Domain segmentation (STARmap PLUS mouse brain)
30 April 2024
Contents
Here, we demonstrate BANKSY domain segmentation on a STARmap PLUS dataset of the mouse brain from Shi et al. (2022).
library(Banksy)
library(data.table)
library(SummarizedExperiment)
library(SpatialExperiment)
library(scater)
library(cowplot)
library(ggplot2)1 Data preprocessing
Data from the study is available from the Single Cell Portal. We
analyze data from well11. The data comprise 1,022 genes profiled at
subcellular resolution in 43,341 cells.
#' Change paths accordingly
gcm_path <- "../data/well11processed_expression_pd.csv.gz"
mdata_path <- "../data/well11_spatial.csv.gz"
#' Gene cell matrix
gcm <- fread(gcm_path)
genes <- gcm$GENE
gcm <- as.matrix(gcm[, -1])
rownames(gcm) <- genes
#' Spatial coordinates and metadata
mdata <- fread(mdata_path, skip = 1)
headers <- names(fread(mdata_path, nrows = 0))
colnames(mdata) <- headers
#' Orient spatial coordinates
xx <- mdata$X
yy <- mdata$Y
mdata$X <- max(yy) - yy
mdata$Y <- max(xx) - xx
mdata <- data.frame(mdata)
rownames(mdata) <- colnames(gcm)
locs <- as.matrix(mdata[, c("X", "Y", "Z")])
#' Create SpatialExperiment
se <- SpatialExperiment(
assay = list(processedExp = gcm),
spatialCoords = locs,
colData = mdata
)2 Running BANKSY
Run BANKSY in domain segmentation mode with lambda=0.8. This places larger
weights on the mean neighborhood expression and azimuthal Gabor filter in
constructing the BANKSY matrix. We adjust the resolution to yield 23 clusters
based on the results from Maher et al. (2023)
(see Fig. 1, 2).
lambda <- 0.8
k_geom <- 30
npcs <- 50
aname <- "processedExp"
se <- Banksy::computeBanksy(se, assay_name = aname, k_geom = k_geom)
set.seed(1000)
se <- Banksy::runBanksyPCA(se, lambda = lambda, npcs = npcs)
set.seed(1000)
se <- Banksy::clusterBanksy(se, lambda = lambda, npcs = npcs, resolution = 0.8)Cluster labels are stored in the colData slot:
head(colData(se))
#> DataFrame with 6 rows and 4 columns
#> X Y clust_M1_lam0.8_k50_res0.8 sample_id
#> <numeric> <numeric> <factor> <character>
#> 1 24225.5 23984.2 10 sample01
#> 2 24849.2 22679.1 10 sample01
#> 3 24488.3 22970.3 10 sample01
#> 4 24371.4 23727.5 10 sample01
#> 5 24362.2 23300.6 10 sample01
#> 6 24644.5 23112.8 10 sample01Visualize clustering results:
cnames <- colnames(colData(se))
cnames <- cnames[grep("^clust", cnames)]
plotColData(se, x = "X", y = "Y", point_size = 0.01, colour_by = cnames[1]) +
scale_color_manual(values = pals::glasbey()) +
coord_equal() +
theme(legend.position = "none")
3 Session information
sessionInfo()
#> 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_US.UTF-8 LC_COLLATE=en_US.UTF-8
#> [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] ExperimentHub_2.12.0 AnnotationHub_3.12.0
#> [3] BiocFileCache_2.12.0 dbplyr_2.5.0
#> [5] spatialLIBD_1.15.4 cowplot_1.1.3
#> [7] scater_1.32.0 ggplot2_3.5.1
#> [9] harmony_1.2.0 Rcpp_1.0.12
#> [11] data.table_1.15.4 scran_1.32.0
#> [13] scuttle_1.14.0 Seurat_5.0.3
#> [15] SeuratObject_5.0.1 sp_2.1-4
#> [17] SpatialExperiment_1.14.0 SingleCellExperiment_1.26.0
#> [19] SummarizedExperiment_1.34.0 Biobase_2.64.0
#> [21] GenomicRanges_1.56.0 GenomeInfoDb_1.40.0
#> [23] IRanges_2.38.0 S4Vectors_0.42.0
#> [25] BiocGenerics_0.50.0 MatrixGenerics_1.16.0
#> [27] matrixStats_1.3.0 Banksy_1.0.0
#> [29] BiocStyle_2.32.0
#>
#> loaded via a namespace (and not attached):
#> [1] bitops_1.0-7 spatstat.sparse_3.0-3
#> [3] doParallel_1.0.17 httr_1.4.7
#> [5] RColorBrewer_1.1-3 tools_4.4.0
#> [7] sctransform_0.4.1 DT_0.33
#> [9] utf8_1.2.4 R6_2.5.1
#> [11] lazyeval_0.2.2 uwot_0.2.2
#> [13] withr_3.0.0 gridExtra_2.3
#> [15] progressr_0.14.0 cli_3.6.2
#> [17] spatstat.explore_3.2-7 fastDummies_1.7.3
#> [19] labeling_0.4.3 sass_0.4.9
#> [21] spatstat.data_3.0-4 ggridges_0.5.6
#> [23] pbapply_1.7-2 Rsamtools_2.20.0
#> [25] dbscan_1.1-12 aricode_1.0.3
#> [27] dichromat_2.0-0.1 sessioninfo_1.2.2
#> [29] parallelly_1.37.1 attempt_0.3.1
#> [31] maps_3.4.2 limma_3.60.0
#> [33] pals_1.8 RSQLite_2.3.6
#> [35] BiocIO_1.14.0 generics_0.1.3
#> [37] ica_1.0-3 spatstat.random_3.2-3
#> [39] dplyr_1.1.4 Matrix_1.7-0
#> [41] ggbeeswarm_0.7.2 fansi_1.0.6
#> [43] abind_1.4-5 lifecycle_1.0.4
#> [45] yaml_2.3.8 edgeR_4.2.0
#> [47] SparseArray_1.4.0 Rtsne_0.17
#> [49] paletteer_1.6.0 grid_4.4.0
#> [51] blob_1.2.4 promises_1.3.0
#> [53] dqrng_0.3.2 crayon_1.5.2
#> [55] miniUI_0.1.1.1 lattice_0.22-6
#> [57] beachmat_2.20.0 mapproj_1.2.11
#> [59] KEGGREST_1.44.0 magick_2.8.3
#> [61] pillar_1.9.0 knitr_1.46
#> [63] metapod_1.12.0 rjson_0.2.21
#> [65] future.apply_1.11.2 codetools_0.2-20
#> [67] leiden_0.4.3.1 glue_1.7.0
#> [69] vctrs_0.6.5 png_0.1-8
#> [71] spam_2.10-0 gtable_0.3.5
#> [73] rematch2_2.1.2 cachem_1.0.8
#> [75] xfun_0.43 S4Arrays_1.4.0
#> [77] mime_0.12 survival_3.6-4
#> [79] RcppHungarian_0.3 iterators_1.0.14
#> [81] tinytex_0.50 fields_15.2
#> [83] statmod_1.5.0 bluster_1.14.0
#> [85] fitdistrplus_1.1-11 ROCR_1.0-11
#> [87] nlme_3.1-164 bit64_4.0.5
#> [89] filelock_1.0.3 RcppAnnoy_0.0.22
#> [91] bslib_0.7.0 irlba_2.3.5.1
#> [93] vipor_0.4.7 KernSmooth_2.23-22
#> [95] colorspace_2.1-0 DBI_1.2.2
#> [97] tidyselect_1.2.1 bit_4.0.5
#> [99] compiler_4.4.0 curl_5.2.1
#> [101] BiocNeighbors_1.22.0 DelayedArray_0.30.0
#> [103] plotly_4.10.4 rtracklayer_1.64.0
#> [105] bookdown_0.39 scales_1.3.0
#> [107] lmtest_0.9-40 rappdirs_0.3.3
#> [109] stringr_1.5.1 digest_0.6.35
#> [111] goftest_1.2-3 spatstat.utils_3.0-4
#> [113] rmarkdown_2.26 benchmarkmeData_1.0.4
#> [115] RhpcBLASctl_0.23-42 XVector_0.44.0
#> [117] htmltools_0.5.8.1 pkgconfig_2.0.3
#> [119] sparseMatrixStats_1.16.0 highr_0.10
#> [121] fastmap_1.1.1 rlang_1.1.3
#> [123] htmlwidgets_1.6.4 UCSC.utils_1.0.0
#> [125] shiny_1.8.1.1 DelayedMatrixStats_1.26.0
#> [127] farver_2.1.1 jquerylib_0.1.4
#> [129] zoo_1.8-12 jsonlite_1.8.8
#> [131] BiocParallel_1.38.0 mclust_6.1.1
#> [133] config_0.3.2 RCurl_1.98-1.14
#> [135] BiocSingular_1.20.0 magrittr_2.0.3
#> [137] GenomeInfoDbData_1.2.12 dotCall64_1.1-1
#> [139] patchwork_1.2.0 munsell_0.5.1
#> [141] viridis_0.6.5 reticulate_1.36.1
#> [143] leidenAlg_1.1.3 stringi_1.8.3
#> [145] zlibbioc_1.50.0 MASS_7.3-60.2
#> [147] plyr_1.8.9 parallel_4.4.0
#> [149] listenv_0.9.1 ggrepel_0.9.5
#> [151] deldir_2.0-4 Biostrings_2.72.0
#> [153] sccore_1.0.5 splines_4.4.0
#> [155] tensor_1.5 locfit_1.5-9.9
#> [157] igraph_2.0.3 spatstat.geom_3.2-9
#> [159] RcppHNSW_0.6.0 reshape2_1.4.4
#> [161] ScaledMatrix_1.12.0 XML_3.99-0.16.1
#> [163] BiocVersion_3.19.1 evaluate_0.23
#> [165] golem_0.4.1 BiocManager_1.30.22
#> [167] foreach_1.5.2 httpuv_1.6.15
#> [169] RANN_2.6.1 tidyr_1.3.1
#> [171] purrr_1.0.2 polyclip_1.10-6
#> [173] benchmarkme_1.0.8 future_1.33.2
#> [175] scattermore_1.2 rsvd_1.0.5
#> [177] xtable_1.8-4 restfulr_0.0.15
#> [179] RSpectra_0.16-1 later_1.3.2
#> [181] viridisLite_0.4.2 tibble_3.2.1
#> [183] GenomicAlignments_1.40.0 memoise_2.0.1
#> [185] beeswarm_0.4.0 AnnotationDbi_1.66.0
#> [187] cluster_2.1.6 shinyWidgets_0.8.6
#> [189] globals_0.16.3