scRNAseq_complete_03-2_Subcelltype_processing
retogerber
2022-12-06
Last updated: 2022-12-06
Checks: 6 1
Knit directory: synovialscrnaseq/
This reproducible R Markdown analysis was created with workflowr (version 1.6.2). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.
The R Markdown file has unstaged changes. To know which version of the R Markdown file created these results, you’ll want to first commit it to the Git repo. If you’re still working on the analysis, you can ignore this warning. When you’re finished, you can run wflow_publish to commit the R Markdown file and build the HTML.
Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.
The command set.seed(20210105) was run prior to running the code in the R Markdown file. Setting a seed ensures that any results that rely on randomness, e.g. subsampling or permutations, are reproducible.
Great job! Recording the operating system, R version, and package versions is critical for reproducibility.
Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.
Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.
Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility.
The results in this page were generated with repository version 816d5c9. See the Past versions tab to see a history of the changes made to the R Markdown and HTML files.
Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated:
Ignored files:
Ignored: '/
Ignored: .Rhistory
Ignored: .Rproj.user/
Ignored: .empty/
Ignored: analysis/.Rhistory
Ignored: analysis/figure/
Ignored: analysis/iSEE_interactive_document.html
Ignored: code/test_files/
Ignored: data/Culemann/
Ignored: data/E-MTAB-8322/
Ignored: data/Synovial scRNA-seq samples - Sheet1.csv
Ignored: data/Zhang_top20_singlecell_cluster_markers_fromGithub.csv
Ignored: data/findMarkers_results.rds
Ignored: data/findMarkers_results_v2.rds
Ignored: data/info/
Ignored: data/syn_sce_tidy_filtered.rds
Ignored: data/syn_sce_tidy_hvg.rds
Ignored: data/syn_sce_tidy_hvg_cms.rds
Ignored: docs/
Ignored: output/Figures_Paper/
Ignored: output/Sample_summaries_RA_comparisons.rds
Ignored: output/Sample_summaries_direct_dissociation.rds
Ignored: output/Sample_summaries_exvivo_treatment.rds
Ignored: output/Suppl_Figure_4d.rds
Ignored: output/barcodes.txt
Ignored: output/count_matrix_unfiltered.mtx
Ignored: output/emptyDrops_result_v4.rds
Ignored: output/emptyDrops_result_v4_tmp.rds
Ignored: output/emptyDrops_result_v4tmptmp.rds
Ignored: output/entropies_fstat_v5_ec.rds
Ignored: output/entropies_fstat_v5_main.rds
Ignored: output/entropies_fstat_v5_mp.rds
Ignored: output/entropies_fstat_v5_sf.rds
Ignored: output/entropies_fstat_v5_tc.rds
Ignored: output/findMarkers_results_v5_ec.rds
Ignored: output/findMarkers_results_v5_main.rds
Ignored: output/findMarkers_results_v5_mp.rds
Ignored: output/findMarkers_results_v5_sf.rds
Ignored: output/findMarkers_results_v5_tc.rds
Ignored: output/findMarkers_results_v6.rds
Ignored: output/findMarkers_results_v6_ec.rds
Ignored: output/findMarkers_results_v6_main.rds
Ignored: output/findMarkers_results_v6_mp.rds
Ignored: output/findMarkers_results_v6_sf.rds
Ignored: output/findMarkers_results_v6_tc.rds
Ignored: output/findMarkers_results_v7_ec.rds
Ignored: output/findMarkers_results_v7_main.rds
Ignored: output/findMarkers_results_v7_mp.rds
Ignored: output/findMarkers_results_v7_sf.rds
Ignored: output/findMarkers_results_v7_tc.rds
Ignored: output/genes.txt
Ignored: output/goana_results_v6_ec.rds
Ignored: output/goana_results_v6_mp.rds
Ignored: output/preprocessing_number_of_cells.rds
Ignored: output/syn_v4_sce_emptyDrops_invivo.rds
Ignored: output/syn_v4_swappedDrops_24300_after.rds
Ignored: output/syn_v4_swappedDrops_24300_before.rds
Ignored: output/syn_v4_swappedDrops_24793_after.rds
Ignored: output/syn_v4_swappedDrops_24793_before.rds
Ignored: output/syn_v5_annot_df_manual.rds
Ignored: output/syn_v5_cluster_cellid_match_invivo.rds
Ignored: output/syn_v5_clustering_lookup_invivo.rds
Ignored: output/syn_v5_clustering_lookup_multiple_invivo.rds
Ignored: output/syn_v5_res_da_Accute_inflammation_invivo.rds
Ignored: output/syn_v5_res_da_Diagnosis_invivo.rds
Ignored: output/syn_v5_res_da_Diagnosis_main_invivo.rds
Ignored: output/syn_v5_res_da_Lymphoid_folicles_invivo.rds
Ignored: output/syn_v5_res_da_Pathotype_invivo.rds
Ignored: output/syn_v5_res_da_Therapy_invivo.rds
Ignored: output/syn_v5_res_da_Vascularisation_bin_invivo.rds
Ignored: output/syn_v5_res_ds_Accute_inflammation_invivo.rds
Ignored: output/syn_v5_res_ds_Diagnosis_invivo.rds
Ignored: output/syn_v5_res_ds_Diagnosis_main_invivo.rds
Ignored: output/syn_v5_res_ds_Lymphoid_folicles_invivo.rds
Ignored: output/syn_v5_res_ds_Pathotype_invivo.rds
Ignored: output/syn_v5_res_ds_Therapy_invivo.rds
Ignored: output/syn_v5_res_ds_Vascularisation_bin_invivo.rds
Ignored: output/syn_v5_sce.rds
Ignored: output/syn_v5_sce_ec_invivo.rds
Ignored: output/syn_v5_sce_filtered_invivo.rds
Ignored: output/syn_v5_sce_hvg_cms_doublet_annot_manual_invivo.rds
Ignored: output/syn_v5_sce_hvg_cms_doublet_cmstest_invivo.rds
Ignored: output/syn_v5_sce_hvg_cms_doublet_invivo.rds
Ignored: output/syn_v5_sce_hvg_cms_doublet_subcluster_invivo.rds
Ignored: output/syn_v5_sce_hvg_invivo.rds
Ignored: output/syn_v5_sce_mp_invivo.rds
Ignored: output/syn_v5_sce_sf_invivo.rds
Ignored: output/syn_v5_sce_tc_invivo.rds
Ignored: output/syn_v5_vst_out_invivo.rds
Ignored: output/syn_v6_cluster_cellid_match_invivo.rds
Ignored: output/syn_v6_clustering_lookup_invivo.rds
Ignored: output/syn_v6_clustering_lookup_multiple_invivo.rds
Ignored: output/syn_v6_sce.rds
Ignored: output/syn_v6_sce_Figure8.rds
Ignored: output/syn_v6_sce_Figure8_dic_ls.rds
Ignored: output/syn_v6_sce_ec_invivo.rds
Ignored: output/syn_v6_sce_filtered_invivo.rds
Ignored: output/syn_v6_sce_hdf5/
Ignored: output/syn_v6_sce_hvg_cms_doublet_invivo.rds
Ignored: output/syn_v6_sce_hvg_cms_doublet_subcluster_invivo.rds
Ignored: output/syn_v6_sce_hvg_invivo.rds
Ignored: output/syn_v6_sce_hvg_marker_genes.rds
Ignored: output/syn_v6_sce_mp_invivo.rds
Ignored: output/syn_v6_sce_sf_invivo.rds
Ignored: output/syn_v6_sce_tc_invivo.rds
Ignored: output/syn_v6_sfig1.rds
Ignored: output/syn_v6_vst_out_invivo.rds
Ignored: output/syn_v7_cluster_cellid_match_invivo.rds
Ignored: output/syn_v7_clustering_lookup_invivo.rds
Ignored: output/syn_v7_clustering_lookup_multiple_invivo.rds
Ignored: output/syn_v7_sce.rds
Ignored: output/syn_v7_sce_ec_invivo.rds
Ignored: output/syn_v7_sce_filtered_invivo.rds
Ignored: output/syn_v7_sce_hdf5/
Ignored: output/syn_v7_sce_hvg_cms_doublet_invivo.rds
Ignored: output/syn_v7_sce_hvg_cms_doublet_subcluster_invivo.rds
Ignored: output/syn_v7_sce_hvg_invivo.rds
Ignored: output/syn_v7_sce_mp_invivo.rds
Ignored: output/syn_v7_sce_sf_invivo.rds
Ignored: output/syn_v7_sce_tc_invivo.rds
Ignored: output/syn_v7_sfig1.rds
Ignored: output/syn_v7_vst_out_invivo.rds
Untracked files:
Untracked: analysis/scRNAseq_complete_01_preprocessing_comparison.Rmd
Untracked: analysis/scRNAseq_complete_04_Annotation_v7.Rmd
Untracked: analysis/test.Rmd
Untracked: code/rebuild_ezRun.R
Untracked: nonhosted_public/
Untracked: singRstudio.sh.bak
Unstaged changes:
Modified: analysis/scRNAseq_complete_01_preprocessing.Rmd
Modified: analysis/scRNAseq_complete_02_HVG_Dimred.Rmd
Modified: analysis/scRNAseq_complete_03-2_Subcelltypes_processing.Rmd
Modified: analysis/scRNAseq_complete_03-3_Subcelltypes_clustering.Rmd
Modified: analysis/scRNAseq_complete_03-4_Subcelltypes_clustering_walktrap.Rmd
Modified: analysis/scRNAseq_complete_03_Batch_Clustering_Doublets.Rmd
Modified: analysis/scRNAseq_complete_04-2_celltype_markers.Rmd
Modified: analysis/scRNAseq_complete_04-2_celltype_markers_subcelltypes.Rmd
Modified: analysis/scRNAseq_complete_Figures.Rmd
Modified: analysis/write_tsv.Rmd
Modified: code/create_hdf5.R
Staged changes:
Modified: analysis/scRNAseq_complete_00_ambient_RNA.Rmd
Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.
These are the previous versions of the repository in which changes were made to the R Markdown (analysis/scRNAseq_complete_03-2_Subcelltypes_processing.Rmd) and HTML (public/scRNAseq_complete_03-2_Subcelltypes_processing.html) files. If you’ve configured a remote Git repository (see ?wflow_git_remote), click on the hyperlinks in the table below to view the files as they were in that past version.
| File | Version | Author | Date | Message |
|---|---|---|---|---|
| html | 3443cc6 | Reto Gerber | 2022-04-25 | Update |
| html | b5b139f | Reto Gerber | 2022-03-29 | Update analysis |
| Rmd | 7d99571 | Reto Gerber | 2022-03-21 | update analysis |
| html | 7d99571 | Reto Gerber | 2022-03-21 | update analysis |
| Rmd | 9133ed1 | Reto Gerber | 2022-03-04 | update to v6 |
| html | 9133ed1 | Reto Gerber | 2022-03-04 | update to v6 |
| html | f2e34e1 | Reto Gerber | 2021-07-29 | Update navbar |
| Rmd | ee1face | Reto Gerber | 2021-07-29 | Add missing scripts |
| html | 222b0d1 | Reto Gerber | 2021-07-29 | Update analysis to v5 |
Set up
suppressPackageStartupMessages({
library(dplyr)
library(ggplot2)
library(purrr)
library(stringr)
library(SummarizedExperiment)
library(SingleCellExperiment)
library(scater)
library(scran)
library(igraph)
library(scuttle)
library(tidySingleCellExperiment)
library(bluster)
library(CellID)
library(BiocParallel)
})
n_workers <- 20
RhpcBLASctl::blas_set_num_threads(n_workers)
bpparam <- BiocParallel::MulticoreParam(workers=n_workers, RNGseed = 123)
here::here()[1] "/home/retger/Synovial/synovialscrnaseq"remove_low_quality_samples <- TRUE
analysis_version <- 7
set.seed(100)tmpfilename <- paste0("syn_v",analysis_version,"_sce_hvg_cms_doublet",dplyr::if_else(remove_low_quality_samples, "_invivo",""),".rds")
syn_sce_tidy_hvg_cms <- readRDS(file = here::here("output",tmpfilename))
tmpfilename <- paste0("syn_v",analysis_version,"_cluster_cellid_match",dplyr::if_else(remove_low_quality_samples, "_invivo",""),".rds")
cluster_assignment <- readRDS(file = here::here("output",tmpfilename))
# list(cluster_names_prop=cluster_names_prop,cluster_names_max=cluster_names_max)
cluster_assignment$cluster_names_max$tc
[1] "4" "9" "13" "21"
$ec
[1] "5" "12"
$sf
[1] "6" "11" "15" "20"
$mp
[1] "7" "8" "14" "16" "18" "22"cluster_assignment$cluster_names_prop$tc
[1] "4" "9" "13" "21"
$ec
[1] "5" "12"
$sf
[1] "6" "11" "15" "20"
$mp
[1] "8" "14" "18" "22"Manually add clusters to main celltypes
# cluster 1 to t-cells
cluster_assignment$cluster_names_max$tc <- c(cluster_assignment$cluster_names_prop$tc,"1")
# cluster 2 to macrophages
cluster_assignment$cluster_names_max$mp <- c(cluster_assignment$cluster_names_prop$mp,"2")if(remove_low_quality_samples){
merge_order <- list(
list("Syn_Bio_053","Syn_Bio_054A",
"Syn_Bio_062","Syn_Bio_064"),#Peripheral_Spondyloarthritis
list("Syn_Bio_023","Syn_Bio_079","Syn_Bio_092"),#Psoriatic_Arthritis
list("Syn_Bio_083","Syn_Bio_084"),#Rheumatoid_Arthritis
list("Syn_Bio_074"),#Seronegative_Polyarthritis
list("Syn_Bio_026","Syn_Bio_028",
list("Syn_Bio_077b","Syn_Bio_077a"),
"Syn_Bio_049","Syn_Bio_050", "Syn_Bio_081",
"Syn_Bio_093","Syn_Bio_096",
list("Syn_Bio_098a","Syn_Bio_098b")
),#Seronegative_Rheumatoid_Arthritis
list("Syn_Bio_091","Syn_Bio_099"),#Spondyloarthritis
list("Syn_Bio_087"),#To_be_determined
list("Syn_Bio_078")#Undiff._Polyarthritis
)
} else{
merge_order <- list(list("Syn_Bio_086"),#special
list("Syn_Bio_053","Syn_Bio_054A",
"Syn_Bio_062","Syn_Bio_064"),#Peripheral_Spondyloarthritis
list("Syn_Bio_023","Syn_Bio_079","Syn_Bio_092"),#Psoriatic_Arthritis
list("Syn_Bio_083","Syn_Bio_084"),#Rheumatoid_Arthritis
list("Syn_Bio_074"),#Seronegative_Polyarthritis
list("Syn_Bio_026","Syn_Bio_028",
list("Syn_Bio_077b","Syn_Bio_077a"),
"Syn_Bio_049","Syn_Bio_050", "Syn_Bio_081",
"Syn_Bio_093","Syn_Bio_096",
list("Syn_Bio_098a","Syn_Bio_098b"),
list(
list("Syn_Bio_055_DMSO","Syn_Bio_072_DMSO","Syn_Bio_094_DMSO"),
list("Syn_Bio_055_Tofa","Syn_Bio_072_Tofa","Syn_Bio_094_Tofa"))
),#Seronegative_Rheumatoid_Arthritis
list("Syn_Bio_075","Syn_Bio_091","Syn_Bio_099"),#Spondyloarthritis
list("Syn_Bio_059","Syn_Bio_080","Syn_Bio_089","Syn_Bio_095"),#Systemic_Sclerosis
list("Syn_Bio_087"),#To_be_determined
list("Syn_Bio_031","Syn_Bio_078")#Undiff._Polyarthritis
)
}
clusters_lookup <- list()Indiviual analysis
Repeat analysis for the four main cell types of interest.
Subclustering - SF
celltype_name_pre <- "sf"HVG selection
set.seed(100)
sce_sub <- syn_sce_tidy_hvg_cms[,syn_sce_tidy_hvg_cms$kgraph_clusters %in% cluster_assignment$cluster_names_max[[celltype_name_pre]]]
assay(sce_sub, "vstresiduals") <- NULL
bpstart(bpparam)
all_gene_var <- modelGeneVar(sce_sub, block=sce_sub$Sample, BPPARAM=bpparam)
bpstop(bpparam)
hvg <- getTopHVGs(all_gene_var, fdr.threshold=0.05)
UpSetR::upset(UpSetR::fromList(list(subset=hvg, all=rownames(syn_sce_tidy_hvg_cms)[rowData(syn_sce_tidy_hvg_cms)$is_hvg])))
rowData(sce_sub)$is_hvg <- rownames(sce_sub) %in% hvgBatch correction
bpstart(bpparam)
temp_sce <- batchelor::multiBatchNorm(sce_sub,
batch=sce_sub$Sample,
subset.row = rownames(sce_sub)[rowData(sce_sub)[["is_hvg"]]],
normalize.all=TRUE,
BPPARAM = bpparam)
bpstop(bpparam)
stopifnot(all(unlist(merge_order) %in% unique(sce_sub$Sample)))
stopifnot(all(unique(sce_sub$Sample) %in% unlist(merge_order)))
bpstart(bpparam)
temp_sce <- batchelor::fastMNN(temp_sce, batch=temp_sce$Sample, prop.k=0.02, merge.order = merge_order,
subset.row = rownames(temp_sce)[rowData(temp_sce)[["is_hvg"]]], correct.all=TRUE,
BPPARAM = bpparam)
bpstop(bpparam)
assay(sce_sub, "reconstructed") <- assay(temp_sce, "reconstructed")
reducedDim(sce_sub, "corrected") <- reducedDim(temp_sce, "corrected")
set.seed(100)
bpstart(bpparam)
sce_sub <- runUMAP(sce_sub, dimred = "corrected", name = "UMAP_corrected",
BPPARAM = bpparam)
bpstop(bpparam)cms
set.seed(123)
bpstart(bpparam)
sce_sub <- CellMixS::cms(sce_sub, k=300, group = "Sample",
dim_red = "PCA", res_name = "unaligned",
BPPARAM = bpparam)
bpstop(bpparam)
bpstart(bpparam)
sce_sub <- CellMixS::cms(sce_sub, k=300, group = "Sample",
dim_red = "corrected", res_name = "MNN",
BPPARAM = bpparam)
bpstop(bpparam)
CellMixS::visHist(sce_sub)
ggpubr::ggarrange(ncol=1,nrow=3,
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="Sample"),
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="cms.MNN"),
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="cms.unaligned")
)
clustering
set.seed(100)
bpstart(bpparam)
graph_sub <- buildSNNGraph(sce_sub, use.dimred="corrected", k=20, BPPARAM = bpparam)
bpstop(bpparam)
clusters <- igraph::cluster_louvain(graph_sub)$membership
colData(sce_sub)[[paste0(celltype_name_pre,"_clusters")]] <- factor(clusters)
clusters_lookup[[celltype_name_pre]] <- data.frame(cell_id = colnames(sce_sub),
cluster = paste0(toupper(celltype_name_pre),"_", as.character(clusters)))
plotReducedDim(sce_sub,"UMAP_corrected",
colour_by=paste0(celltype_name_pre,"_clusters"),
text_by=paste0(celltype_name_pre,"_clusters"))+
theme(legend.position = c(1.01,0.7),
legend.background = element_rect(color="grey",fill = "white"),
legend.margin = margin(10,10,10,10), plot.margin = margin(t=10,r=250,b=0,l=10))
save
tmpfilename <- paste0("syn_v",analysis_version,"_sce_",celltype_name_pre,dplyr::if_else(remove_low_quality_samples, "_invivo",""),".rds")
saveRDS(sce_sub, file = here::here("output",tmpfilename))Subclustering - EC
celltype_name_pre <- "ec"HVG selection
set.seed(100)
sce_sub <- syn_sce_tidy_hvg_cms[,syn_sce_tidy_hvg_cms$kgraph_clusters %in% cluster_assignment$cluster_names_max[[celltype_name_pre]]]
assay(sce_sub, "vstresiduals") <- NULL
bpstart(bpparam)
all_gene_var <- modelGeneVar(sce_sub, block=sce_sub$Sample, BPPARAM=bpparam)Warning in regularize.values(x, y, ties, missing(ties), na.rm = na.rm):
collapsing to unique 'x' valuesbpstop(bpparam)
hvg <- getTopHVGs(all_gene_var, fdr.threshold=0.05)
UpSetR::upset(UpSetR::fromList(list(subset=hvg, all=rownames(syn_sce_tidy_hvg_cms)[rowData(syn_sce_tidy_hvg_cms)$is_hvg])))
rowData(sce_sub)$is_hvg <- rownames(sce_sub) %in% hvgBatch correction
bpstart(bpparam)
temp_sce <- batchelor::multiBatchNorm(sce_sub,
batch=sce_sub$Sample,
subset.row = rownames(sce_sub)[rowData(sce_sub)[["is_hvg"]]],
normalize.all=TRUE,
BPPARAM = bpparam)
bpstop(bpparam)
stopifnot(all(unlist(merge_order) %in% unique(sce_sub$Sample)))
stopifnot(all(unique(sce_sub$Sample) %in% unlist(merge_order)))
bpstart(bpparam)
temp_sce <- batchelor::fastMNN(temp_sce, batch=temp_sce$Sample, prop.k=0.02, merge.order = merge_order,
subset.row = rownames(temp_sce)[rowData(temp_sce)[["is_hvg"]]], correct.all=TRUE,
BPPARAM = bpparam)
bpstop(bpparam)
assay(sce_sub, "reconstructed") <- assay(temp_sce, "reconstructed")
reducedDim(sce_sub, "corrected") <- reducedDim(temp_sce, "corrected")
set.seed(100)
bpstart(bpparam)
sce_sub <- runUMAP(sce_sub, dimred = "corrected", name = "UMAP_corrected",
BPPARAM = bpparam)
bpstop(bpparam)cms
set.seed(123)
bpstart(bpparam)
sce_sub <- CellMixS::cms(sce_sub, k=300, group = "Sample",
dim_red = "PCA", res_name = "unaligned",
BPPARAM = bpparam)
bpstop(bpparam)
bpstart(bpparam)
sce_sub <- CellMixS::cms(sce_sub, k=300, group = "Sample",
dim_red = "corrected", res_name = "MNN",
BPPARAM = bpparam)
bpstop(bpparam)
CellMixS::visHist(sce_sub)
ggpubr::ggarrange(ncol=1,nrow=3,
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="Sample"),
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="cms.MNN"),
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="cms.unaligned")
)
clustering
set.seed(100)
bpstart(bpparam)
graph_sub <- buildSNNGraph(sce_sub, use.dimred="corrected", k=20, BPPARAM = bpparam)
bpstop(bpparam)
clusters <- igraph::cluster_louvain(graph_sub)$membership
colData(sce_sub)[[paste0(celltype_name_pre,"_clusters")]] <- factor(clusters)
clusters_lookup[[celltype_name_pre]] <- data.frame(cell_id = colnames(sce_sub),
cluster = paste0(toupper(celltype_name_pre),"_", as.character(clusters)))
plotReducedDim(sce_sub,"UMAP_corrected",
colour_by=paste0(celltype_name_pre,"_clusters"),
text_by=paste0(celltype_name_pre,"_clusters"))+
theme(legend.position = c(1.01,0.7),
legend.background = element_rect(color="grey",fill = "white"),
legend.margin = margin(10,10,10,10), plot.margin = margin(t=10,r=250,b=0,l=10))
save
tmpfilename <- paste0("syn_v",analysis_version,"_sce_",celltype_name_pre,dplyr::if_else(remove_low_quality_samples, "_invivo",""),".rds")
saveRDS(sce_sub, file = here::here("output",tmpfilename))Subclustering - Macrophages
celltype_name_pre <- "mp"HVG selection
set.seed(100)
sce_sub <- syn_sce_tidy_hvg_cms[,syn_sce_tidy_hvg_cms$kgraph_clusters %in% cluster_assignment$cluster_names_max[[celltype_name_pre]]]
assay(sce_sub, "vstresiduals") <- NULL
bpstart(bpparam)
all_gene_var <- modelGeneVar(sce_sub, block=sce_sub$Sample, BPPARAM=bpparam)
bpstop(bpparam)
hvg <- getTopHVGs(all_gene_var, fdr.threshold=0.05)
UpSetR::upset(UpSetR::fromList(list(subset=hvg, all=rownames(syn_sce_tidy_hvg_cms)[rowData(syn_sce_tidy_hvg_cms)$is_hvg])))
rowData(sce_sub)$is_hvg <- rownames(sce_sub) %in% hvgBatch correction
bpstart(bpparam)
temp_sce <- batchelor::multiBatchNorm(sce_sub,
batch=sce_sub$Sample,
subset.row = rownames(sce_sub)[rowData(sce_sub)[["is_hvg"]]],
normalize.all=TRUE,
BPPARAM = bpparam)
bpstop(bpparam)
stopifnot(all(unlist(merge_order) %in% unique(sce_sub$Sample)))
stopifnot(all(unique(sce_sub$Sample) %in% unlist(merge_order)))
bpstart(bpparam)
temp_sce <- batchelor::fastMNN(temp_sce, batch=temp_sce$Sample, prop.k=0.02, merge.order = merge_order,
subset.row = rownames(temp_sce)[rowData(temp_sce)[["is_hvg"]]], correct.all=TRUE,
BPPARAM = bpparam)
bpstop(bpparam)
assay(sce_sub, "reconstructed") <- assay(temp_sce, "reconstructed")
reducedDim(sce_sub, "corrected") <- reducedDim(temp_sce, "corrected")
set.seed(100)
bpstart(bpparam)
sce_sub <- runUMAP(sce_sub, dimred = "corrected", name = "UMAP_corrected",
BPPARAM = bpparam)
bpstop(bpparam)cms
set.seed(123)
bpstart(bpparam)
sce_sub <- CellMixS::cms(sce_sub, k=300, group = "Sample",
dim_red = "PCA", res_name = "unaligned",
BPPARAM = bpparam)
bpstop(bpparam)
bpstart(bpparam)
sce_sub <- CellMixS::cms(sce_sub, k=300, group = "Sample",
dim_red = "corrected", res_name = "MNN",
BPPARAM = bpparam)
bpstop(bpparam)
CellMixS::visHist(sce_sub)
ggpubr::ggarrange(ncol=1,nrow=3,
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="Sample"),
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="cms.MNN"),
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="cms.unaligned")
)
clustering
set.seed(100)
bpstart(bpparam)
graph_sub <- buildSNNGraph(sce_sub, use.dimred="corrected", k=20, BPPARAM = bpparam)
bpstop(bpparam)
clusters <- igraph::cluster_louvain(graph_sub)$membership
colData(sce_sub)[[paste0(celltype_name_pre,"_clusters")]] <- factor(clusters)
clusters_lookup[[celltype_name_pre]] <- data.frame(cell_id = colnames(sce_sub),
cluster = paste0(toupper(celltype_name_pre),"_", as.character(clusters)))
plotReducedDim(sce_sub,"UMAP_corrected",
colour_by=paste0(celltype_name_pre,"_clusters"),
text_by=paste0(celltype_name_pre,"_clusters"))+
theme(legend.position = c(1.01,0.7),
legend.background = element_rect(color="grey",fill = "white"),
legend.margin = margin(10,10,10,10), plot.margin = margin(t=10,r=250,b=0,l=10))
save
tmpfilename <- paste0("syn_v",analysis_version,"_sce_",celltype_name_pre,dplyr::if_else(remove_low_quality_samples, "_invivo",""),".rds")
saveRDS(sce_sub, file = here::here("output",tmpfilename))Subclustering - T-cells
celltype_name_pre <- "tc"HVG selection
set.seed(100)
sce_sub <- syn_sce_tidy_hvg_cms[,syn_sce_tidy_hvg_cms$kgraph_clusters %in% cluster_assignment$cluster_names_max[[celltype_name_pre]]]
assay(sce_sub, "vstresiduals") <- NULL
bpstart(bpparam)
all_gene_var <- modelGeneVar(sce_sub, block=sce_sub$Sample, BPPARAM=bpparam)
bpstop(bpparam)
hvg <- getTopHVGs(all_gene_var, fdr.threshold=0.05)
UpSetR::upset(UpSetR::fromList(list(subset=hvg, all=rownames(syn_sce_tidy_hvg_cms)[rowData(syn_sce_tidy_hvg_cms)$is_hvg])))
rowData(sce_sub)$is_hvg <- rownames(sce_sub) %in% hvgBatch correction
bpstart(bpparam)
temp_sce <- batchelor::multiBatchNorm(sce_sub,
batch=sce_sub$Sample,
subset.row = rownames(sce_sub)[rowData(sce_sub)[["is_hvg"]]],
normalize.all=TRUE,
BPPARAM = bpparam)
bpstop(bpparam)
stopifnot(all(unlist(merge_order) %in% unique(sce_sub$Sample)))
stopifnot(all(unique(sce_sub$Sample) %in% unlist(merge_order)))
bpstart(bpparam)
temp_sce <- batchelor::fastMNN(temp_sce, batch=temp_sce$Sample, prop.k=0.02, merge.order = merge_order,
subset.row = rownames(temp_sce)[rowData(temp_sce)[["is_hvg"]]], correct.all=TRUE,
BPPARAM = bpparam)
bpstop(bpparam)
assay(sce_sub, "reconstructed") <- assay(temp_sce, "reconstructed")
reducedDim(sce_sub, "corrected") <- reducedDim(temp_sce, "corrected")
set.seed(100)
bpstart(bpparam)
sce_sub <- runUMAP(sce_sub, dimred = "corrected", name = "UMAP_corrected",
BPPARAM = bpparam)
bpstop(bpparam)cms
set.seed(123)
bpstart(bpparam)
sce_sub <- CellMixS::cms(sce_sub, k=300, group = "Sample",
dim_red = "PCA", res_name = "unaligned",
BPPARAM = bpparam)
bpstop(bpparam)
bpstart(bpparam)
sce_sub <- CellMixS::cms(sce_sub, k=300, group = "Sample",
dim_red = "corrected", res_name = "MNN",
BPPARAM = bpparam)
bpstop(bpparam)
CellMixS::visHist(sce_sub)
ggpubr::ggarrange(ncol=1,nrow=3,
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="Sample"),
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="cms.MNN"),
plotReducedDim(sce_sub,"UMAP_corrected", colour_by="cms.unaligned")
)
clustering
set.seed(100)
bpstart(bpparam)
graph_sub <- buildSNNGraph(sce_sub, use.dimred="corrected", k=20, BPPARAM = bpparam)
bpstop(bpparam)
clusters <- igraph::cluster_louvain(graph_sub)$membership
colData(sce_sub)[[paste0(celltype_name_pre,"_clusters")]] <- factor(clusters)
clusters_lookup[[celltype_name_pre]] <- data.frame(cell_id = colnames(sce_sub),
cluster = paste0(toupper(celltype_name_pre),"_", as.character(clusters)))
plotReducedDim(sce_sub,"UMAP_corrected",
colour_by=paste0(celltype_name_pre,"_clusters"),
text_by=paste0(celltype_name_pre,"_clusters"))+
theme(legend.position = c(1.01,0.7),
legend.background = element_rect(color="grey",fill = "white"),
legend.margin = margin(10,10,10,10), plot.margin = margin(t=10,r=250,b=0,l=10))
save
tmpfilename <- paste0("syn_v",analysis_version,"_sce_",celltype_name_pre,dplyr::if_else(remove_low_quality_samples, "_invivo",""),".rds")
saveRDS(sce_sub, file = here::here("output",tmpfilename))Combine clustering results
tmpfilename <- paste0("syn_v",analysis_version,"_clustering_lookup",dplyr::if_else(remove_low_quality_samples, "_invivo",""),".rds")
saveRDS(clusters_lookup, here::here("output",tmpfilename))
colData(syn_sce_tidy_hvg_cms)$combined_clusters <- ""
names(clusters_lookup)[1] "sf" "ec" "mp" "tc"for(sub_name in names(clusters_lookup)){
celllabelmatch <- match(clusters_lookup[[sub_name]]$cell_id,
colnames(syn_sce_tidy_hvg_cms))
celllabelmatch <- celllabelmatch[!is.na(celllabelmatch)]
colData(syn_sce_tidy_hvg_cms)$combined_clusters[celllabelmatch] <- clusters_lookup[[sub_name]]$cluster
}
colData(syn_sce_tidy_hvg_cms)$combined_clusters[colData(syn_sce_tidy_hvg_cms)$combined_clusters == ""] <-
colData(syn_sce_tidy_hvg_cms)$kgraph_clusters[colData(syn_sce_tidy_hvg_cms)$combined_clusters == ""]tmpfilename <- paste0("syn_v",analysis_version,"_sce_hvg_cms_doublet_subcluster",dplyr::if_else(remove_low_quality_samples, "_invivo",""),".rds")
saveRDS(syn_sce_tidy_hvg_cms, file = here::here("output",tmpfilename))4th Part: Annotation
sessionInfo()R version 4.0.3 (2020-10-10)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 20.04 LTS
Matrix products: default
BLAS/LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.8.so
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=C
[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
attached base packages:
[1] parallel stats4 stats graphics grDevices utils datasets
[8] methods base
other attached packages:
[1] gdtools_0.2.3 BiocParallel_1.24.1
[3] CellID_0.1.0 SeuratObject_4.0.0
[5] Seurat_4.0.0 bluster_1.0.0
[7] tidySingleCellExperiment_1.0.0 scuttle_1.0.4
[9] igraph_1.2.6 scran_1.18.7
[11] scater_1.18.6 SingleCellExperiment_1.12.0
[13] SummarizedExperiment_1.20.0 Biobase_2.50.0
[15] GenomicRanges_1.42.0 GenomeInfoDb_1.26.7
[17] IRanges_2.24.1 S4Vectors_0.28.1
[19] BiocGenerics_0.36.1 MatrixGenerics_1.2.1
[21] matrixStats_0.58.0 stringr_1.4.0
[23] purrr_0.3.4 ggplot2_3.3.3
[25] dplyr_1.0.4 workflowr_1.6.2
loaded via a namespace (and not attached):
[1] scattermore_0.7 ggthemes_4.2.4
[3] tidyr_1.1.2 knitr_1.31
[5] irlba_2.3.3 DelayedArray_0.16.3
[7] data.table_1.13.6 rpart_4.1-15
[9] RCurl_1.98-1.2 generics_0.1.0
[11] RhpcBLASctl_0.20-137 cowplot_1.1.1
[13] RANN_2.6.1 proxy_0.4-24
[15] future_1.21.0 spatstat.data_2.0-0
[17] httpuv_1.5.5 assertthat_0.2.1
[19] viridis_0.5.1 xfun_0.21
[21] hms_1.0.0 evaluate_0.14
[23] promises_1.2.0.1 DEoptimR_1.0-8
[25] fansi_0.4.2 readxl_1.3.1
[27] DBI_1.1.1 htmlwidgets_1.5.3
[29] kSamples_1.2-9 ellipsis_0.3.1
[31] RSpectra_0.16-0 backports_1.2.1
[33] ggpubr_0.4.0 deldir_0.2-10
[35] sparseMatrixStats_1.2.1 vctrs_0.3.6
[37] here_1.0.1 TTR_0.24.2
[39] ROCR_1.0-11 abind_1.4-5
[41] CellMixS_1.6.1 batchelor_1.6.3
[43] RcppEigen_0.3.3.9.1 withr_2.4.1
[45] robustbase_0.93-7 vcd_1.4-8
[47] sctransform_0.3.2.9008 xts_0.12.1
[49] goftest_1.2-2 svglite_1.2.3.2
[51] cluster_2.1.1 lazyeval_0.2.2
[53] laeken_0.5.1 crayon_1.4.1
[55] SuppDists_1.1-9.5 edgeR_3.32.1
[57] pkgconfig_2.0.3 labeling_0.4.2
[59] nlme_3.1-152 vipor_0.4.5
[61] nnet_7.3-15 rlang_0.4.10
[63] globals_0.14.0 lifecycle_1.0.0
[65] miniUI_0.1.1.1 rsvd_1.0.3
[67] cellranger_1.1.0 rprojroot_2.0.2
[69] polyclip_1.10-0 RcppHNSW_0.3.0
[71] lmtest_0.9-38 Matrix_1.3-2
[73] carData_3.0-4 boot_1.3-27
[75] zoo_1.8-8 beeswarm_0.2.3
[77] whisker_0.4 ggridges_0.5.3
[79] png_0.1-7 viridisLite_0.3.0
[81] bitops_1.0-6 KernSmooth_2.23-18
[83] DelayedMatrixStats_1.12.3 parallelly_1.23.0
[85] rstatix_0.7.0 ggsignif_0.6.0
[87] beachmat_2.6.4 scales_1.1.1
[89] magrittr_2.0.1 plyr_1.8.6
[91] hexbin_1.28.2 ica_1.0-2
[93] zlibbioc_1.36.0 compiler_4.0.3
[95] dqrng_0.2.1 RColorBrewer_1.1-2
[97] pcaMethods_1.82.0 fitdistrplus_1.1-3
[99] cli_2.3.0 XVector_0.30.0
[101] listenv_0.8.0 patchwork_1.1.1
[103] pbapply_1.4-3 ggplot.multistats_1.0.0
[105] MASS_7.3-53.1 mgcv_1.8-34
[107] tidyselect_1.1.0 stringi_1.5.3
[109] forcats_0.5.1 highr_0.8
[111] yaml_2.2.1 BiocSingular_1.6.0
[113] askpass_1.1 locfit_1.5-9.4
[115] ggrepel_0.9.1 grid_4.0.3
[117] fastmatch_1.1-0 tools_4.0.3
[119] future.apply_1.7.0 rio_0.5.16
[121] foreign_0.8-81 git2r_0.28.0
[123] gridExtra_2.3 smoother_1.1
[125] scatterplot3d_0.3-41 farver_2.0.3
[127] Rtsne_0.15 digest_0.6.27
[129] shiny_1.6.0 Rcpp_1.0.6
[131] broom_0.7.4 car_3.0-10
[133] later_1.1.0.1 RcppAnnoy_0.0.18
[135] httr_1.4.2 colorspace_2.0-0
[137] fs_1.5.0 tensor_1.5
[139] ranger_0.12.1 reticulate_1.18
[141] umap_0.2.7.0 splines_4.0.3
[143] uwot_0.1.10 statmod_1.4.35
[145] spatstat.utils_2.0-0 sp_1.4-5
[147] systemfonts_1.0.1 plotly_4.9.3
[149] xtable_1.8-4 jsonlite_1.7.2
[151] spatstat_1.64-1 UpSetR_1.4.0
[153] destiny_3.4.0 R6_2.5.0
[155] pillar_1.4.7 htmltools_0.5.1.1
[157] mime_0.10 tictoc_1.0
[159] glue_1.4.2 fastmap_1.1.0
[161] VIM_6.1.0 BiocNeighbors_1.8.2
[163] class_7.3-18 codetools_0.2-18
[165] fgsea_1.16.0 ResidualMatrix_1.0.0
[167] lattice_0.20-41 tibble_3.0.6
[169] curl_4.3 ggbeeswarm_0.6.0
[171] leiden_0.3.7 zip_2.1.1
[173] openxlsx_4.2.3 openssl_1.4.3
[175] survival_3.2-7 limma_3.46.0
[177] rmarkdown_2.6 munsell_0.5.0
[179] e1071_1.7-4 GenomeInfoDbData_1.2.4
[181] haven_2.3.1 reshape2_1.4.4
[183] gtable_0.3.0