
##   8       1        1        2       1       2       1      12       3
gridExtra::grid.arrange(
    plotTSNE(sce.lawlor, colour_by="label"),
    plotTSNE(sce.lawlor, colour_by="islet unos id"),
    ncol=2
)
Obligatory $t$-SNE plots of the Lawlor pancreas dataset. Each point represents a cell that is colored by cluster (left) or batch (right).

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 4.3.1 (2023-06-16)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 22.04.3 LTS

Matrix products: default
BLAS:   /home/biocbuild/bbs-3.18-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] BiocSingular_1.18.0         scran_1.30.0               
 [3] scater_1.30.0               ggplot2_3.4.4              
 [5] scuttle_1.12.0              ensembldb_2.26.0           
 [7] AnnotationFilter_1.26.0     GenomicFeatures_1.54.0     
 [9] AnnotationDbi_1.64.0        AnnotationHub_3.10.0       
[11] BiocFileCache_2.10.0        dbplyr_2.3.4               
[13] Matrix_1.6-1.1              scRNAseq_2.15.0            
[15] SingleCellExperiment_1.24.0 SummarizedExperiment_1.32.0
[17] Biobase_2.62.0              GenomicRanges_1.54.0       
[19] GenomeInfoDb_1.38.0         IRanges_2.36.0             
[21] S4Vectors_0.40.0            BiocGenerics_0.48.0        
[23] MatrixGenerics_1.14.0       matrixStats_1.0.0          
[25] BiocStyle_2.30.0            rebook_1.12.0              

loaded via a namespace (and not attached):
  [1] rstudioapi_0.15.0             jsonlite_1.8.7               
  [3] CodeDepends_0.6.5             magrittr_2.0.3               
  [5] ggbeeswarm_0.7.2              farver_2.1.1                 
  [7] rmarkdown_2.25                BiocIO_1.12.0                
  [9] zlibbioc_1.48.0               vctrs_0.6.4                  
 [11] memoise_2.0.1                 Rsamtools_2.18.0             
 [13] DelayedMatrixStats_1.24.0     RCurl_1.98-1.12              
 [15] htmltools_0.5.6.1             S4Arrays_1.2.0               
 [17] progress_1.2.2                curl_5.1.0                   
 [19] BiocNeighbors_1.20.0          SparseArray_1.2.0            
 [21] sass_0.4.7                    bslib_0.5.1                  
 [23] cachem_1.0.8                  GenomicAlignments_1.38.0     
 [25] igraph_1.5.1                  mime_0.12                    
 [27] lifecycle_1.0.3               pkgconfig_2.0.3              
 [29] rsvd_1.0.5                    R6_2.5.1                     
 [31] fastmap_1.1.1                 GenomeInfoDbData_1.2.11      
 [33] shiny_1.7.5.1                 digest_0.6.33                
 [35] colorspace_2.1-0              dqrng_0.3.1                  
 [37] irlba_2.3.5.1                 ExperimentHub_2.10.0         
 [39] RSQLite_2.3.1                 beachmat_2.18.0              
 [41] labeling_0.4.3                filelock_1.0.2               
 [43] fansi_1.0.5                   httr_1.4.7                   
 [45] abind_1.4-5                   compiler_4.3.1               
 [47] bit64_4.0.5                   withr_2.5.1                  
 [49] BiocParallel_1.36.0           viridis_0.6.4                
 [51] DBI_1.1.3                     biomaRt_2.58.0               
 [53] rappdirs_0.3.3                DelayedArray_0.28.0          
 [55] bluster_1.12.0                rjson_0.2.21                 
 [57] tools_4.3.1                   vipor_0.4.5                  
 [59] beeswarm_0.4.0                interactiveDisplayBase_1.40.0
 [61] httpuv_1.6.12                 glue_1.6.2                   
 [63] restfulr_0.0.15               promises_1.2.1               
 [65] grid_4.3.1                    Rtsne_0.16                   
 [67] cluster_2.1.4                 generics_0.1.3               
 [69] gtable_0.3.4                  hms_1.1.3                    
 [71] metapod_1.10.0                ScaledMatrix_1.10.0          
 [73] xml2_1.3.5                    utf8_1.2.4                   
 [75] XVector_0.42.0                ggrepel_0.9.4                
 [77] BiocVersion_3.18.0            pillar_1.9.0                 
 [79] stringr_1.5.0                 limma_3.58.0                 
 [81] later_1.3.1                   dplyr_1.1.3                  
 [83] lattice_0.22-5                rtracklayer_1.62.0           
 [85] bit_4.0.5                     tidyselect_1.2.0             
 [87] locfit_1.5-9.8                Biostrings_2.70.0            
 [89] knitr_1.44                    gridExtra_2.3                
 [91] bookdown_0.36                 ProtGenerics_1.34.0          
 [93] edgeR_4.0.0                   xfun_0.40                    
 [95] statmod_1.5.0                 stringi_1.7.12               
 [97] lazyeval_0.2.2                yaml_2.3.7                   
 [99] evaluate_0.22                 codetools_0.2-19             
[101] tibble_3.2.1                  BiocManager_1.30.22          
[103] graph_1.80.0                  cli_3.6.1                    
[105] xtable_1.8-4                  munsell_0.5.0                
[107] jquerylib_0.1.4               Rcpp_1.0.11                  
[109] dir.expiry_1.10.0             png_0.1-8                    
[111] XML_3.99-0.14                 parallel_4.3.1               
[113] ellipsis_0.3.2                blob_1.2.4                   
[115] prettyunits_1.2.0             sparseMatrixStats_1.14.0     
[117] bitops_1.0-7                  viridisLite_0.4.2            
[119] scales_1.2.1                  purrr_1.0.2                  
[121] crayon_1.5.2                  rlang_1.1.1                  
[123] cowplot_1.1.1                 KEGGREST_1.42.0              

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.