-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathSeurat_function.R
1458 lines (1269 loc) · 59.5 KB
/
Seurat_function.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
##------20190724 ZJH
##------20201205 ZJH update functions
##------20210103 ZJH update functions: SelectCells,
##------of neuron related plots and find DEGs between each celltypes
library(Seurat)
library(ggplot2)
library(RColorBrewer)
library(Seurat)
library(openxlsx)
library("grid")
library("ggplotify")
library(magrittr)
library(dplyr)
library(ggsignif)
library("ggplot2")
library(cowplot)
library(openxlsx)
##-------------------------------------------------1. create_seurat_filter_cell---------------------------------------------------------------##
Create_seurat_filter_cell <- function(exp_count,pr_name,min_cell,Or_ident,mt_pattern,rp_pattern){
# CreateSeuratObject : filter gene and cell by min.cells(Include features detected in at least this many cells) and min.features(Include cells where at least this many features are detected.)
exp_count.seurat <- CreateSeuratObject(exp_count,
project = pr_name,
min.cells = min_cell)
[email protected]$orig.ident <- as.factor(Or_ident)
[email protected] <- exp_count.seurat$orig.ident
## Check the mt gene and rp gene
exp_count.seurat[["percent.mt"]] <- PercentageFeatureSet(object = exp_count.seurat, pattern = mt_pattern)
exp_count.seurat[["percent.rp"]] <- PercentageFeatureSet(object = exp_count.seurat, pattern = rp_pattern)
print(levels([email protected]))
print(table(exp_count.seurat$orig.ident))
print(head(x = [email protected], 5))
print(dim([email protected]))
return(exp_count.seurat)
}
##-------------------------------------------------------2.QC------------------------------------------------------------------------------------##
QC <- function(exp_count.seurat,name_pre){
folder_name="1.QC"
if (file.exists(folder_name)){
print("1.QC existed.")
}
else{
dir.create(folder_name)
}
file_name=paste(name_pre,"Violinplot.pdf",sep="_")
pdf(file = file_name, height = 10, width = 20)
print(VlnPlot(object = exp_count.seurat, features = c("nFeature_RNA", "nCount_RNA","percent.mt","percent.rp", "S.Score", "G2M.Score"),
ncol = 2,pt.size=0, group.by = "orig.ident"))
plot1=FeatureScatter(object = exp_count.seurat, feature1 = "nCount_RNA", feature2 = "nFeature_RNA")
plot2=FeatureScatter(object = exp_count.seurat, feature1 = "nCount_RNA", feature2 = "percent.mt")
plot3=FeatureScatter(object = exp_count.seurat, feature1 = "nCount_RNA", feature2 = "percent.rp")
plot4=FeatureScatter(object = exp_count.seurat, feature1 = "nCount_RNA", feature2 = "S.Score")
plot5=FeatureScatter(object = exp_count.seurat, feature1 = "nCount_RNA", feature2 = "G2M.Score")
print(CombinePlots(plots=list(plot1,plot2,plot3,plot4, plot5),ncol=3))
dev.off()
#copy files to 1.QC
result_files <- c(file_name)
file.copy(result_files, folder_name ,overwrite = T)
file.remove(result_files)
print("Based on the Violin plot, you can filter the cell by the nFeature_RNA ,nCount_RNA and/or percent.mt")
}
##-------------------------------------------------3.Subset_data_QC------------------------------------------------------------##
Subset_data_QC <- function(exp_count.seurat, min_gene, min_mt_percent, min_rp_percent, min_ncountRNA, max_gene, max_ncountRNA){
exp_count.seurat <- subset(x = exp_count.seurat, subset = nFeature_RNA > min_gene
& percent.mt < min_mt_percent & percent.rp < min_rp_percent & nCount_RNA > min_ncountRNA &
nFeature_RNA < max_gene & nCount_RNA < max_ncountRNA)
# QC
qc_pdf_name <- paste("filtered_by_gene",min_gene,"UMI",min_ncountRNA,"percent.mt",min_mt_percent,"percent.rp",min_rp_percent,sep="_")
QC(exp_count.seurat,qc_pdf_name)
dim(exp_count.seurat)
print(levels(exp_count.seurat$orig.ident))
return(exp_count.seurat)
}
##-------------------------------------------------4.Select_pc-----------------------------------------------------------##
Select_pc <- function(exp_count.seurat,condition,npc_plot){
folder_name="2.RunPCA"
if (file.exists(folder_name)){
print("2.RunPCA existed.")
}
else{
dir.create(folder_name)
}
# Normalize Find VariableGene and Scale Data
exp_count.seurat <- SCTransform(exp_count.seurat,vars.to.regress = c("percent.rp","percent.mt"))
# RunPCA
exp_count.seurat <- RunPCA(exp_count.seurat,verbose = FALSE, npcs = npc_plot)
# Select PCs:1:30pcs
selected_pcs_name <- paste0(condition,"_Selected_PCs.pdf")
pdf(file = selected_pcs_name,height = 8, width = 8)
print(ElbowPlot(object = exp_count.seurat, ndims = npc_plot, reduction = "pca"))
dev.off()
# Plot
pca_plot_name <- paste0(condition, "_PCA_plot",".pdf")
pdf(pca_plot_name,8,8)
print(DimPlot(exp_count.seurat,reduction="pca",label = TRUE,pt.size=1.1,group.by="ident",shape.by="orig.ident"))
dev.off()
#Copy files to 2.RunPCA
result_files <- c(selected_pcs_name,pca_plot_name)
file.copy(result_files, folder_name ,overwrite = T)#拷贝文件
file.remove(result_files) #移除拷贝完的文件
return(exp_count.seurat)
}
##-------------------------------------------------5.Plot_cluster-------------------------------------------------##
# RunUMAP and RunTSNE and Then find cluster
Plot_cluster <- function(exp_count.seurat,condition,npc_used,resolution_number, k_param){
folder_name="3.PlotCluster"
if (file.exists(folder_name)){
print("3.PlotCluster existed.")
}
else{
dir.create(folder_name)
}
# RunTSNE and RunUMAP
exp_count.seurat <- RunTSNE(exp_count.seurat,dims = 1:npc_used,verbose = FALSE,check_duplicates = FALSE)
exp_count.seurat <- RunUMAP(exp_count.seurat, dims = 1:npc_used,verbose = FALSE)
# FinderNeighbors
exp_count.seurat <- FindNeighbors(exp_count.seurat, dims = 1:npc_used, verbose = FALSE, k.param = k_param)
exp_count.seurat <- FindClusters(exp_count.seurat, verbose = FALSE, resolution = resolution_number)
# RenameIdents from zero to one
levels_define <- as.numeric(levels(exp_count.seurat))
new.cluster.ids <- levels_define + 1
names(new.cluster.ids) <- levels(exp_count.seurat)
exp_count.seurat <- RenameIdents(exp_count.seurat, new.cluster.ids)
combined_cluster_plotname <- paste0(condition, "_combined_cluster_resolution_",resolution_number,".pdf")
pdf(combined_cluster_plotname,7,7)
print(DimPlot(exp_count.seurat,reduction="umap",label = TRUE, pt.size = 1.2,
label.size = 4.5, repel=TRUE, group.by="ident"))
print(DimPlot(exp_count.seurat,reduction="tsne", label = TRUE, pt.size = 1.2,
label.size = 4.5, repel=TRUE, group.by="ident"))
dev.off()
split_clustering_plot_name <- paste0(condition, "_split_clustering_resolution_", resolution_number, ".pdf")
pdf(split_clustering_plot_name,14,7)
print(DimPlot(exp_count.seurat,reduction="umap",label = TRUE, pt.size = 1.2,
label.size = 4.5, group.by="ident", repel=TRUE, split.by="orig.ident"))
print(DimPlot(exp_count.seurat,reduction="tsne", label = TRUE, pt.size = 1.2,
label.size = 4.5, group.by="ident", repel=TRUE, split.by="orig.ident"))
dev.off()
# Copy files to 2.Cluster
file.copy(combined_cluster_plotname, folder_name ,overwrite = T)
file.remove(combined_cluster_plotname)
file.copy(split_clustering_plot_name, folder_name ,overwrite = T)
file.remove(split_clustering_plot_name)
return(exp_count.seurat)
}
##-------------------------------------------------6.DEGs between clusters---------------------------------------------------##
FindmarkerForCluster <- function(exp_count.seurat,condition){
folder_name="4.DEGsBetweenCluster"
if (file.exists(folder_name)){
print("3.DEGsBetweenCluster file existed.")
}
else{
dir.create(folder_name)
}
exp_count.markers <- FindAllMarkers(object = exp_count.seurat, only.pos = TRUE, min.pct = 0.1, logfc.threshold = 0.1)
index <- exp_count.markers$p_val_adj < 0.05
exp_count.markers <- exp_count.markers[index,]
save_name <- paste0(condition,"_DEGsbetweenClusters.csv")
write.csv(exp_count.markers,save_name)
file.copy(save_name, folder_name,overwrite = T)
file.remove(save_name)
return(exp_count.markers)
}
##----------------------------------------------------------7. Topmarker----------------------------------------------------##
TopMarkersInCluster <- function(exp_count.markers,condition,top_num){
library(dplyr)
folder_name="4.DEGsBetweenCluster"
if (file.exists(folder_name)){
print("4.DEGsBetweenCluster file existed.")
}
else{
dir.create(folder_name)
}
#将readsCountSM.markers传给group_by，group_by按cluster 排序，再将结果传给top_n，top_n按avg_logFC排序，显示每个类中的前两个
top_marker <- exp_count.markers %>% group_by(cluster) %>% top_n(n = top_num, wt = avg_log2FC)
file_name=paste(condition,"_top_marker",top_num,".csv",sep="")
write.csv(top_marker,file =file_name)
file.copy(file_name, folder_name,overwrite = T)
return(top_marker)
}
##------------------------------------------------- 8.Plot Feature ------------------------------------------##
Plot_Features <- function(exp_count.seurat,Features_used,condition,height,width){
folder_name="5.Plot_Features"
if (file.exists(folder_name)){
print("5.Plot_Features file existed")
}
else{
dir.create(folder_name)
}
# VlnPlot
#name1 <- paste("VlnPlot_",condition,".pdf",sep="")
#pdf(file = name1,height = height, width = width)
#print(VlnPlot(object = exp_count.seurat, features = Features_used))
#dev.off()
#file.copy(name1, folder_name,overwrite = T)#拷贝文件
#file.remove(name1)
# FeaturePlot
#name2 <- paste("FeaturePlot_",condition,".pdf",sep="")
#pdf(file = name2,height = height, width = width)
#print(FeaturePlot(object = exp_count.seurat, features = Features_used,cols = c("lightgrey", "red"),min.cutoff = 0, max.cutoff = 1,order=T,reduction="tsne"))
#dev.off()
#file.copy(name2, folder_name,overwrite = T)#拷贝文件
#file.remove(name2)
#HeatMap
name3 <- paste("HeatMap_",condition,".pdf",sep="")
pdf(file = name3,height = height, width = width)
print(DoHeatmap(exp_count.seurat, features = Features_used) + NoLegend())
dev.off()
file.copy(name3, folder_name,overwrite = T)#拷贝文件
file.remove(name3)
}
##-----------------------------------------------------------------9. celltype identification----------------------------------------##
# CHETAH
Classification_CHETAH <- function(input_object,ref_object,n_gene_used)
{
input_object_count <- input_object@assays$RNA@counts
input_object_reduced_tsne <- Embeddings(input_object, reduction = "tsne")#使用embeddings 函数来调用tsne 值
all.equal(rownames(input_object_reduced_tsne), colnames(input_object_count))
# 创建input_object
input_cell <- SingleCellExperiment(assays = list(counts = input_object_count),
reducedDims = SimpleList(TSNE = input_object_reduced_tsne))
## Classification
input_cell <- CHETAHclassifier(input = input_cell,
ref_cells = ref_object, n_genes = n_gene_used)
pdf("PlotCHETAH_confidence_score_0.1.pdf",18,12)
PlotCHETAH(input_cell)
dev.off()
input_object$celltype_CHETAH_confidence_score_0.1 <- as.factor(input_cell$celltype_CHETAH)
## confidence score =0
input_cell <- Classify(input_cell, 0)
pdf("PlotCHETAH_confidence_score_0.pdf",8,8)
PlotCHETAH(input = input_cell, tree = FALSE)
dev.off()
input_object$celltype_CHETAH_confidence_score_0 <- as.factor(input_cell$celltype_CHETAH)
return(input_object)
}
# Seurat
Make_ref_data_seurat <- function(ref_exp,ref_celltype){
ref.seurat <- CreateSeuratObject(counts = ref_exp)
ref.seurat <- SCTransform(ref.seurat)
ref.seurat <- RunPCA(object = ref.seurat, npcs = 30, verbose = FALSE)
ref.seurat$celltype <- ref_celltype
saveRDS(ref.seurat, file = "ref_seurat.rds")
return(ref.seurat)
}
Classification_Seurat <- function(input_object,ref.seurat){
input_object.anchors <- FindTransferAnchors(reference =ref.seurat, query = input_object,
dims = 1:30,project.query = T,k.filter=150,k.anchor=5)
predictions <- TransferData(anchorset = input_object.anchors, refdata = ref.seurat$celltype,
dims = 1:30)
input_object <- AddMetaData(object = input_object, metadata = predictions)
input_object$celltype_Seurat <- input_object$predicted.id
return(input_object)
}
## SingleR (http://comphealth.ucsf.edu/SingleR/).
#library(SingleR)
#singler = CreateSinglerSeuratObject(counts.file, annot, project.name,
# min.genes = 500, technology, species = "Human" (or "Mouse"), citation,
# normalize.gene.length = F, min.cells = 2, npca = 10
# regress.out = "nUMI", reduce.seurat.object = T)
#counts.file = 'GSE74923_L1210_CD8_processed_data.txt'
# This file was probably proccessed with Excel as there are duplicate gene names
# (1-Mar, 2-Mar, etc.). They were removed manually.
#annot.file = 'GSE74923_L1210_CD8_processed_data.txt_types.txt' # a table with two columns
# # cell name and the original identity (CD8 or L1210)
#singler = CreateSinglerSeuratObject(counts.file, annot.file, 'GSE74923',
# variable.genes='de', regress.out='nUMI',
# technology='C1', species='Mouse',
# citation='Kimmerling et al.', reduce.file.size = F,
# normalize.gene.length = T)
#save(singler,file='GSE74923.RData'))
# The object can then be saved and uploaded to the SingleR web-app for further analysis and visualization or using functions available in the SingleR package (see vignette).
#save(singler,file=paste0(project.name,'.RData')
##-------------------------------------10. celltype hist between different condition----------------------------------------##
## sample_info <- as.factor(sample_info)
## cell_type <- as.factor(exp_count.seurat$celltype)
Sample_celltype_hist <- function(sample_info,cell_type,save_name1){
print("sample_info and cell_type must be factor")
Sample_celltype <- table(sample_info[cell_type == levels(cell_type)[1]])
col_name <- levels(cell_type)[1]
len = length(levels(cell_type))
for (i in 2:len)
{
celltype <- table(sample_info[cell_type==levels(cell_type)[i]])
col_name_1 <- levels(cell_type)[i]
Sample_celltype <- cbind(Sample_celltype,celltype)
col_name <- cbind(col_name,col_name_1)
}
colnames(Sample_celltype) <- col_name
save_name1 <- paste0(save_name1,".csv")
write.csv(Sample_celltype,file = save_name1)
return(Sample_celltype)}
Plot_celltype_hist <- function(plot_data,pdf_name){
## ggplot2
library(ggplot2)
p1 <- ggplot(plot_data,aes(x=celltype,y=cell_number,fill=type))+geom_bar(stat="identity",position ="dodge")
p1 <- p1 + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) #移除网格线背景以及加坐标线
p1 <- p1 + theme(axis.text.x = element_text(size = 10, color = "black", face="bold",
vjust = 0.8, hjust = 0.6, angle = 30))
pdf(paste0(pdf_name,".pdf"),8,8)
print(p1)
dev.off()
}
##-------------------------12.差异基因数目的barplot----------------##
### computer up/down gene number
gene_number_stats <- function(file, sheetnames, FC_cutoff, sig_method, sig_cutoff){
if(sig_method == "p_val"){
stats <- c()
for(i in sheetnames){
DEGs <- read.xlsx(file, sheet = i, colNames = T, rowNames = T)
up_index <- DEGs$avg_log2FC > FC_cutoff & DEGs$p_val < sig_cutoff
up_num <- nrow(DEGs[up_index, ])
down_index <- DEGs$avg_log2FC < -FC_cutoff & DEGs$p_val < sig_cutoff
down_num <- nrow(DEGs[down_index, ])
tmp <- c(up_num, down_num, i)
stats <- rbind(stats,tmp)
}
}
else{
stats <- c()
for(i in sheetnames){
DEGs <- read.xlsx(file, sheet = i, colNames = T, rowNames = T)
up_index <- DEGs$avg_log2FC > FC_cutoff & DEGs$p_val_adj < sig_cutoff
up_num <- nrow(DEGs[up_index, ])
down_index <- DEGs$avg_log2FC < -FC_cutoff & DEGs$p_val_adj < sig_cutoff
down_num <- nrow(DEGs[down_index, ])
tmp <- c(up_num, down_num, i)
stats <- rbind(stats, tmp)
}
}
stats <- data.frame(stats)
stats[,1] <- as.numeric(as.character(stats[,1]))
stats[,2] <- as.numeric(as.character(stats[,2]))
colnames(stats) <- c("up","down","celltype")
return(stats)
}
###plot
Plot_function <- function(){
library(ggplot2)
plot_data <- data.frame(number=c(data$up,-1*data$down),
celltype=rep(data$celltype,2),
ud=c(rep("up",dim(data)[1]),rep("down",dim(data)[1])))
plot_data$celltype <- factor(plot_data$celltype,levels=c("all_cells",levels(Aggregated_seurat$celltype_assign)[-3]))
range(plot_data$number)
p1 <- ggplot(plot_data, aes(x=celltype, y=number, fill = ud))+
geom_col(position = position_dodge(width = 0), width = 0.6, size = 0.3, colour = "black")+
theme(panel.grid = element_blank(),
panel.background = element_rect(color = "black", fill = "transparent"),
legend.position="none",
axis.text.x = element_text(angle = 45, hjust = 1)) +
labs(x = "celltype", y = "number") +
geom_hline(yintercept = 0, size = 0.3) +
scale_fill_manual(values=c("#377EB8","#E41A1C"))+
scale_y_continuous(breaks = seq(-500, 300, 100), labels = as.character(abs(seq(-500, 300, 100))), limits = c(-500, 300))
ggsave("20200802_Figure_4_Female_DEG_number_barplot.pdf", p1, width = 5, height = 4)
}
####-------画特定基因在特定细胞中表达的featureplot图--------------------##
FeaturePlot_SpecificGene <- function(subtype_object, intertest_DEGs, gene_num, pic_name){
raw_exp <- FetchData(subtype_object, vars = intertest_DEGs$GeneName)
raw_exp$celltype <- subtype_object$celltype
new_exp <- matrix(0, nrow(raw_exp), ncol(raw_exp)-1)
j = 1
for (i in intertest_DEGs$Celltypes)
{
if(j <= gene_num){
print(i)
index <- raw_exp$celltype==i
new_exp[index, j] <- raw_exp[index, j]
j <- j + 1}
else{
break
}
}
colnames(new_exp) <- colnames(raw_exp)[1:gene_num]
rownames(new_exp) <- rownames(raw_exp)
new_exp <- as.data.frame(new_exp)
subtype_object <- AddMetaData(subtype_object, metadata = new_exp,
col.name = colnames(new_exp))
metrics <- colnames(new_exp)
p3 <- FeaturePlot(subtype_object, features = metrics,
col = c("lightgrey", "red"),
split.by = "orig.ident",
reduction = "tsne", min.cutoff = 'q10')
ggsave(pic_name, p3, width = 10, height = 40)
return(subtype_object)
}
##------------找给定一个seurat中每类细胞类型中的实验组和对照组之间的差异基因--------------##
##--20201205
FindDEGsFromCelltypes <- function(Aggregated_seurat, cell_type, treat,
control,logfc_threshold,
min_pct_cell, file_prefix){
results_DEGs_wilcox <- list()
results_DEGs_bimod <- list()
results_DEGs_roc <- list()
for (i in 1:length(cell_type)){
## 找每一类Obesity 和 Control的差异基因
subset_seurat <- subset(x = Aggregated_seurat, subset = celltype_assign == cell_type[i])
print(cell_type[i])
## wilcox
print("Wilcox test")
DEGs_treat_VS_Control_wilcox <- FindMarkers(subset_seurat,slot="data",
logfc.threshold = logfc_threshold,ident.1 = treat,
ident.2 = control,min.pct = min_pct_cell,
group.by = "orig.ident",test.use = "wilcox",
min.cells.group = 1)
results_DEGs_wilcox[[i]] <- DEGs_treat_VS_Control_wilcox
## bimod
print("Biomod test")
DEGs_treat_VS_Control_bimod <- FindMarkers(subset_seurat,slot="data",
logfc.threshold = logfc_threshold,ident.1 = treat,
ident.2 = control,min.pct = min_pct_cell,
group.by = "orig.ident",test.use = "bimod",
min.cells.group = 1)
results_DEGs_bimod[[i]] <- DEGs_treat_VS_Control_bimod
## roc
print("Roc test")
DEGs_treat_VS_Control_roc <- FindMarkers(subset_seurat,slot="data",
logfc.threshold = logfc_threshold,ident.1 = treat,
ident.2 = control,min.pct = min_pct_cell,
group.by = "orig.ident",test.use = "roc",
min.cells.group = 1)
results_DEGs_roc[[i]] <- DEGs_treat_VS_Control_roc
}
## 把list的列名重命名
names(results_DEGs_wilcox) <- as.character(cell_type)#重命名
names(results_DEGs_bimod) <- as.character(cell_type)#重命名
names(results_DEGs_roc) <- as.character(cell_type)#重命名
## 将结果写到excel中
write.xlsx(results_DEGs_wilcox, file = paste0(file_prefix, "_wilcox_DEGs_Treat_VS_Control.xlsx"),
col.names = T, row.names = T)
write.xlsx(results_DEGs_bimod, file = paste0(file_prefix, "_biomod_DEGs_Treat_VS_Control.xlsx"),
col.names = T, row.names = T)
write.xlsx(results_DEGs_roc, file = paste0(file_prefix, "_roc_DEGs_Treat_VS_Control.xlsx"),
col.names = T, row.names = T)
}
FindDEGsFromCelltypesWilcoxtest <- function(Aggregated_seurat, cell_type, treat,
control,logfc_threshold,
min_pct_cell, file_prefix){
results_DEGs_wilcox <- list()
for (i in 1:length(cell_type)){
## 找每一类Obesity 和 Control的差异基因
subset_seurat <- subset(x = Aggregated_seurat, subset = celltype_assign == cell_type[i])
print(cell_type[i])
## wilcox
print("Wilcox test")
DEGs_treat_VS_Control_wilcox <- FindMarkers(subset_seurat,slot="data",
logfc.threshold = logfc_threshold,ident.1 = treat,
ident.2 = control,min.pct = min_pct_cell,
group.by = "orig.ident",test.use = "wilcox",
min.cells.group = 1)
results_DEGs_wilcox[[i]] <- DEGs_treat_VS_Control_wilcox
}
## 把list的列名重命名
names(results_DEGs_wilcox) <- as.character(cell_type)#重命名
## 将结果写到excel中
write.xlsx(results_DEGs_wilcox, file = paste0(file_prefix, "_wilcox_DEGs_Treat_VS_Control.xlsx"),
col.names = T, row.names = T, overwrite = T)
}
####--------------------神经元分析经常画的一些图------------------------#####
## 聚类图和细胞marker图的参数
#reduction_type <- "umap"
#group_by_type <- "celltype_assign"
#pdf_prefix <- "Ob_female_6wk"
#celltype_markers <- c("Snap25", "Syt1", "Slc17a6",
#"Slc32a1", "Gad1", "Gad2",
#"Apoe", "Hcrt", "Pomc",
#"Agrp", "Cartpt", "Npy")
## 每类细胞数目比例图的plot_data
#plot_data <- table(Aggregated_seurat$orig.ident, Aggregated_seurat$celltype_assign)
#plot_data <- data.frame(Percentage = c(plot_data[1, ]/sum(plot_data[1, ]),
#plot_data[2, ]/sum(plot_data[2, ])),
#Treat = c(rep("Control", 4), rep("Ob", 4)),
#Type = rep(colnames(plot_data), 2))
#plot_data$Type <- factor(plot_data$Type,levels = c("Apoe", "GABA", "Glu", "Hybrid"))
## 印记基因热图所用的marker
#imprinted_genes <- c("Nap1l5", "Ndn", "Peg3", "Snrpn", "Ube3a", "Gnas")
NeuronSubtypePlots <- function(Aggregated_seurat, reduction_type, group_by_type,
celltype_markers, plot_data, imprinted_genes)
{
## 1.聚类图：总的聚类图和分开的聚类图
# 总的聚类图
p1 <- DimPlot(Aggregated_seurat, reduction=reduction_type, label = TRUE,
pt.size = 0.8 , label.size = 4.5, repel=TRUE,
group.by = group_by_type)
pdf(paste0("F1A_", pdf_prefix, "_neuron_clustering_assign_celltyp.pdf"), 6, 5)
print(p1)
dev.off()
# 分开的聚类图
p2 <- DimPlot(Aggregated_seurat, reduction=reduction_type, label = TRUE,
pt.size = 0.8 , label.size = 4.5, repel=TRUE,
group.by = group_by_type, split.by = "orig.ident")
pdf(paste0("F1B_", pdf_prefix, "_neuron_clustering_assign_celltyp_split.pdf"), 10, 5)
print(p2)
dev.off()
## 2.鉴定神经元用的marker的featureplot图和小提琴图
# 用的marker的featureplot图
p3 <- FeaturePlot(Aggregated_seurat, features = celltype_markers, ncol = 4,
reduction = reduction_type, cols = c("lightgrey", "red"),
pt.size = 0.8)
pdf(paste0("F2A_", pdf_prefix, "_featureplot_of_neurons_markers.pdf"), 16, 12)
print(p3)
dev.off()
# 用的marker的小提琴图
p4 <- VlnPlot(Aggregated_seurat, features = celltype_markers, ncol = 4, pt.size = 0)
pdf(paste0("F2B_", pdf_prefix, "_vlnplot_of_neurons_markers.pdf"), 16, 12)
print(p4)
dev.off()
## 3.各个神经元类别在对照组和实验组之间的比例图
p5 <- ggplot(plot_data, aes(x = Treat,y = Percentage, fill = Type)) + geom_bar(position="stack", stat="identity")
p5 <- p5 + theme_classic()
# p <- p + facet_wrap(~Sex)
pdf(paste0("F3_", pdf_prefix, "_Neuron_percentage.pdf"), 4, 4)
print(p5)
dev.off()
## 4.神经元印记基因的热图
cor_len <- length(unique(Aggregated_seurat$celltype_assign))
annotation_col <- data.frame(subtype = Aggregated_seurat$celltype_assign,
row.names = rownames([email protected]))
heatmap_byGroup <- FetchData(object = Aggregated_seurat,
vars = imprinted_genes,
slot = 'scale.data')
heatmap_byGroup <- cbind(heatmap_byGroup,annotation_col)
heatmap_byGroup <- arrange(heatmap_byGroup, subtype)
heatmap_byGroup_order <- heatmap_byGroup[, 1:length(imprinted_genes)]
colorCount=18
getPalette = colorRampPalette(RColorBrewer::brewer.pal(9, "Set1"))
col1=getPalette(colorCount)[1:cor_len]
names(col1) <- levels(annotation_col$subtype)
ann_colors = list(group = col1)
pdf(paste0("F4_", pdf_prefix, "_Neuron_subtype_imprinted_DEGs.pdf"),9,3)
print(pheatmap(t(heatmap_byGroup_order), #fontsize_row=3,
colorRampPalette(rev(brewer.pal(n = 7, name = "RdYlBu")))(10),
breaks=seq(-1, 1, length.out = 10),
treeheight_row=0.5, treeheight_col=1,
border_color='grey', cluster_cols = F,cluster_rows = F,
fontsize_row = 18,
annotation_col = annotation_col,
annotation_colors = ann_colors,
show_colnames = F,show_rownames = T, scale='none',
border=TRUE, angle_col = "0",
fontsize = 10))
dev.off()
}
####------20201226 Pieplot of specific genes----------####
# Barplot
blank_theme <- theme_minimal()+
theme(
axis.title.x = element_blank(),
axis.title.y = element_blank(),
panel.border = element_blank(),
panel.grid=element_blank(),
axis.ticks = element_blank(),
plot.title=element_text(size=14, face="bold"),
axis.text.x = element_blank()
)
Pieplot <- function(plot_data){
bp<- ggplot(plot_data, aes(x=gene, y=value, fill=color, label = label))+
geom_bar(width = 0.5, stat = "identity", position = "stack") + facet_wrap(~treat)
# bp <- bp + scale_y_continuous(limits=c(0, 100), breaks=seq(0,90,20))
bp <- bp + blank_theme
bp <- bp + geom_text(size = 3, colour = "black", fontface = "bold",
position=position_stack(0.5))
pie <- bp + coord_polar("y", start=0)
pie <- pie + scale_fill_manual(values=c("#E69F00", "#56B4E9"))
pie <- pie + theme(legend.title = element_blank(), legend.position = "top")
return(pie)
}
Pieplot_Func <- function(seurat_object, used_features, celltype, pre_fix, width, height){
CellPertage <- DotPlot(seurat_object, features = used_features,
group.by = "celltype", split.by = "orig.ident")$data
id <- as.matrix(as.character(CellPertage$id))
len <- nrow(CellPertage)
CellPertage$celltype <- matrix(unlist(strsplit(id, '_')), len, 2, byrow = TRUE)[1:len, 1]
CellPertage$treat <- matrix(unlist(strsplit(id, '_')), len, 2, byrow = TRUE)[1:len, 2]
CellPertage$colors <- NULL
write.csv(CellPertage, paste0(pre_fix, "_cell_percentage.csv"))
pdf(paste0("Pieplot_of_", pre_fix,"_cell_percentage.pdf"), width, height)
for (j in celltype){
print(j)
CellPertage_temp <- CellPertage[CellPertage$celltype==j, ]
library(ggplot2)
plot_data <- data.frame(value = c(CellPertage_temp$pct.exp, 100-CellPertage_temp$pct.exp))
plot_data$treat <- c(CellPertage_temp$treat,CellPertage_temp$treat)
plot_data$color <- as.factor(c(rep("Expressed", length(CellPertage_temp$pct.exp)),
rep("Unexpressed", length(CellPertage_temp$pct.exp))))
genenames <- as.character(CellPertage_temp$features.plot)
plot_data$gene <- as.factor(c(rep(genenames, 2)))
plot_data$label <- paste0(round(plot_data$value, 2), "%")
p_plots <- list()
for (i in used_features){
p_plot <- paste0("p_",i)
index <- plot_data$gene==i
temp <- plot_data[index, ]
temp_plot <- Pieplot(temp)
assign(p_plot,temp_plot)
p_plots[[i]] <- get(p_plot)
}
plots <- plot_grid(plotlist=p_plots, labels = j)
print(plots)
}
dev.off()
}
## Usage
### Pomc Cartpt Gal Trh
## features_1 <- c("Pomc", "Cartpt", "Gal", "Trh")
# Female
## celltype <- c("GABA1", "GABA2", "GABA3", "GABA4", "GABA5",
#"GABA6", "GABA7", "Glu1", "Glu2", "Apoe")
## pre_fix <- "Female_Pomc_Cartpt_Gal_Trh"
## width <- 8
## height <- 5
##Pieplot_Func(female, features_1, celltype, pre_fix, width, height)
##------------20210104 Selected Cells and then reclustering--------------##
SelectCells <- function(Aggregated_seurat, n.cells, seed.use){
## select samples
cellid <- rownames([email protected])
index_control <- Aggregated_seurat$orig.ident=="WT"
control_cellid <- cellid[index_control]
index_ob <- Aggregated_seurat$orig.ident=="ob/ob"
ob_cellid <- cellid[index_ob]
set.seed(seed.use)
subset_controlcellid <- sample(control_cellid, n.cells, replace=TRUE)
subset_obcellid <- sample(ob_cellid, n.cells, replace=TRUE)
select_cellid <- as.vector(c(subset_controlcellid, subset_obcellid))
Aggregated_seurat$cellnames <- rownames([email protected])
subset_Object <- subset(Aggregated_seurat,
cells = select_cellid)
return(subset_Object)
}
Reclustering <- function(Aggregated_seurat, npc_used, resolution_number, k_param){
Aggregated_seurat <- SCTransform(Aggregated_seurat,
vars.to.regress = c("percent.rp","percent.mt"))
# RunPCA
set.seed(123)
Aggregated_seurat <- RunPCA(Aggregated_seurat, dims = 1:npc_used,
verbose = FALSE, check_duplicates = FALSE)
Aggregated_seurat <- RunTSNE(Aggregated_seurat, dims = 1:npc_used,
verbose = FALSE, check_duplicates = FALSE)
Aggregated_seurat <- RunUMAP(Aggregated_seurat, dims = 1:npc_used,
verbose = FALSE)
# FinderNeighbors
Aggregated_seurat <- FindNeighbors(Aggregated_seurat, dims = 1:npc_used,
verbose = FALSE, k.param = k_param)
Aggregated_seurat <- FindClusters(Aggregated_seurat, verbose = FALSE,
resolution = resolution_number)
# RenameIdents
levels_define <- as.numeric(levels(Aggregated_seurat))
new.cluster.ids <- levels_define + 1
names(new.cluster.ids) <- levels(Aggregated_seurat)
Aggregated_seurat <- RenameIdents(Aggregated_seurat, new.cluster.ids)
return(Aggregated_seurat)
}
ClusteringPlot <- function(Aggregated_seurat, prefix, pt.size, reduction_type,
resolution_number){
# FIgure 1
p1 <- DimPlot(Aggregated_seurat, reduction = reduction_type, label = TRUE,pt.size = pt.size,
label.size = 4, repel = TRUE, group.by = "ident") + NoLegend()
p1 <- p1 + labs(title = paste0("Clustering resolution", resolution_number))
p1 <- p1 + theme(plot.title = element_text(hjust = 0.5))
p2 <- DimPlot(Aggregated_seurat, reduction = reduction_type, label = TRUE,
pt.size = pt.size, label.size = 4, repel = TRUE,group.by = "celltype_assign") + NoLegend()
p2 <- p2 + labs(title = "Celltype Identification")
p2 <- p2 + theme(plot.title = element_text(hjust = 0.5))
pdf1_name <- paste(prefix, "F1_clustering", reduction_type, resolution_number,
".pdf", sep = "_")
plots <- plot_grid(p1, p2, ncol = 2, rel_widths = c(1.8,2), labels = c("A","B"), label_size = 20)
pdf(pdf1_name, 15, 8)
print(plots)
dev.off()
# FIgure 2
p2 <- DimPlot(Aggregated_seurat, reduction = reduction_type, label = TRUE,
group.by = "celltype_assign", pt.size = pt.size, label.size = 4,
repel=TRUE, split.by = "orig.ident") + NoLegend()
pdf2_name <- paste(prefix, "F2_clustering_split_by_sample", reduction_type, resolution_number,
".pdf", sep = "_")
pdf(pdf2_name, 15, 8)
print(p2)
dev.off()
}
MarkersPlot <- function(Aggregated_seurat, markers, names, prefix, reduction_type){
p_celltype_plots <- list()
clusters <- levels([email protected])
for (i in 1:length(markers)){
p_celltype <- paste0("p_celltype_",i)
temp <- VlnPlot(object = Aggregated_seurat, features = markers[i], ncol = 1, pt.size = 0)
temp <- temp + NoLegend()
temp <- temp + labs(title=paste0(clusters[i], " ",markers[i],"(", names[i], ")"))
temp <- temp + theme(axis.title.y=element_blank()) + theme(plot.title = element_text(hjust = 0.5))
assign(p_celltype, temp)
p_celltype_plots[[i]] <- get(p_celltype)
}
plots_celltype <- plot_grid(plotlist = p_celltype_plots, ncol = 2)
pdf3_name <- paste0(prefix, "_F3_Vlnplot_marker_each_celltype", ".pdf")
pdf(pdf3_name, 10, 13)
print(plots_celltype)
dev.off()
pdf4_name <- paste(prefix,"F4_Featureplot_marker",".pdf", sep = "_")
p <- FeaturePlot(Aggregated_seurat, features = markers,
reduction = reduction_type, ncol = 4, col = c("lightgrey", "red"))
ggsave(pdf4_name, p, width = 20, height = 12)
}
# err_prob_list <- c(0.01, 0.05, 0.1, 0.15, 0.2)
CellnumberTest <- function(Aggregated_seurat, prefix, control_name, treat_name, err_prob_list, random_num){
# 1. calculate cell number
sample_info <- as.factor(Aggregated_seurat$orig.ident)
cell_type <- Aggregated_seurat$celltype_assign
obs.counts <- Sample_celltype_hist(sample_info,cell_type,
paste0(prefix, "condition_cell_type"))
# 2. cellnumber test
print(obs.counts)
res.table.ControlVsObesity = c()
## Go through a series of error probabilities
for (err_prob in err_prob_list) {
tip.exp <- generateNull(obs.counts, n=random_num, p=err_prob);
## Control vs Obesity
res.1 = two.class.test(obs.counts, tip.exp, cond.control=control_name,
cond.treatment=treat_name, to.plot=F)
res.table.ControlVsObesity = rbind(res.table.ControlVsObesity, res.1)
}
rownames(res.table.ControlVsObesity) = as.character(paste0("error_p_", err_prob_list))
res.table.ControlVsObesity <- t(res.table.ControlVsObesity)
# 3.save results
cellnumber <- t(obs.counts)
cellnumber_pct <- data.frame(control_percent = round(cellnumber[,1]/sum(cellnumber[,1]),4) * 100,
obesity_percent = round(cellnumber[,2]/sum(cellnumber[,2]),4) * 100,
control_cellnumber = cellnumber[,1],
obesity_cellnumber = cellnumber[,2])
results <- cbind(cellnumber_pct, res.table.ControlVsObesity)
write.csv(results, paste(prefix, "cellnumber_test.csv", sep = "_"))
}
HeatmapPlot <- function(Aggregated_seurat, features_1, pdfname){
annotation_col <- data.frame([email protected],
row.names=rownames([email protected]))
colnames(annotation_col) <- 'group'
heatmap_byGroup <- FetchData(object = Aggregated_seurat, vars = features_1, slot='scale.data')
heatmap_byGroup <- cbind(heatmap_byGroup, annotation_col)
heatmap_byGroup <- arrange(heatmap_byGroup, group)
heatmap_byGroup_1 <- as.matrix(t(heatmap_byGroup[ ,1:dim(heatmap_byGroup)[2]-1]))
anno_col <- data.frame(group = heatmap_byGroup[ ,dim(heatmap_byGroup)[2]])
colnames(heatmap_byGroup_1) <- rownames(anno_col)
all.equal(rownames(anno_col), colnames(heatmap_byGroup_1))
pdf(pdfname,12,20)
pheatmap::pheatmap(heatmap_byGroup_1, #fontsize_row=3,
color=colorRampPalette(rev(brewer.pal(n = 8, name ="RdYlBu")))(100),
breaks=seq(-2, 2, length.out = 100),
treeheight_row=10, treeheight_col=2,
border_color='grey',cluster_cols = F,cluster_rows = F,
fontsize_row = 18,
annotation_col = anno_col,
#annotation_colors = ann_colors,
show_colnames = F,scale='none',
border=TRUE)
dev.off()
}
#####--------Basic function to convert human to mouse gene names--------######
# from https://www.r-bloggers.com/2016/10/converting-mouse-to-human-gene-names-with-biomart-package/
# Basic function to convert human to mouse gene names
convertHumanGeneList <- function(x){
#require("biomaRt")
#human = useMart("ensembl", dataset = "hsapiens_gene_ensembl")
#mouse = useMart("ensembl", dataset = "mmusculus_gene_ensembl")
#genesV2 = getLDS(attributes = c("hgnc_symbol"), filters = "hgnc_symbol", values = x , mart = human, attributesL = c("mgi_symbol"), martL = mouse, uniqueRows=T)
#humanx <- unique(genesV2[, 2])
# Print the first 6 genes found to the screen
#print(head(humanx))
#return(humanx)
}
######--------Basic function to convert mouse to human gene names#--------#########
#musGenes <- c("Hmmr", "Tlx3", "Cpeb4")
# Basic function to convert mouse to human gene names
convertMouseGeneList <- function(x){
#require("biomaRt")
#human = useMart("ensembl", dataset = "hsapiens_gene_ensembl")
#mouse = useMart("ensembl", dataset = "mmusculus_gene_ensembl")
#genesV2 = getLDS(attributes = c("mgi_symbol"), filters = "mgi_symbol", values = x , mart = mouse, attributesL = c("hgnc_symbol"), martL = human, uniqueRows=T)
#humanx <- unique(genesV2[, 2])
# Print the first 6 genes found to the screen
#print(head(humanx))
#return(humanx)
}
genes <- convertMouseGeneList(musGenes)
#####
#\' 使用R基本绘图函数绘制y轴不连续的柱形图
#\'
#\' 绘制y轴不连续的柱形图，具有误差线添加功能。断点位置通过btm和top参数设置，如果不设置，函数可自动计算合适的断点位置。
#\' @title gap.barplot function
#\' @param df 长格式的data.frame，即数据框中每一列为一组绘图数据。
#\' @param y.cols 用做柱形图y值的数据列（序号或名称），一列为一组。
#\' @param sd.cols 与y值列顺序对应的误差值的数据列（序号或名称）。
#\' @param btm 低位断点。如果btm和top均不设置，程序将自动计算和设置断点位置。
#\' @param top 高位断点。
#\' @param min.range 自动计算断点的阈值：最大值与最小值的最小比值
#\' @param max.fold 自动计算断点时最大值与下方数据最大值的最大倍数比
#\' @param ratio 断裂后上部与下部y轴长度的比例。
#\' @param gap.width y轴断裂位置的相对物理宽度（非坐标轴实际刻度）
#\' @param brk.type 断点类型，可设为normal或zigzag
#\' @param brk.bg 断点处的背景颜色
#\' @param brk.srt 断点标记线旋转角度
#\' @param brk.size 断点标记线的大小（长度）
#\' @param brk.col 断点标记线的颜色
#\' @param brk.lwd 断点标记线的线宽
#\' @param cex.error 误差线相对长度，默认为1
#\' @param ... 其他传递给R基本绘图函数barplot的参数
#\' @return 返回barplot的原始返回值，即柱形图的x坐标
#\' @examples
#\' datax <- na.omit(airquality)[,1:4]
#\' cols <- cm.colors(ncol(datax))
#\' layout(matrix(1:6, ncol=2))
#\' set.seed(0)
#\' for (ndx in 1:6){
#\' dt <- datax[sample(rownames(datax), 10), ]
#\' par(mar=c(0.5,2,0.5,0.5))
#\' brkt <- sample(c(\'normal\', \'zigzag\'), 1)
#\' gap.barplot(dt, col=cols, brk.type=brkt, max.fold=5, ratio=2)
#\' }
#\' @author ZG Zhao
#\' @export
gap.barplot <- function(df, y.cols=1:ncol(df), sd.cols=NULL, btm=NULL, top=NULL, min.range=10, max.fold=5, ratio=1, gap.width=1,
brk.type='normal', brk.bg='white', brk.srt=135, brk.size=1, brk.col='black', brk.lwd=1, cex.error=1, ...){
if (missing(df)) stop('No data provided.')
if (is.numeric(y.cols)) ycol <- y.cols else ycol <- colnames(df)==y.cols
if (!is.null(sd.cols))
if (is.numeric(sd.cols)) scol <- sd.cols else scol <- colnames(df)==sd.cols
## Arrange data
opts <- options()
options(warn=-1)
y <- t(df[, ycol])
colnames(y) <- NULL
if(missing(sd.cols)) sdx <- 0 else sdx <- t(df[, scol])
sdu <- y + sdx
sdd <- y - sdx
ylim <- c(0, max(sdu) * 1.05)
## 如果没有设置btm或top，自动计算
if (is.null(btm) | is.null(top)){
autox <- .auto.breaks(dt=sdu, min.range=min.range, max.fold=max.fold)
if (autox$flag){
btm <- autox$btm
top <- autox$top
} else {
xx <- barplot(y, beside=TRUE, ylim=ylim, ...)
if (!missing(sd.cols)) errorbar(xx, y, sdu - y, horiz=FALSE, cex=cex.error)
box()
return(invisible(xx))
}
}
## Set up virtual y limits
halflen <- btm - ylim[1]
xlen <- halflen * 0.1 * gap.width
v_tps1 <- btm + xlen # virtual top positions
v_tps2 <- v_tps1 + halflen * ratio
v_ylim <- c(ylim[1], v_tps2)
r_tps1 <- top # real top positions
r_tps2 <- ylim[2]
## Rescale data
lmx <- summary(lm(c(v_tps1, v_tps2)~c(r_tps1, r_tps2)))
lmx <- lmx$coefficients
sel1 <- y > top
sel2 <- y >=btm & y <=top
y[sel1] <- y[sel1] * lmx[2] + lmx[1]
y[sel2] <- btm + xlen/2
sel1 <- sdd > top
sel2 <- sdd >=btm & sdd <=top
sdd[sel1] <- sdd[sel1] * lmx[2] + lmx[1]
sdd[sel2] <- btm + xlen/2
sel1 <- sdu > top
sel2 <- sdu >=btm & sdu <=top