Cell Line L3 Analysis
1. load libraries
2. Load Seurat Object
#Load Seurat Object L3
load("../Documents/1-SS-STeps/4-Analysis_and_Robj_Marie/analyse juillet 2023/ObjetsR/L3_B.Robj")
L3 <- L3_B
L3## An object of class Seurat
## 36629 features across 6428 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 = L3) <- "cell_line"
L3[["percent.rb"]] <- PercentageFeatureSet(L3, pattern = "^RP[SL]")
VlnPlot(L3, 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 18417 by 6428## Model formula is y ~ log_umi## Get Negative Binomial regression parameters per gene## Using 2000 genes, 5000 cells## Found 177 outliers - those will be ignored in fitting/regularization step## Second step: Get residuals using fitted parameters for 18417 genes## Computing corrected count matrix for 18417 genes## Calculating gene attributes## Wall clock passed: Time difference of 24.08633 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
L3 <- SCTransform(L3, 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 18417 by 6428## Model formula is y ~ log_umi## Get Negative Binomial regression parameters per gene## Using 2000 genes, 5000 cells## Found 177 outliers - those will be ignored in fitting/regularization step## Second step: Get residuals using fitted parameters for 18417 genes## Computing corrected count matrix for 18417 genes## Calculating gene attributes## Wall clock passed: Time difference of 19.97999 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 <- L3@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
L3 <- RunPCA(L3,
features = Variables_genes_after_exclusion,
do.print = TRUE,
pcs.print = 1:5,
genes.print = 15,
npcs = 50)## PC_ 1
## Positive: B2M, VIM, S100A6, MALAT1, SQSTM1, LGALS1, CCL5, MAL, MT2A, CDKN1A
## S100A11, CCR7, H3F3B, RPL10, TMSB4X, PRKCE, ISG15, JUNB, RRM2, S100A4
## S100P, NCF2, OPTN, CD82, MYL6, WARS, PLAAT4, PMAIP1, HMSD, CRIP1
## Negative: NPM1, HSP90AB1, UBE2S, HSPE1, HSPD1, NCL, FABP5, MTDH, SERBP1, RPL17
## NME1, PTMA, SRM, HSP90AA1, HNRNPAB, CCT8, RPL23, HNRNPA2B1, RAN, ODC1
## CCT6A, JPT1, MRPL3, SET, MDH2, RPS17, TOMM40, LTB, FKBP4, CCT2
## PC_ 2
## Positive: HIST1H4C, HIST1H1B, HIST1H1E, HIST1H1C, TOP2A, HIST1H1A, HIST1H1D, MALAT1, RRM2, NLGN1
## WWOX, HIST1H3B, TUBB, PSAT1, HIST1H2AH, HIST1H3G, ATAD2, PRKDC, HIST1H3D, MKI67
## RPS16, SMC1A, EEF2, NSD2, HIST1H2AL, RPL28, PCLAF, SOS1, DUT, TMPO
## Negative: LGALS1, S100A11, TMSB4X, ACTG1, ACTB, TMSB10, PTTG1, MYL6, S100A4, HSPA8
## TNFRSF4, CDC20, S100A6, B2M, CRIP1, PRR13, CCL17, VIM, ANXA1, JPT1
## ANXA2, NQO1, MYL12A, SH3BGRL3, HTATIP2, S100A10, IL13, CLIC1, IL32, YWHAZ
## PC_ 3
## Positive: RANBP17, RPL18A, PGAP1, LINC02398, RPL13A, AC024901.1, RPL36, AREG, TMEM178B, ACSL4
## EEF2, MYO1D, NLGN1, LY75, LMNA, RPS15, EMP3, AHNAK, NCALD, REL
## RPS28, MAST4, RPS19, NFAT5, COL19A1, JUN, HSPA8, CADM1, MIR155HG, APP
## Negative: HIST1H4C, TUBA1B, H2AFZ, RRM2, GAPDH, TUBB, PCLAF, HIST1H1E, HIST1H1A, TYMS
## HIST1H1B, UBE2T, HIST1H1D, PPIA, MT-CO3, HMGB2, PKMYT1, TUBA4A, H1FX, CD74
## H2AFX, CFL1, UBE2I, SIVA1, CENPM, CENPU, UBE2C, HIST1H3B, DUT, TPI1
## PC_ 4
## Positive: FN1, AC011586.2, RPS12, ADAM19, RPL12, STC1, LINC01170, GAB2, PLPP1, FAM107B
## ATP8B4, RNF213, CHST11, FAM189A1, DENND1B, DOCK2, KCNQ5, RPS4X, CACNA1D, ARHGEF6
## RUNX1, AGBL1, RYR2, FGGY, ARL6IP5, CYTOR, MSI2, NCOA2, AC112770.1, SLC26A4
## Negative: AREG, RPS15, RPL18A, RPL36, CD70, GPX4, HMSD, RPS16, EMP3, RPS5
## RPS19, RPS28, LAYN, UBA52, PDCD5, CYTIP, RPL18, HPGDS, TXNDC17, PRMT1
## RANBP17, AC024901.1, GJB2, LTA, PHLDA2, CYB5A, LINC00276, SNRPD2, RPL28, ATP1B2
## PC_ 5
## Positive: SQSTM1, KLF6, PPP1R15A, GATA2, TXN, CXCL8, GRINA, EEF1A2, IL1A, DUSP1
## RAP1B, CCL5, ZFAS1, HSPA9, CHCHD10, GAS5, IER3, FOS, CCL3, NAMPT
## PRPS2, RPS12, RGS1, PMAIP1, FOSB, EBI3, HIST1H1E, HSPA5, RPL12, CNBP
## Negative: RPS15, CA10, RPL28, RPL36, RPL18A, RPL18, RPS19, RPS28, GPX4, RPS16
## S100A4, UBA52, RPS5, CSMD3, CFI, LINC02398, NDUFA11, CD3D, SPINT2, GMFG
## AREG, AL023574.1, SEC11C, UQCR11, OAZ1, LSM7, EVI2B, COX6C, RALYL, SNRPD2
# 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 L3$cell_line is a factor or character vector containing cell line names
data <- as.data.frame(table(L3$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 <- L3[["pca"]]@stdev / sum(L3[["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] 10# TEST-2
# get significant PCs
stdv <- L3[["pca"]]@stdev
sum.stdv <- sum(L3[["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] 10# 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
L3 <- FindNeighbors(L3,
dims = 1:min.pc,
verbose = FALSE)
# understanding resolution
L3 <- FindClusters(L3,
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: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.9231
## Number of communities: 3
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.8776
## Number of communities: 5
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.8398
## Number of communities: 5
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.8074
## Number of communities: 6
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7804
## Number of communities: 6
## Elapsed time: 0 seconds
## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7571
## 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: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7397
## 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: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7251
## 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: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7129
## 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: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.7003
## 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: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.6873
## 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: 6428
## Number of edges: 194888
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.6764
## Number of communities: 14
## Elapsed time: 0 seconds# non-linear dimensionality reduction --------------
L3 <- RunUMAP(L3,
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(L3,group.by = "cell_line",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.0.1",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.0.2",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.0.3",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.0.4",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.0.5",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.0.6",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.0.7",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.0.8",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.0.9",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.1",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
group.by = "SCT_snn_res.1.1",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3,
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
L3 <- RunAzimuth(L3, 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 220 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## 15:44:06 Read 6428 rows## 15:44:06 Processing block 1 of 1## 15:44:06 Commencing smooth kNN distance calibration using 1 thread with target n_neighbors = 20
## 15:44:06 Initializing by weighted average of neighbor coordinates using 1 thread
## 15:44:06 Commencing optimization for 67 epochs, with 128560 positive edges
## 15:44:07 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: 2.28114891052246DimPlot(L3, group.by = "predicted.celltype.l2",
reduction = "umap",
label.size = 3,
repel = T,
label = F)
DimPlot (L3, 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(L3, ref = ref)## | | | 0%[1] "Using assay SCT for query"## Pre-filtering cells with scGate...##
## ### Detected a total of 2353 pure 'Target' cells (36.61% of total)## [1] "4075 out of 6428 ( 63% ) 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): GNLY, GZMK, CCL20, MT1G, CD177, CGA, G0S2, IL17A, CCL4L2,
## XCL1, IGFL2, GZMH, TRDC, XCL2, LAIR2, IL10, ACTG2, RASD1, DIRAS3, KLRD1, NPPC,
## IL17RB, ZBED2, LGMN, CD7, CXCR6, HOPX, MS4A6A, TYROBP, TUBA3D, ADGRG1, IL2,
## PVALB, MRC1, TNFSF4, TASL, ID1, CPA5, TMIGD2, PRF1, H1-4, METTL7A, IL26, CLIC3,
## H2AZ1, CTSW, ACP5, PTGER2, GPR25, TNFSF8, TNS3, ELAPOR1, CCL3L3, POLR1F, ENC1,
## BTLA, KIR3DL2, F5, AHSP, CCR2, FAIM2, GIMAP7, CDKN2B, GIMAP4, HTRA1, CCND1,
## PLAC8, LIMS2, H1-2, CDKN2A, H1-0, WARS1, H1-3, ASCL2, DTHD1, H2BC11, GPX1,
## FCMR, H2AC6, IRAG2, H1-10, MYO7A, FASLG, SLA, CCR5, SCML1, H3C10, NAP1L2,
## HS3ST3B1, FGFBP2, IL22, KRT81, PDLIM4, UCP3, ZNF683, ECEL1, HSD11B1, ARC,
## GFPT2, H4C3, RORC, 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 2353 pure 'Target' cells (36.61% of total)## [1] "4075 out of 6428 ( 63% ) 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): GNLY, GZMK, CCL20, MT1G, CD177, CGA, G0S2, IL17A, CCL4L2,
## XCL1, IGFL2, GZMH, TRDC, XCL2, LAIR2, IL10, ACTG2, RASD1, DIRAS3, KLRD1, NPPC,
## IL17RB, ZBED2, LGMN, CD7, CXCR6, HOPX, MS4A6A, TYROBP, TUBA3D, ADGRG1, IL2,
## PVALB, MRC1, TNFSF4, TASL, ID1, CPA5, TMIGD2, PRF1, H1-4, METTL7A, IL26, CLIC3,
## H2AZ1, CTSW, ACP5, PTGER2, GPR25, TNFSF8, TNS3, ELAPOR1, CCL3L3, POLR1F, ENC1,
## BTLA, KIR3DL2, F5, AHSP, CCR2, FAIM2, GIMAP7, CDKN2B, GIMAP4, HTRA1, CCND1,
## PLAC8, LIMS2, H1-2, CDKN2A, H1-0, WARS1, H1-3, ASCL2, DTHD1, H2BC11, GPX1,
## FCMR, H2AC6, IRAG2, H1-10, MYO7A, FASLG, SLA, CCR5, SCML1, H3C10, NAP1L2,
## HS3ST3B1, FGFBP2, IL22, KRT81, PDLIM4, UCP3, ZNF683, ECEL1, HSD11B1, ARC,
## GFPT2, H4C3, RORC, GSTM2## | |======================================================================| 100%## Creating slots functional.cluster and functional.cluster.conf in query objectUMAPPlot(L3, 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(L3))
#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$pruned.labels)
#add the annotations to the Seurat object metadata
L3@meta.data$monaco.main <- monaco.main$pruned.labels
L3@meta.data$monaco.fine <- monaco.fine$pruned.labels
#
L3@meta.data$hpca.main <- hpca.main$pruned.labels
L3@meta.data$hpca.fine <- hpca.fine$pruned.labels
#
L3@meta.data$dice.main <- dice.main$pruned.labels
L3@meta.data$dice.fine <- dice.fine$pruned.labels
#
L3@meta.data$bpe.main <- bpe.main$pruned.labels
L3@meta.data$bpe.fine <- bpe.fine$pruned.labels
# Monaco Annotations
DimPlot(L3, group.by = "monaco.main",
reduction = "umap",
label.size = 3,
repel = T,
label = F)
DimPlot(L3, group.by = "monaco.main",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)## Warning: ggrepel: 3 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps
DimPlot(L3, group.by = "monaco.fine",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)## Warning: ggrepel: 6 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps
# HPCA Annotations
DimPlot(L3, 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(L3, group.by = "hpca.fine",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
# DICE Annotations
DimPlot(L3, group.by = "dice.main",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3, group.by = "dice.fine",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
# BPE Annotations
DimPlot(L3, group.by = "bpe.main",
reduction = "umap",
label.size = 3,
repel = T,
label = T, label.box = T)
DimPlot(L3, 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