Cell Line L7 Analysis
1. load libraries
2. Load Seurat Object
#Load Seurat Object L7
load("../Documents/1-SS-STeps/4-Analysis_and_Robj_Marie/analyse juillet 2023/ObjetsR/L7.Robj")
L7## An object of class Seurat
## 36629 features across 5331 samples within 2 assays
## Active assay: RNA (36601 features, 0 variable features)
## 2 layers present: counts, data
## 1 other assay present: ADT3. QC
# Set identity classes to an existing column in meta data
Idents(object = L7) <- "cell_line"
L7[["percent.rb"]] <- PercentageFeatureSet(L7, pattern = "^RP[SL]")
VlnPlot(L7, features = c("nFeature_RNA",
"nCount_RNA",
"percent.mito",
"percent.rb"),
ncol = 4, pt.size = 0.1) &
theme(plot.title = element_text(size=10))
## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
##FeatureScatter is typically used to visualize feature-feature
relationships ##for anything calculated by the object, ##i.e. columns in
object metadata, PC scores etc.
## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x' ## Assign
Cell-Cycle Scores
## Running SCTransform on assay: RNA## vst.flavor='v2' set. Using model with fixed slope and excluding poisson genes.## Calculating cell attributes from input UMI matrix: log_umi## Variance stabilizing transformation of count matrix of size 18407 by 5331## Model formula is y ~ log_umi## Get Negative Binomial regression parameters per gene## Using 2000 genes, 5000 cells## Found 112 outliers - those will be ignored in fitting/regularization step## Second step: Get residuals using fitted parameters for 18407 genes## Computing corrected count matrix for 18407 genes## Calculating gene attributes## Wall clock passed: Time difference of 22.08922 secs## Determine variable features## Place corrected count matrix in counts slot## Set default assay to SCT## Warning: The following features are not present in the object: MLF1IP, not
## searching for symbol synonyms## Warning: The following features are not present in the object: FAM64A, HN1, not
## searching for symbol synonyms4. Normalize data
# Apply SCTransform
L7 <- SCTransform(L7, vars.to.regress = c("percent.rb","percent.mito", "CC.Difference"),
do.scale=TRUE,
do.center=TRUE,
verbose = TRUE)## Running SCTransform on assay: RNA## vst.flavor='v2' set. Using model with fixed slope and excluding poisson genes.## Calculating cell attributes from input UMI matrix: log_umi## Variance stabilizing transformation of count matrix of size 18407 by 5331## Model formula is y ~ log_umi## Get Negative Binomial regression parameters per gene## Using 2000 genes, 5000 cells## Found 112 outliers - those will be ignored in fitting/regularization step## Second step: Get residuals using fitted parameters for 18407 genes## Computing corrected count matrix for 18407 genes## Calculating gene attributes## Wall clock passed: Time difference of 17.75149 secs## Determine variable features## Regressing out percent.rb, percent.mito, CC.Difference## Centering and scaling data matrix## Place corrected count matrix in counts slot## Set default assay to SCT5. Perform PCA
Variables_genes <- L7@assays$SCT@var.features
# Exclude genes starting with "HLA-" AND "Xist" AND "TRBV, TRAV"
Variables_genes_after_exclusion <- Variables_genes[!grepl("^HLA-|^XIST|^TRBV|^TRAV", Variables_genes)]
# These are now standard steps in the Seurat workflow for visualization and clustering
L7 <- RunPCA(L7,
features = Variables_genes_after_exclusion,
do.print = TRUE,
pcs.print = 1:5,
genes.print = 15,
npcs = 50)## PC_ 1
## Positive: HSP90AB1, NPM1, HSPD1, HSPE1, NCL, HSP90AA1, HSPA9, SRM, HMGA1, PPP1R14B
## CHCHD10, EIF4G1, NME1, NME2, CANX, RPL23, C1QBP, FCER2, ATP5F1B, PABPC1
## PPBP, MIR155HG, LRPPRC, CCT5, TOMM40, RPL35A, PPP2R2B, HNRNPAB, MTDH, RPL22L1
## Negative: S100A4, IL32, S100A6, SH3BGRL3, S100A11, LAPTM5, ARHGDIB, TMSB4X, FXYD5, B2M
## CRIP1, CD52, IFITM2, LGALS1, LGALS3, GMFG, CORO1A, COTL1, LCK, EMP3
## RAC2, LSP1, S1PR4, ACTB, CNN2, TMSB10, ITM2B, CD99, IFITM1, EVL
## PC_ 2
## Positive: PDE4D, DENND4A, MBNL1, RABGAP1L, NCALD, GRAMD1B, WWOX, MACROD2, ELMO1, NEAT1
## SPOCK1, RAD51B, CDKAL1, ARHGAP15, MSC-AS1, PDE7A, LRBA, TRIO, RPS6KA5, INPP4B
## PVT1, AHR, EXT1, ATP8B4, PDE3B, MACF1, SOS1, DOCK2, CASK, FTX
## Negative: TUBA1B, UBE2S, JPT1, HSPE1, PRELID1, CYCS, PTMA, H2AFZ, ATP5MC3, HSP90AA1
## NDUFAB1, HMGB1, RPL35, TUBB4B, TOMM40, RANBP1, EIF5A, PPIB, MRPL12, CYC1
## HNRNPAB, TXN, NOP16, NME1, CDC20, RPL7, ATP5F1B, PPP1R14B, DYNLL1, EIF5
## PC_ 3
## Positive: IQCG, CSF2, TNFSF9, IL2RG, SRGN, IFNG, IL32, PHLDA1, CD2, PGK1
## CCL3, BNIP3, PKM, PGAM1, LINC01480, CCL1, CD40LG, PLIN2, CCL4, FAM162A
## RGCC, GADD45G, ZFP36L1, KLF6, SLC16A3, C4orf3, CD82, DUSP4, LDHA, SH3BGRL3
## Negative: RRM2, TYMS, STMN1, HIST1H4C, SMC4, TUBB, TUBA1B, HIST1H1E, ATAD2, MKI67
## H2AFZ, HIST1H1C, HMGB2, LMNB1, HIST1H1D, PCLAF, PKMYT1, NUSAP1, TOP2A, TK1
## PCNA, KIFC1, H2AFX, DHFR, DUT, HIST2H2AC, ESCO2, CDCA2, ASF1B, ZWINT
## PC_ 4
## Positive: CCR7, CD74, KRT7, DNAJC5B, CYP1B1, EBI3, MT2A, AC099552.1, CD44, AC097518.2
## BIRC3, PRDX1, AC011990.1, NFKB2, BATF3, KCNMA1, ITPR1, IGFBP4, LINC02341, AC114977.1
## BLK, AKR1A1, CSF2, CBR3, NFKBIA, CCL5, LTA, CCL1, CD82, TNFSF11
## Negative: BNIP3, FAM162A, LDHA, PLIN2, PGK1, ENO1, GPI, GAPDH, UBALD2, TPI1
## BNIP3L, HILPDA, MIF, PTPRR, PFKFB4, SLC2A3, MXI1, CYTIP, INSIG2, ALDOA
## AK4, KIF2A, NMU, PGAM1, RIMKLA, PLAAT3, SLC16A3, C4orf3, PHF19, P4HA1
## PC_ 5
## Positive: CDC20, MRPS12, RNF213, UBE2S, YWHAZ, CCNB1, DANCR, EFHD2, S100A11, CAPN2
## SFPQ, NOP16, IQGAP2, SLC9A3R1, BAK1, ITGA4, CELF2, NMU, LMNA, DYNLL1
## SYNCRIP, TENM3, SH3BP1, FUT7, PTTG1, PLK1, CCND2, ARL6IP1, HSPH1, INPP4A
## Negative: CD74, HIST1H4C, HIST1H1E, LDHA, ARID5A, FAM162A, HIST1H1C, GAPDH, BNIP3, PKM
## CCR7, RRM2, TUBB, CD70, HILPDA, MT2A, PGK1, MIF, TRAF1, BNIP3L
## EEF1A1, CCL5, ENO1, KIF2A, DUT, PMAIP1, ALDOC, TYMS, PFKFB4, STAT1
# Perform
PCA TEST
library(ggplot2)
library(RColorBrewer)
# Assuming you have 10 different cell lines, generating a color palette with 10 colors
cell_line_colors <- brewer.pal(10, "Set3")
# Assuming L7$cell_line is a factor or character vector containing cell line names
data <- as.data.frame(table(L7$cell_line))
colnames(data) <- c("cell_line", "nUMI") # Change column name to nUMI
ncells <- ggplot(data, aes(x = cell_line, y = nUMI, fill = cell_line)) +
geom_col() +
theme_classic() +
geom_text(aes(label = nUMI),
position = position_dodge(width = 0.9),
vjust = -0.25) +
scale_fill_manual(values = cell_line_colors) +
theme(axis.text.x = element_text(angle = 45, hjust = 1),
plot.title = element_text(hjust = 0.5)) + # Adjust the title position
ggtitle("Filtered cells per sample") +
xlab("Cell lines") + # Adjust x-axis label
ylab("Frequency") # Adjust y-axis label
print(ncells)
# TEST-1
# given that the output of RunPCA is "pca"
# replace "so" by the name of your seurat object
pct <- L7[["pca"]]@stdev / sum(L7[["pca"]]@stdev) * 100
cumu <- cumsum(pct) # Calculate cumulative percents for each PC
# Determine the difference between variation of PC and subsequent PC
co2 <- sort(which((pct[-length(pct)] - pct[-1]) > 0.1), decreasing = T)[1] + 1
# last point where change of % of variation is more than 0.1%. -> co2
co2## [1] 16# TEST-2
# get significant PCs
stdv <- L7[["pca"]]@stdev
sum.stdv <- sum(L7[["pca"]]@stdev)
percent.stdv <- (stdv / sum.stdv) * 100
cumulative <- cumsum(percent.stdv)
co1 <- which(cumulative > 90 & percent.stdv < 5)[1]
co2 <- sort(which((percent.stdv[1:length(percent.stdv) - 1] -
percent.stdv[2:length(percent.stdv)]) > 0.1),
decreasing = T)[1] + 1
min.pc <- min(co1, co2)
min.pc## [1] 16# Create a dataframe with values
plot_df <- data.frame(pct = percent.stdv,
cumu = cumulative,
rank = 1:length(percent.stdv))
# Elbow plot to visualize
ggplot(plot_df, aes(cumulative, percent.stdv, label = rank, color = rank > min.pc)) +
geom_text() +
geom_vline(xintercept = 90, color = "grey") +
geom_hline(yintercept = min(percent.stdv[percent.stdv > 5]), color = "grey") +
theme_bw()
6. Clustering
L7 <- FindNeighbors(L7,
dims = 1:min.pc,
verbose = FALSE)
# understanding resolution
L7 <- FindClusters(L7,
resolution = c(0.1, 0.2, 0.3, 0.4, 0.5, 0.6,
0.7,0.8, 0.9, 1, 1.1, 1.2))## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.9247
## Number of communities: 4
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.8913
## Number of communities: 7
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.8692
## Number of communities: 8
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.8478
## Number of communities: 9
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.8274
## Number of communities: 10
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.8137
## Number of communities: 10
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7992
## Number of communities: 10
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7850
## Number of communities: 11
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7726
## Number of communities: 12
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7612
## Number of communities: 14
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7510
## Number of communities: 13
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 5331
## Number of edges: 182548
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7426
## Number of communities: 15
## Elapsed time: 0 seconds# non-linear dimensionality reduction --------------
L7 <- RunUMAP(L7,
dims = 1:min.pc,
verbose = FALSE)## Warning: The default method for RunUMAP has changed from calling Python UMAP via reticulate to the R-native UWOT using the cosine metric
## To use Python UMAP via reticulate, set umap.method to 'umap-learn' and metric to 'correlation'
## This message will be shown once per session# note that you can set `label = TRUE` or use the Label Clusters function to help label
# individual clusters
DimPlot(L7,group.by = "cell_line",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.0.1",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.0.2",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.0.3",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.0.4",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.0.5",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.0.6",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.0.7",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.0.8",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.0.9",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.1",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.1.1",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7,
group.by = "SCT_snn_res.1.2",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
7. Azimuth Annotation
## Warning: The following packages are already installed and will not be
## reinstalled: pbmcref# The RunAzimuth function can take a Seurat object as input
L7 <- RunAzimuth(L7, reference = "pbmcref")## Warning: Overwriting miscellanous data for model## Warning: Adding a dimensional reduction (refUMAP) without the associated assay
## being present
## Warning: Adding a dimensional reduction (refUMAP) without the associated assay
## being present## detected inputs from HUMAN with id type Gene.name## reference rownames detected HUMAN with id type Gene.name## Normalizing query using reference SCT model## Warning: 113 features of the features specified were not present in both the reference query assays.
## Continuing with remaining 4887 features.## Projecting cell embeddings## Finding query neighbors## Finding neighborhoods## Finding anchors## Found 210 anchors## Finding integration vectors## Finding integration vector weights## Predicting cell labels
## Predicting cell labels## Warning: Feature names cannot have underscores ('_'), replacing with dashes
## ('-')## Predicting cell labels## Warning: Feature names cannot have underscores ('_'), replacing with dashes
## ('-')##
## Integrating dataset 2 with reference dataset## Finding integration vectors## Integrating data## Warning: Keys should be one or more alphanumeric characters followed by an
## underscore, setting key from integrated_dr_ to integrateddr_## Computing nearest neighbors## Running UMAP projection## Warning in RunUMAP.default(object = neighborlist, reduction.model =
## reduction.model, : Number of neighbors between query and reference is not equal
## to the number of neighbors within reference## 16:39:16 Read 5331 rows## 16:39:16 Processing block 1 of 1## 16:39:16 Commencing smooth kNN distance calibration using 1 thread with target n_neighbors = 20
## 16:39:16 Initializing by weighted average of neighbor coordinates using 1 thread
## 16:39:16 Commencing optimization for 67 epochs, with 106620 positive edges
## 16:39:16 Finished## Warning: No assay specified, setting assay as RNA by default.## Projecting reference PCA onto query
## Finding integration vector weights
## Projecting back the query cells into original PCA space
## Finding integration vector weights
## Computing scores:
## Finding neighbors of original query cells
## Finding neighbors of transformed query cells
## Computing query SNN
## Determining bandwidth and computing transition probabilities
## Total elapsed time: 1.81497502326965DimPlot(L7, group.by = "predicted.celltype.l2",
reduction = "umap",
label.size = 3,
repel = T,
label = F)
DimPlot (L7, group.by = "predicted.celltype.l2",
reduction = "umap",
label.size = 3,
repel = T,
label = T)
8. Cell type annotation using ProjectTils
#Load reference atlas and query data
ref <- readRDS(file = "CD4T_human_ref_v1.rds")
#Run Projection algorithm
query.projected <- Run.ProjecTILs(L7, ref = ref)## | | | 0%[1] "Using assay SCT for query"## Pre-filtering cells with scGate...##
## ### Detected a total of 1406 pure 'Target' cells (26.37% of total)## [1] "3925 out of 5331 ( 74% ) non-pure cells removed. Use filter.cells=FALSE to avoid pre-filtering"
## [1] "Aligning query to reference map for batch-correction..."## Warning: Layer counts isn't present in the assay object[[assay]]; returning
## NULL## Preparing PCA embeddings for objects...## Warning: Number of dimensions changing from 50 to 20##
## Projecting corrected query onto Reference PCA space
##
## Projecting corrected query onto Reference UMAP space## Warning: Not all features provided are in this Assay object, removing the
## following feature(s): CXCL13, GZMK, MT1G, CD177, CCL22, G0S2, IGFL2, GZMH,
## TRDC, CH25H, FOXP3, IL10, IL1RL1, IL1R2, ACTG2, KLRB1, DIRAS3, KLRD1, THBS1,
## NPPC, IL17RB, COCH, FXYD2, CD200, HOPX, MS4A6A, LAYN, TYROBP, TUBA3D, IL2,
## FCER1G, PVALB, EOMES, MRC1, TASL, H1-4, METTL7A, ZNF80, LRRC32, H2AZ1, ACP5,
## GPR25, ELAPOR1, CCL3L3, POLR1F, CXCR5, MMP9, FCRL3, ADTRP, IL1R1, PECAM1, AHSP,
## CCR2, GIMAP4, HTRA1, LIMS2, H1-2, H1-0, FLT1, WARS1, H1-3, CAMK1, ASCL2, DTHD1,
## H2BC11, NELL2, GPX1, STAC, H2AC6, IRAG2, H1-10, SCML1, PTGIR, CLEC7A, DAPK2,
## PON2, H3C10, HS3ST3B1, FBLN7, TRIM2, FGFBP2, IL22, KRT81, SLC28A3, IL13, ECEL1,
## HES1, HSD11B1, CPE, ARC, NLRP3, NT5E, H4C3, HS3ST1, GSTM2## | |======================================================================| 100%## Creating slots functional.cluster and functional.cluster.conf in query object
#Plot the predicted composition of the query in terms of reference T cell subtypes
plot.statepred.composition(ref, query.projected, metric = "Percent")
## | | | 0%[1] "Using assay SCT for query"## Pre-filtering cells with scGate...
##
## ### Detected a total of 1406 pure 'Target' cells (26.37% of total)## [1] "3925 out of 5331 ( 74% ) non-pure cells removed. Use filter.cells=FALSE to avoid pre-filtering"
## [1] "Aligning query to reference map for batch-correction..."## Warning: Layer counts isn't present in the assay object[[assay]]; returning
## NULL## Preparing PCA embeddings for objects...## Warning: Number of dimensions changing from 50 to 20##
## Projecting corrected query onto Reference PCA space
##
## Projecting corrected query onto Reference UMAP space## Warning: Not all features provided are in this Assay object, removing the
## following feature(s): CXCL13, GZMK, MT1G, CD177, CCL22, G0S2, IGFL2, GZMH,
## TRDC, CH25H, FOXP3, IL10, IL1RL1, IL1R2, ACTG2, KLRB1, DIRAS3, KLRD1, THBS1,
## NPPC, IL17RB, COCH, FXYD2, CD200, HOPX, MS4A6A, LAYN, TYROBP, TUBA3D, IL2,
## FCER1G, PVALB, EOMES, MRC1, TASL, H1-4, METTL7A, ZNF80, LRRC32, H2AZ1, ACP5,
## GPR25, ELAPOR1, CCL3L3, POLR1F, CXCR5, MMP9, FCRL3, ADTRP, IL1R1, PECAM1, AHSP,
## CCR2, GIMAP4, HTRA1, LIMS2, H1-2, H1-0, FLT1, WARS1, H1-3, CAMK1, ASCL2, DTHD1,
## H2BC11, NELL2, GPX1, STAC, H2AC6, IRAG2, H1-10, SCML1, PTGIR, CLEC7A, DAPK2,
## PON2, H3C10, HS3ST3B1, FBLN7, TRIM2, FGFBP2, IL22, KRT81, SLC28A3, IL13, ECEL1,
## HES1, HSD11B1, CPE, ARC, NLRP3, NT5E, H4C3, HS3ST1, GSTM2## | |======================================================================| 100%## Creating slots functional.cluster and functional.cluster.conf in query objectUMAPPlot(L7, group.by = "functional.cluster",
reduction = "umap",
label.size = 3,
repel = T,
label = T)
9.Cell type annotation using SingleR
## see ?celldex and browseVignettes('celldex') for documentation## loading from cache## see ?celldex and browseVignettes('celldex') for documentation## loading from cache## see ?celldex and browseVignettes('celldex') for documentation
## loading from cache## see ?celldex and browseVignettes('celldex') for documentation## loading from cache## see ?celldex and browseVignettes('celldex') for documentation
## loading from cache## see ?celldex and browseVignettes('celldex') for documentation## loading from cache## see ?celldex and browseVignettes('celldex') for documentation
## loading from cache## see ?celldex and browseVignettes('celldex') for documentation## loading from cache#convert our Seurat object to single cell experiment (SCE)
sce <- as.SingleCellExperiment(DietSeurat(L7))
#using SingleR
monaco.main <- SingleR(test = sce,assay.type.test = 1,ref = monaco.ref,labels = monaco.ref$label.main)
monaco.fine <- SingleR(test = sce,assay.type.test = 1,ref = monaco.ref,labels = monaco.ref$label.fine)
hpca.main <- SingleR(test = sce,assay.type.test = 1,ref = hpca.ref,labels = hpca.ref$label.main)
hpca.fine <- SingleR(test = sce,assay.type.test = 1,ref = hpca.ref,labels = hpca.ref$label.fine)
dice.main <- SingleR(test = sce,assay.type.test = 1,ref = dice.ref,labels = dice.ref$label.main)
dice.fine <- SingleR(test = sce,assay.type.test = 1,ref = dice.ref,labels = dice.ref$label.fine)
bpe.main <- SingleR(test = sce,assay.type.test = 1,ref = bpe.ref,labels = bpe.ref$label.main)
bpe.fine <- SingleR(test = sce,assay.type.test = 1,ref = bpe.ref,labels = bpe.ref$label.fine)
#summary of general cell type annotations
#table(monaco.main$pruned.labels)
#table(hpca.main$pruned.labels)
#table(dice.main$pruned.labels)
#table(bpe.main$pruned.labels)
#The finer cell types annotations are you after, the harder they are to get reliably.
#This is where comparing many databases, as well as using individual markers from literature,
#would all be very valuable.
#table(monaco.fine$pruned.labels)
#table(hpca.fine$pruned.labels)
#table(dice.fine$pruned.labels)
#table(bpe.fine$pruned.labels)
#add the annotations to the Seurat object metadata
L7@meta.data$monaco.main <- monaco.main$pruned.labels
L7@meta.data$monaco.fine <- monaco.fine$pruned.labels
#
L7@meta.data$hpca.main <- hpca.main$pruned.labels
L7@meta.data$hpca.fine <- hpca.fine$pruned.labels
#
L7@meta.data$dice.main <- dice.main$pruned.labels
L7@meta.data$dice.fine <- dice.fine$pruned.labels
#
L7@meta.data$bpe.main <- bpe.main$pruned.labels
L7@meta.data$bpe.fine <- bpe.fine$pruned.labels
# Monaco Annotations
DimPlot(L7, group.by = "monaco.main",
reduction = "umap",
label.size = 3,
repel = T,
label = F)
DimPlot(L7, group.by = "monaco.main",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7, group.by = "monaco.fine",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)## Warning: ggrepel: 1 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps
# HPCA Annotations
DimPlot(L7, group.by = "hpca.main",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)## Warning: ggrepel: 2 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps
DimPlot(L7, group.by = "hpca.fine",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
# DICE Annotations
DimPlot(L7, group.by = "dice.main",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7, group.by = "dice.fine",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
# BPE Annotations
DimPlot(L7, group.by = "bpe.main",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L7, group.by = "bpe.fine",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
10. clusTree
## Loading required package: ggraph##
## Attaching package: 'ggraph'## The following object is masked from 'package:sp':
##
## geometry