-
Notifications
You must be signed in to change notification settings - Fork 0
/
05-integr_singlecell_cbps.R
838 lines (603 loc) · 32.4 KB
/
05-integr_singlecell_cbps.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
#integr CBPs datasets thanks to hematomap
out<-"outputs/05-integr_singlecell_cbps"
dir.create(out)
source("scripts/utils/new_utils.R")
source("../singlecell/scripts/utils/HTO_utils.R")
library(Seurat)
####QC filtering and Demultiplexing data ####
# remove low quality/outlyers cells (doublet with nGene/RNA hi/lo or percent.mt Hi)
# threshold : if not 2 subpop : median +/- 4* median absolute deviation(mad), else cutoff based on distribution
#cbp4####
sample<-"cbp4"
umis<- Read10X("~/RUN/Run_539_10x_standard/Output/cellranger_count_cbp4_tri/single_cell_barcode_539_HTO_cbp4b/outs/filtered_feature_bc_matrix/")$`Gene Expression`
cbp4_all <- CreateSeuratObject(counts = umis,project = sample)
cbp4_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp4_all, pattern = "^MT-")
VlnPlot(object = cbp4_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
p1<-VlnPlot(object = cbp4_all, features ="percent.mt")+geom_hline(yintercept = median(cbp4_all$percent.mt)+ 4*mad(cbp4_all$percent.mt) )
p2<-VlnPlot(object = cbp4_all, features ="nCount_RNA")+geom_hline(yintercept = 60000)
p3<-VlnPlot(object = cbp4_all, features ="nFeature_RNA")+geom_hline(yintercept = 7000 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp4_QC_cells_metrics.png"))
cbp4_qc<-subset(cbp4_all,percent.mt<median(cbp4_all$percent.mt)+ 4*mad(cbp4_all$percent.mt)&nCount_RNA<60000&nFeature_RNA<7000)
#reassign samples
cbp4.htos<-as.matrix(Read10X("~/RUN/Run_539_10x_standard/Output/cellranger_count_cbp4_tri/single_cell_barcode_539_HTO_cbp4b/outs/filtered_feature_bc_matrix/")$`Antibody Capture`)
rownames(cbp4.htos)<-c("ctrlM555",
"ctrlM518",
"ctrlM537",
"lgaF551",
"lgaF543")
cbp4_qc[["HTO"]] <- CreateAssayObject(counts = cbp4.htos[,colnames(cbp4_qc)])
# Normalize HTO data, here we use centered log-ratio (CLR) transformation
cbp4_qc <- NormalizeData(cbp4_qc, assay = "HTO", normalization.method = "CLR")
cbp4_qc <- HTODemux(cbp4_qc, assay = "HTO",positive.quantile = 0.95)
table(cbp4_qc$HTO_classification.global)
# Doublet Negative Singlet
# 1017 2373 2487
#sex based recovery
cbp4_qc<-checkHTOSex(cbp4_qc,gene_male="RPS4Y1",gene_female="XIST")
# calculating pct of real singlet male/female cells expressing sex marker ( save in 'misc' of 'HTO' assay):
# for male : 82 % express the male gene
# for female : 97% express the female gene
#as ++ doublet/singlet, and only 82% male cells express RPS4y1, split in 2 based hi / lo RNA count
VlnPlot(object = cbp4_all, features ="nCount_RNA")+geom_hline(yintercept = 8000)
cbp4_qc_lo<-subset(cbp4_qc,nCount_RNA<8000)
cbp4_qc_lo <- NormalizeData(cbp4_qc_lo, assay = "HTO", normalization.method = "CLR")
cbp4_qc_lo <- HTODemux(cbp4_qc_lo, assay = "HTO",positive.quantile = 0.9999)
table(cbp4_qc_lo$HTO_classification.global)
# Doublet Negative Singlet
# 323 605 739
cbp4_qc_lo<-checkHTOSex(cbp4_qc_lo,gene_male="RPS4Y1",gene_female="XIST")
# for male : 52 % express the male gene
# for female : 85 % express the female gene
cbp4_qc_lo<-sexBasedHTOAssign(cbp4_qc_lo)
cbp4_qc_lo_s<-subset(cbp4_qc_lo,new.HTO_classif.global=="Singlet")
cbp4_qc_hi<-subset(cbp4_qc,nCount_RNA>=8000)
cbp4_qc_hi <- NormalizeData(cbp4_qc_hi, assay = "HTO", normalization.method = "CLR")
cbp4_qc_hi <- HTODemux(cbp4_qc_hi, assay = "HTO",positive.quantile = 0.95)
table(cbp4_qc_hi$HTO_classification.global)
# Doublet Negative Singlet
# 880 1511 1819
cbp4_qc_hi<-checkHTOSex(cbp4_qc_hi,gene_male="RPS4Y1",gene_female="XIST")
# for male : 100 % express the male gene
# for female : 100 % express the female gene
cbp4_qc_hi<-sexBasedHTOAssign(cbp4_qc_hi)
# Bad_HTO_assign Doublet Negative Singlet
# 112 1201 446 2451
cbp4_qc_hi_s<-subset(cbp4_qc_hi,new.HTO_classif.global=="Singlet")
#merge the 2
cbp4_qc_s<-merge(cbp4_qc_hi_s,cbp4_qc_lo_s)
cbp4_qc_s$sample<-cbp4_qc_s$new.ID
table(cbp4_qc_s$sample)
# ctrlM518 ctrlM537 ctrlM555 lgaF543 lgaF551
# 678 470 619 1013 745
saveRDS(cbp4_qc_s,fp(out,"cbp4.rds"))
#cbp2####
sample<-"cbp2"
umis<- Read10X("~/RUN/Run_539_10x_standard/Output/cellranger_count_cbp2b_tri/single_cell_barcode_539_HTO_cbp2b/outs/filtered_feature_bc_matrix/")$`Gene Expression`
cbp2_all <- CreateSeuratObject(counts = umis,project = sample)
cbp2_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp2_all, pattern = "^MT-")
VlnPlot(object = cbp2_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
p1<-VlnPlot(object = cbp2_all, features ="percent.mt")+geom_hline(yintercept = median(cbp2_all$percent.mt)+ 4*mad(cbp2_all$percent.mt) )
p2<-VlnPlot(object = cbp2_all, features ="nCount_RNA")+geom_hline(yintercept = 60000)
p3<-VlnPlot(object = cbp2_all, features ="nFeature_RNA")+geom_hline(yintercept = 6500 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp2_QC_cells_metrics.png"))
cbp2_qc<-subset(cbp2_all,percent.mt<median(cbp2_all$percent.mt)+ 4*mad(cbp2_all$percent.mt)&nCount_RNA<60000&nFeature_RNA<6500)
#reassign samples
cbp2.htos<-as.matrix(Read10X("~/RUN/Run_539_10x_standard/Output/cellranger_count_cbp2b_tri/single_cell_barcode_539_HTO_cbp2b/outs/filtered_feature_bc_matrix/")$`Antibody Capture`)
rownames(cbp2.htos)<-c("ctrlF528",
"ctrlM539",
"iugrM553",
"iugrM558",
"lgaM549",
"lgaF532")
cbp2_qc[["HTO"]] <- CreateAssayObject(counts = cbp2.htos[,colnames(cbp2_qc)])
cbp2_qc <- NormalizeData(cbp2_qc, assay = "HTO", normalization.method = "CLR")
cbp2_qc <- HTODemux(cbp2_qc, assay = "HTO",positive.quantile = 0.9999)
table(cbp2_qc$HTO_classification.global)
# Doublet Negative Singlet
# 9536 1143 98
#as ++ doublet split in 2 based hi / lo RNA count
VlnPlot(object = cbp2_qc, features ="nCount_RNA",group.by = "orig.ident")+geom_hline(yintercept = 6000)
cbp2_qc_lo<-subset(cbp2_qc,nCount_RNA<6000)
cbp2_qc_lo <- NormalizeData(cbp2_qc_lo, assay = "HTO", normalization.method = "CLR")
cbp2_qc_lo <- HTODemux(cbp2_qc_lo, assay = "HTO",positive.quantile = 0.95)
table(cbp2_qc_lo$HTO_classification.global)
# Doublet Negative Singlet
# 1749 1002 49
cbp2_qc_lo<-checkHTOSex(cbp2_qc_lo,gene_male="RPS4Y1",gene_female="XIST")
# for male : 52 % express the male gene
# for female : 36 % express the female gene
cbp2_qc_hi<-subset(cbp2_qc,nCount_RNA>=6000)
cbp2_qc_hi <- NormalizeData(cbp2_qc_hi, assay = "HTO", normalization.method = "CLR")
cbp2_qc_hi <- HTODemux(cbp2_qc_hi, assay = "HTO",positive.quantile = 0.99)
table(cbp2_qc_hi$HTO_classification.global)
# Doublet Negative Singlet
# 1323 1844 4810
cbp2_qc_hi<-checkHTOSex(cbp2_qc_hi,gene_male="RPS4Y1",gene_female="XIST")
# for male : 99 % express the male gene
# for female : 98 % express the female gene
cbp2_qc_hi<-sexBasedHTOAssign(cbp2_qc_hi)
# Bad_HTO_assign Doublet Negative Singlet
# 84 1596 503 5794
#merge the 2
cbp2_qc<-merge(cbp2_qc_hi,cbp2_qc_lo)
#demultiplex on SNP info
mat_gt<-fread("../lineage_tracing/outputs/cbp2/25pct_det_and_variables_snps_barcodes_genotype_matrices_imputed.tsv")
mat_gt<-as.matrix(data.frame(mat_gt,row.names = "snp"))
dim(mat_gt)
colnames(mat_gt)<-str_replace(colnames(mat_gt),"\\.","-")
sum(colnames(cbp2_qc)%in%colnames(mat_gt)) #8292
cbp2_qc_snp<-cbp2_qc[,colnames(cbp2_qc)%in%colnames(mat_gt)]
cbp2_qc_snp[["SNP"]]<-CreateAssayObject(data=mat_gt[,colnames(cbp2_qc_snp)])
DefaultAssay(cbp2_qc_snp)<-"SNP"
cbp2_qc_snp<-FindVariableFeatures(cbp2_qc_snp)
cbp2_qc_snp<-ScaleData(cbp2_qc_snp)
cbp2_qc_snp<-RunPCA(cbp2_qc_snp)
cbp2_qc_snp<-RunUMAP(cbp2_qc_snp,dims = 1:10)
cbp2_qc_snp<-FindNeighbors(cbp2_qc_snp,dims = 1:10)
cbp2_qc_snp<-FindClusters(cbp2_qc_snp,resolution = 0.5)
DimPlot(cbp2_qc_snp,label=T,group.by = c("seurat_clusters","new.ID"))
new.idents<-c("lgaM549",
"iugrM558",
"ctrlM539",
"lgaF532",
"iugrM553",
"lgaM549",
"ctrlF528",
"Doublet",
"ctrlM539",
"iugrM558",
"Doublet",
"Doublet",
"Doublet",
"Doublet"
)
names(new.idents)<-levels(cbp2_qc_snp)
cbp2_qc_snp<-RenameIdents(cbp2_qc_snp,new.idents)
DimPlot(cbp2_qc_snp,label = T)
cbp2_qc_snp[["snp.ID"]]<-Idents(cbp2_qc_snp)
cbp2_qc_s<-subset(cbp2_qc_snp,snp.ID!="Doublet")
cbp2_qc_s$sample<-cbp2_qc_s$snp.ID
table(cbp2_qc_s$sample)
# lgaM549 iugrM558 ctrlM539 lgaF532 iugrM553 ctrlF528
# 2459 1381 1358 929 768 692
saveRDS(cbp2_qc_s,fp(out,"cbp2.rds"))
#cbp3####
sample<-"cbp3"
umis<- Read10X("../singlecell/datasets/cbp3/filtered_feature_bc_matrix/")
cbp3_all <- CreateSeuratObject(counts = umis,project = sample)
cbp3_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp3_all, pattern = "^MT-")
VlnPlot(object = cbp3_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
p1<-VlnPlot(object = cbp3_all, features ="percent.mt")+geom_hline(yintercept = median(cbp3_all$percent.mt)+ 4*mad(cbp3_all$percent.mt) )
p2<-VlnPlot(object = cbp3_all, features ="nCount_RNA")+geom_hline(yintercept = 60000)
p3<-VlnPlot(object = cbp3_all, features ="nFeature_RNA")+geom_hline(yintercept = 6500 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp3_QC_cells_metrics.png"))
cbp3_qc<-subset(cbp3_all,percent.mt<median(cbp3_all$percent.mt)+ 4*mad(cbp3_all$percent.mt)&nCount_RNA<60000&nFeature_RNA<6500)
#reassign samples
cbp3.htos<-as.matrix(Read10X("../singlecell/datasets/cbp3/HTO_CBP3/umi_count/",gene.column = 1))[1:3,]
rownames(cbp3.htos)<-c("ctrlF523",
"iugrF524",
"lgaF559")
colnames(cbp3.htos)<-paste0(colnames(cbp3.htos),"-1")
sum(colnames(cbp3_qc)%in%colnames(cbp3.htos)) #2570/2590
common_bc<-intersect(colnames(cbp3_qc),colnames(cbp3.htos))
cbp3_qc<-cbp3_qc[,common_bc]
cbp3_qc[["HTO"]] <- CreateAssayObject(counts = cbp3.htos[,common_bc])
# Normalize HTO data, here we use centered log-ratio (CLR) transformation
cbp3_qc <- NormalizeData(cbp3_qc, assay = "HTO", normalization.method = "CLR")
cbp3_qc <- HTODemux(cbp3_qc, assay = "HTO",positive.quantile = 0.98)
table(cbp3_qc$HTO_classification.global)
#++soublet need split in 2
VlnPlot(object = cbp3_all, features ="nFeature_RNA")+geom_hline(yintercept = 800 )
cbp3_qc_lo<-subset(cbp3_qc,nFeature_RNA<800)
cbp3_qc_lo <- NormalizeData(cbp3_qc_lo, assay = "HTO", normalization.method = "CLR")
cbp3_qc_lo <- HTODemux(cbp3_qc_lo, assay = "HTO",positive.quantile = 0.95)
table(cbp3_qc_lo$HTO_classification.global)
# Doublet Negative Singlet
# 321 411 108
cbp3_qc_hi<-subset(cbp3_qc,nFeature_RNA>=800)
cbp3_qc_hi <- NormalizeData(cbp3_qc_hi, assay = "HTO", normalization.method = "CLR")
cbp3_qc_hi <- HTODemux(cbp3_qc_hi, assay = "HTO",positive.quantile = 0.98)
table(cbp3_qc_hi$HTO_classification.global)
# Doublet Negative Singlet
# 278 723 729
cbp3_qc<-merge(cbp3_qc_hi,cbp3_qc_lo)
cbp3_qc_s<-subset(cbp3_qc,HTO_classification.global=="Singlet")
cbp3_qc_s$sample<-cbp3_qc_s$hash.ID
table(cbp3_qc_s$sample)
# ctrlF523 iugrF524 lgaF559
# 206 368 263
saveRDS(cbp3_qc_s,fp(out,"cbp3.rds"))
#cbp8####
sample<-"cbp8"
umis<- Read10X("~/RUN/Run_554_10x_standard/Output/cellranger_count_hto/single_cell_barcode_run_554_cbp8/outs/filtered_feature_bc_matrix/")$`Gene Expression`
cbp8_all <- CreateSeuratObject(counts = umis,project = sample)
cbp8_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp8_all, pattern = "^MT-")
VlnPlot(object = cbp8_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
p1<-VlnPlot(object = cbp8_all, features ="percent.mt")+geom_hline(yintercept = median(cbp8_all$percent.mt)+ 4*mad(cbp8_all$percent.mt) )
p2<-VlnPlot(object = cbp8_all, features ="nCount_RNA")+geom_hline(yintercept = 40000)
p3<-VlnPlot(object = cbp8_all, features ="nFeature_RNA")+geom_hline(yintercept = 6000 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp8_QC_cells_metrics.png"))
cbp8_qc<-subset(cbp8_all,percent.mt<median(cbp8_all$percent.mt)+ 4*mad(cbp8_all$percent.mt)&nCount_RNA<40000&nFeature_RNA<6000)
#reassign samples
cbp8.htos<-as.matrix(Read10X("~/RUN/Run_554_10x_standard/Output/cellranger_count_hto/single_cell_barcode_run_554_cbp8/outs/filtered_feature_bc_matrix/")$`Antibody Capture`)
rownames(cbp8.htos)<-c("ctrlM503",
"ctrlM530",
"lgaM556",
"lgaM496")
cbp8_qc[["HTO"]] <- CreateAssayObject(counts = cbp8.htos[,colnames(cbp8_qc)])
# Normalize HTO data, here we use centered log-ratio (CLR) transformation
cbp8_qc <- NormalizeData(cbp8_qc, assay = "HTO", normalization.method = "CLR")
cbp8_qc <- HTODemux(cbp8_qc, assay = "HTO",positive.quantile = 0.97)
table(cbp8_qc$HTO_classification.global)
# Doublet Negative Singlet
# 843 1367 3652
cbp8_qc_s<-subset(cbp8_qc,HTO_classification.global=="Singlet")
cbp8_qc_s$sample<-cbp8_qc_s$hash.ID
saveRDS(cbp8_qc_s,fp(out,"cbp8.rds"))
#cbp0_ctrl####
sample<-"cbp0_ctrl"
umis<- as.matrix(data.frame(fread("../singlecell/datasets/cbp0/CBP547.csv"),row.names=1))
dim(umis) #33538 6965
head(rownames(umis))
gene_trad<-TransEnsembltoSymbol(rownames(umis))
umis_f<-umis[gene_trad[hgnc_symbol!=""]$ensembl_gene_id,]
rownames(umis_f)<-gene_trad[hgnc_symbol!=""]$hgnc_symbol
sum(duplicated(rownames(umis_f))) #8
cbp0_ctrl_all <- CreateSeuratObject(counts = umis_f,project = sample)
cbp0_ctrl_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp0_ctrl_all, pattern = "^MT-")
VlnPlot(object = cbp0_ctrl_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
p1<-VlnPlot(object = cbp0_ctrl_all, features ="percent.mt")+
geom_hline(yintercept = median(cbp0_ctrl_all$percent.mt)+ 4*mad(cbp0_ctrl_all$percent.mt) )
p2<-VlnPlot(object = cbp0_ctrl_all, features ="nCount_RNA")+
geom_hline(yintercept = median(cbp0_ctrl_all$nCount_RNA)-2*mad(cbp0_ctrl_all$nCount_RNA))+
geom_hline(yintercept = 14000)
p3<-VlnPlot(object = cbp0_ctrl_all, features ="nFeature_RNA")+
geom_hline(yintercept = median(cbp0_ctrl_all$nFeature_RNA)-2*mad(cbp0_ctrl_all$nFeature_RNA) )+
geom_hline(yintercept = 2900)
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp0_ctrl_QC_cells_metrics.png"))
cbp0_ctrl_qc<-subset(cbp0_ctrl_all,percent.mt<median(cbp0_ctrl_all$percent.mt)+ 4*mad(cbp0_ctrl_all$percent.mt)&
nCount_RNA>median(cbp0_ctrl_all$nCount_RNA)-2*mad(cbp0_ctrl_all$nCount_RNA)&nCount_RNA<14000&
nFeature_RNA>median(cbp0_ctrl_all$nFeature_RNA)-2*mad(cbp0_ctrl_all$nFeature_RNA)&nFeature_RNA<2900)
cbp0_ctrl_qc #6478 cells
cbp0_ctrl_qc$sample<-"ctrlF547"
saveRDS(cbp0_ctrl_qc,fp(out,"cbp0_ctrl.rds"))
#cbp0_lga####
sample<-"cbp0_lga"
umis<- as.matrix(data.frame(fread("../singlecell/datasets/cbp0/CBP552.csv"),row.names=1))
dim(umis) #33538 3020
umis_f<-umis[gene_trad[hgnc_symbol!=""]$ensembl_gene_id,]
rownames(umis_f)<-gene_trad[hgnc_symbol!=""]$hgnc_symbol
sum(duplicated(rownames(umis_f))) #8
cbp0_lga_all <- CreateSeuratObject(counts = umis_f,project = sample)
cbp0_lga_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp0_lga_all, pattern = "^MT-")
VlnPlot(object = cbp0_lga_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
p1<-VlnPlot(object = cbp0_lga_all, features ="percent.mt")+geom_hline(yintercept = median(cbp0_lga_all$percent.mt)+ 4*mad(cbp0_lga_all$percent.mt) )
p2<-VlnPlot(object = cbp0_lga_all, features ="nCount_RNA")+geom_hline(yintercept = median(cbp0_lga_all$nCount_RNA)-2*mad(cbp0_lga_all$nCount_RNA))+geom_hline(yintercept = 20000)
p3<-VlnPlot(object = cbp0_lga_all, features ="nFeature_RNA")+geom_hline(yintercept = median(cbp0_lga_all$nFeature_RNA)-2*mad(cbp0_lga_all$nFeature_RNA) )+geom_hline(yintercept = 4000)
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp0_lga_QC_cells_metrics.png"))
cbp0_lga_qc<-subset(cbp0_lga_all,percent.mt<median(cbp0_lga_all$percent.mt)+ 4*mad(cbp0_lga_all$percent.mt)&nCount_RNA>median(cbp0_lga_all$nCount_RNA)-2*mad(cbp0_lga_all$nCount_RNA)&nCount_RNA<20000&nFeature_RNA>median(cbp0_lga_all$nFeature_RNA)-2*mad(cbp0_lga_all$nFeature_RNA)&nFeature_RNA<4000)
cbp0_lga_qc #2897 cells
cbp0_lga_qc$sample<-"lgaF552"
saveRDS(cbp0_lga_qc,fp(out,"cbp0_lga.rds"))
#cbp6a####
sample<-"cbp6a"
umis<- Read10X("~/RUN/Run_538_10x_standard/Output/cellranger_count/run_538_10x-cbp6-a/outs/filtered_feature_bc_matrix/")
cbp6a_all <- CreateSeuratObject(counts = umis,project = sample)
cbp6a_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp6a_all, pattern = "^MT-")
VlnPlot(object = cbp6a_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
thr_mt<-median(cbp6a_all$percent.mt)+ 4*mad(cbp6a_all$percent.mt)
thr_ge<-median(cbp6a_all$nFeature_RNA)- 2*mad(cbp6a_all$nFeature_RNA)
thr_rn<-median(cbp6a_all$nCount_RNA)-2*mad(cbp6a_all$nCount_RNA)
min(cbp6a_all$nCount_RNA)
p1<-VlnPlot(object = cbp6a_all, features ="percent.mt")+geom_hline(yintercept = thr_mt )
p2<-VlnPlot(object = cbp6a_all, features ="nCount_RNA")+geom_hline(yintercept = 30000)
p3<-VlnPlot(object = cbp6a_all, features ="nFeature_RNA")+geom_hline(yintercept = 220 )+geom_hline(yintercept = 5000 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp6a_QC_cells_metrics.png"))
cbp6a_qc<-subset(cbp6a_all,percent.mt<median(cbp6a_all$percent.mt)+ 4*mad(cbp6a_all$percent.mt)&nCount_RNA<30000&nFeature_RNA>220&nFeature_RNA<5000)
#reassign samples
VlnPlot(cbp6a_qc,c("XIST","RPS4Y1"),group.by = "orig.ident")
sum(cbp6a_qc@assays$RNA@data["XIST",]>0)
sum(cbp6a_qc@assays$RNA@data["RPS4Y1",]>0)
cbp6a_qc@meta.data[cbp6a_qc@assays$RNA@data["XIST",]>0&
cbp6a_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"Doublet"
cbp6a_qc@meta.data[cbp6a_qc@assays$RNA@data["XIST",]>0&
cbp6a_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"ctrlF544"
cbp6a_qc@meta.data[cbp6a_qc@assays$RNA@data["XIST",]==0&
cbp6a_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"lgaM556"
cbp6a_qc@meta.data[cbp6a_qc@assays$RNA@data["XIST",]==0&
cbp6a_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"Negative"
table(cbp6a_qc@meta.data$sex.ID)
# ctrlF544 Doublet lgaM556 Negative
# 2098 456 3596 2435
cbp6a_qc_s<-subset(cbp6a_qc,sex.ID!="Doublet"&sex.ID!="Negative")
cbp6a_qc_s$sample<-cbp6a_qc_s$sex.ID
table(cbp6a_qc_s$sample)
# ctrlF544 lgaM556
# 2098 3596
saveRDS(cbp6a_qc_s,fp(out,"cbp6a.rds"))
#cbp6b####
sample<-"cbp6b"
umis<- Read10X("~/RUN/Run_538_10x_standard/Output/cellranger_count/run_538_10x-cbp6-b/outs/filtered_feature_bc_matrix/")
cbp6b_all <- CreateSeuratObject(counts = umis,project = sample)
cbp6b_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp6b_all, pattern = "^MT-")
VlnPlot(object = cbp6b_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
thr_mt<-median(cbp6b_all$percent.mt)+ 4*mad(cbp6b_all$percent.mt)
thr_ge<-median(cbp6b_all$nFeature_RNA)- 2*mad(cbp6b_all$nFeature_RNA)
thr_rn<-median(cbp6b_all$nCount_RNA)-2*mad(cbp6b_all$nCount_RNA)
p1<-VlnPlot(object = cbp6b_all, features ="percent.mt")+geom_hline(yintercept = thr_mt )
p2<-VlnPlot(object = cbp6b_all, features ="nCount_RNA")+geom_hline(yintercept = 30000)
p3<-VlnPlot(object = cbp6b_all, features ="nFeature_RNA")+geom_hline(yintercept = thr_ge )+geom_hline(yintercept = 5000 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp6b_QC_cells_metrics.png"))
cbp6b_qc<-subset(cbp6b_all,percent.mt< thr_mt&nCount_RNA<30000&nFeature_RNA>thr_ge&nFeature_RNA<5000)
#reassign samples
VlnPlot(cbp6b_qc,c("XIST","RPS4Y1"),group.by = "orig.ident")
sum(cbp6b_qc@assays$RNA@data["XIST",]>0)
sum(cbp6b_qc@assays$RNA@data["RPS4Y1",]>0)
cbp6b_qc@meta.data[cbp6b_qc@assays$RNA@data["XIST",]>0&
cbp6b_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"Doublet"
cbp6b_qc@meta.data[cbp6b_qc@assays$RNA@data["XIST",]>0&
cbp6b_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"ctrlF545"
cbp6b_qc@meta.data[cbp6b_qc@assays$RNA@data["XIST",]==0&
cbp6b_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"lgaM533"
cbp6b_qc@meta.data[cbp6b_qc@assays$RNA@data["XIST",]==0&
cbp6b_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"Negative"
table(cbp6b_qc@meta.data$sex.ID)
# ctrlF545 Doublet lgaM533 Negative
# 875 66 2437 363
cbp6b_qc_s<-subset(cbp6b_qc,sex.ID!="Doublet"&sex.ID!="Negative")
cbp6b_qc_s$sample<-cbp6b_qc_s$sex.ID
table(cbp6b_qc_s$sample)
# ctrlF545 lgaM533
# 875 2437
saveRDS(cbp6b_qc_s,fp(out,"cbp6b.rds"))
#cbp6c####
sample<-"cbp6c"
umis<- Read10X("~/RUN/Run_538_10x_standard/Output/cellranger_count/run_538_10x-cbp6-c/outs/filtered_feature_bc_matrix/")
cbp6c_all <- CreateSeuratObject(counts = umis,project = sample)
cbp6c_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp6c_all, pattern = "^MT-")
VlnPlot(object = cbp6c_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
thr_mt<-median(cbp6c_all$percent.mt)+ 4*mad(cbp6c_all$percent.mt)
thr_ge<-median(cbp6c_all$nFeature_RNA)- 2*mad(cbp6c_all$nFeature_RNA)
thr_rn<-median(cbp6c_all$nCount_RNA)-2*mad(cbp6c_all$nCount_RNA)
p1<-VlnPlot(object = cbp6c_all, features ="percent.mt")+geom_hline(yintercept = thr_mt )
p2<-VlnPlot(object = cbp6c_all, features ="nCount_RNA")+geom_hline(yintercept = 30000)
p3<-VlnPlot(object = cbp6c_all, features ="nFeature_RNA")+geom_hline(yintercept = 5000 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp6c_QC_cells_metrics.png"))
cbp6c_qc<-subset(cbp6c_all,percent.mt< thr_mt&nCount_RNA<30000&nFeature_RNA<5000)
#reassign samples
VlnPlot(cbp6c_qc,c("XIST","RPS4Y1"),group.by = "orig.ident")
sum(cbp6c_qc@assays$RNA@data["XIST",]>0)
sum(cbp6c_qc@assays$RNA@data["RPS4Y1",]>0)
cbp6c_qc@meta.data[cbp6c_qc@assays$RNA@data["XIST",]>0&
cbp6c_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"Doublet"
cbp6c_qc@meta.data[cbp6c_qc@assays$RNA@data["XIST",]>0&
cbp6c_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"ctrlF541"
cbp6c_qc@meta.data[cbp6c_qc@assays$RNA@data["XIST",]==0&
cbp6c_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"lgaM526"
cbp6c_qc@meta.data[cbp6c_qc@assays$RNA@data["XIST",]==0&
cbp6c_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"Negative"
table(cbp6c_qc@meta.data$sex.ID)
# ctrlF541 Doublet lgaM526 Negative
# 2346 114 1550 426
cbp6c_qc_s<-subset(cbp6c_qc,sex.ID!="Doublet"&sex.ID!="Negative")
cbp6c_qc_s$sample<-cbp6c_qc_s$sex.ID
table(cbp6c_qc_s$sample)
# ctrlF541 lgaM526
# 2346 1550
saveRDS(cbp6c_qc_s,fp(out,"cbp6c.rds"))
#cbp7a####
sample<-"cbp7a"
umis<- Read10X("~/RUN/Run_554_10x_standard/Output/cellranger_count/single_cell_barcode_run_554_10xcbp7-a/outs/filtered_feature_bc_matrix/")
cbp7a_all <- CreateSeuratObject(counts = umis,project = sample)
cbp7a_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp7a_all, pattern = "^MT-")
VlnPlot(object = cbp7a_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
thr_mt<-median(cbp7a_all$percent.mt)+ 4*mad(cbp7a_all$percent.mt)
thr_ge<-median(cbp7a_all$nFeature_RNA)- 2*mad(cbp7a_all$nFeature_RNA)
thr_rn<-median(cbp7a_all$nCount_RNA)-2*mad(cbp7a_all$nCount_RNA)
p1<-VlnPlot(object = cbp7a_all, features ="percent.mt")+geom_hline(yintercept = thr_mt )
p2<-VlnPlot(object = cbp7a_all, features ="nCount_RNA")+geom_hline(yintercept = 35000)
p3<-VlnPlot(object = cbp7a_all, features ="nFeature_RNA")+geom_hline(yintercept = 6000 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp7a_QC_cells_metrics.png"))
cbp7a_qc<-subset(cbp7a_all,percent.mt<thr_mt&nCount_RNA<35000&nFeature_RNA<6000)
#reassign samples
VlnPlot(cbp7a_qc,c("XIST","RPS4Y1"),group.by = "orig.ident")
sum(cbp7a_qc@assays$RNA@data["XIST",]>0)
sum(cbp7a_qc@assays$RNA@data["RPS4Y1",]>0)
sum(cbp7a_qc@assays$RNA@data["XIST",]>0&cbp7a_qc@assays$RNA@data["RPS4Y1",]>0) #408 doublet
cbp7a_qc@meta.data[cbp7a_qc@assays$RNA@data["XIST",]>0&
cbp7a_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"Doublet"
cbp7a_qc@meta.data[cbp7a_qc@assays$RNA@data["XIST",]>0&
cbp7a_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"lgaF554"
cbp7a_qc@meta.data[cbp7a_qc@assays$RNA@data["XIST",]==0&
cbp7a_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"ctrlM555"
cbp7a_qc@meta.data[cbp7a_qc@assays$RNA@data["XIST",]==0&
cbp7a_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"Negative"
table(cbp7a_qc@meta.data$sex.ID)
# ctrlM555 Doublet lgaF554 Negative
# 1990 408 2800 573
cbp7a_qc_s<-subset(cbp7a_qc,sex.ID!="Doublet"&sex.ID!="Negative")
cbp7a_qc_s$sample<-cbp7a_qc_s$sex.ID
table(cbp7a_qc_s$sample)
# ctrlM555 lgaF554
# 1990 2800
saveRDS(cbp7a_qc_s,fp(out,"cbp7a.rds"))
#cbp7b####
sample<-"cbp7b"
umis<- Read10X("~/RUN/Run_554_10x_standard/Output/cellranger_count/single_cell_barcode_run_554_10xcbp7-b/outs/filtered_feature_bc_matrix/")
cbp7b_all <- CreateSeuratObject(counts = umis,project = sample)
cbp7b_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp7b_all, pattern = "^MT-")
VlnPlot(object = cbp7b_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
thr_mt<-median(cbp7b_all$percent.mt)+ 4*mad(cbp7b_all$percent.mt)
thr_ge<-median(cbp7b_all$nFeature_RNA)- 2*mad(cbp7b_all$nFeature_RNA)
thr_rn<-median(cbp7b_all$nCount_RNA)-2*mad(cbp7b_all$nCount_RNA)
p1<-VlnPlot(object = cbp7b_all, features ="percent.mt")+geom_hline(yintercept = thr_mt )
p2<-VlnPlot(object = cbp7b_all, features ="nCount_RNA")+geom_hline(yintercept = 35000)
p3<-VlnPlot(object = cbp7b_all, features ="nFeature_RNA")+geom_hline(yintercept = 6000 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp7b_QC_cells_metrics.png"))
cbp7b_qc<-subset(cbp7b_all,percent.mt<thr_mt&nCount_RNA<35000&nFeature_RNA<6000)
#reassign samples
VlnPlot(cbp7b_qc,c("XIST","RPS4Y1"),group.by = "orig.ident")
sum(cbp7b_qc@assays$RNA@data["XIST",]>0)
sum(cbp7b_qc@assays$RNA@data["RPS4Y1",]>0)
sum(cbp7b_qc@assays$RNA@data["XIST",]>0&cbp7b_qc@assays$RNA@data["RPS4Y1",]>0) #419 doublet
cbp7b_qc@meta.data[cbp7b_qc@assays$RNA@data["XIST",]>0&
cbp7b_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"Doublet"
cbp7b_qc@meta.data[cbp7b_qc@assays$RNA@data["XIST",]>0&
cbp7b_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"lgaF521"
cbp7b_qc@meta.data[cbp7b_qc@assays$RNA@data["XIST",]==0&
cbp7b_qc@assays$RNA@data["RPS4Y1",]>0,"sex.ID"]<-"ctrlM518"
cbp7b_qc@meta.data[cbp7b_qc@assays$RNA@data["XIST",]==0&
cbp7b_qc@assays$RNA@data["RPS4Y1",]==0,"sex.ID"]<-"Negative"
table(cbp7b_qc@meta.data$sex.ID)
# ctrlM518 Doublet lgaF521 Negative
# 2009 469 3576 532
cbp7b_qc_s<-subset(cbp7b_qc,sex.ID!="Doublet"&sex.ID!="Negative")
cbp7b_qc_s$sample<-cbp7b_qc_s$sex.ID
table(cbp7b_qc_s$sample)
# ctrlM518 lgaF521
# 2009 3576
saveRDS(cbp7b_qc_s,fp(out,"cbp7b.rds"))
#cbp7c####
sample<-"cbp7c"
umis<- Read10X("~/RUN/Run_554_10x_standard/Output/cellranger_count/single_cell_barcode_run_554_10xcbp7-c/outs/filtered_feature_bc_matrix/")
cbp7c_all <- CreateSeuratObject(counts = umis,project = sample)
cbp7c_all[["percent.mt"]] <- PercentageFeatureSet(object = cbp7c_all, pattern = "^MT-")
VlnPlot(object = cbp7c_all, features = c("nFeature_RNA", "nCount_RNA", "percent.mt"))
#take the 4 four median absolute deviations above the median
thr_mt<-median(cbp7c_all$percent.mt)+ 4*mad(cbp7c_all$percent.mt)
thr_ge<-median(cbp7c_all$nFeature_RNA)- 2*mad(cbp7c_all$nFeature_RNA)
thr_rn<-median(cbp7c_all$nCount_RNA)-2*mad(cbp7c_all$nCount_RNA)
p1<-VlnPlot(object = cbp7c_all, features ="percent.mt")+geom_hline(yintercept = thr_mt )
p2<-VlnPlot(object = cbp7c_all, features ="nCount_RNA")+geom_hline(yintercept = 35000)
p3<-VlnPlot(object = cbp7c_all, features ="nFeature_RNA")+geom_hline(yintercept = 6000 )
p1|p2|p3+plot_layout(guides = "collect")
ggsave(fp(out,"cbp7c_QC_cells_metrics.png"))
cbp7c_qc<-subset(cbp7c_all,percent.mt<thr_mt&nCount_RNA<35000&nFeature_RNA<6000)
cbp7c_qc #2823 cells
cbp7c_qc$sample<-"ctrlM537"
saveRDS(cbp7c_qc,fp(out,"cbp7c.rds"))
#### INTEGRATION ####
#based on Hematomap (see 01-Make_Hematomap )
options(future.globals.maxSize = 50000 * 1024^2)
out<-"outputs/06-integr_singlecell_cbps"
source("../methyl/scripts/utils/new_utils.R")
library(Seurat)
library(parallel)
hmap<-readRDS("outputs/04-make_hematomap/hematomap_ctrls_sans_stress.rds")
hmap
DefaultAssay(hmap)<-"integrated"
hmap[["pca.annoy.neighbors"]] <- LoadAnnoyIndex(object = hmap[["pca.annoy.neighbors"]], file = "outputs/05-make_hematomap/reftmp.idx")
cbps_run<-c("cbp0_ctrl","cbp0_lga",
paste0("cbp",2:4),
ps("cbp6",c("a","b","c")),ps("cbp7",c("a","b","c")),
"cbp8")
length(cbps_run)#12
cbps_list<-lapply(cbps_run, function(run)readRDS(fp(out,ps(run,".rds"))))
cbps_list
cbps_list<-mclapply(cbps_list,function(x){
#message("calculate CC.Difference for",x@project.name)
if(!"S.Score"%in%colnames(x@meta.data)){
x<-SCTransform(x,method = "glmGamPoi")
x <- CellCycleScoring(x,s.features = cc.genes$s.genes,
g2m.features = cc.genes$g2m.genes,
set.ident = TRUE,
search=TRUE)
}
x$CC.Difference <- x$S.Score - x$G2M.Score
return(x)
},mc.cores = 6)
cbps_list<-mclapply(cbps_list, SCTransform,vars.to.regress=c("percent.mt","CC.Difference"),
return.only.var.genes=F,
method = "glmGamPoi",mc.cores = 6)
anchors <- list()
for (i in 1:length(cbps_list)) {
anchors[[i]] <- FindTransferAnchors(
reference = hmap,
query = cbps_list[[i]],
k.filter = NA,
reference.reduction = "pca",
reference.neighbors = "pca.annoy.neighbors",
dims = 1:50
)
}
for (i in 1:length(cbps_list)) {
cbps_list[[i]] <- MapQuery(
anchorset = anchors[[i]],
query = cbps_list[[i]],
reference = hmap,
refdata = list(
cell_type = "cell_type",
lineage = "lineage"),
reference.reduction = "pca",
reduction.model = "ref.umap"
)
}
# Merge the queries
cbps <- merge(cbps_list[[1]], cbps_list[2:length(cbps_list)],merge.dr = c("ref.pca","ref.umap"))
p<-DimPlot(cbps, reduction = "ref.umap", group.by = "predicted.cell_type", label = TRUE, repel = TRUE, label.size = 3) + NoLegend()
ggsave(fp(out,"predicted_cell_type.png"),plot=p)
DimPlot(cbps, reduction = "ref.umap", group.by = "predicted.lineage", label = TRUE, repel = TRUE, label.size = 3) + NoLegend()
#add metadata
cbps[["group"]]<-str_extract(cbps@meta.data$sample,"ctrl|lga|iugr")
cbps[["sex"]]<-str_extract(cbps@meta.data$sample,"M|F")
cbps[["group_sex"]]<-paste0(cbps@meta.data$group,cbps@meta.data$sex)
cbps$hto<-cbps$orig.ident%in%c("cbp2","cbp3","cbp4","cbp8")
cbps$batch<-cbps$orig.ident
cbps[["group_hto"]]<-paste0(cbps@meta.data$group,cbps@meta.data$hto)
cbps[["sample_hto"]]<-paste0(cbps@meta.data$sample,cbps@meta.data$hto)
cbps[["ambigous"]]<-cbps@meta.data$sample%in%c("iugrM558","lgaF559")
cbps[["cell_type_hmap"]]<-cbps$predicted.cell_type
cbps[["lineage_hmap"]]<-cbps$predicted.lineage
cbps$differentiated<-cbps$lineage_hmap%in%c("Mk/Er","18","DC","T cell","B cell")
#denovo umap
cbps <- RunUMAP(cbps, reduction = 'ref.pca', dims = 1:30,reduction.name = "denovo.umap",n.components = 2)
DimPlot(cbps, group.by = 'cell_type_hmap',reduction = "denovo.umap", label = TRUE)
DimPlot(cbps, group.by = 'lineage_hmap',reduction = "denovo.umap", label = TRUE)
saveRDS(cbps,fp(out,"cbps.rds"))
####CHECK INTEGR OK####
cbps_f<-subset(cbps,lineage_hmap!="18"&ambigous==F&group!="iugr"&hto==T)
rm(cbps)
#check qu'on a bien tout
#all
cbps_f# 12685 cells
length(unique(cbps_f$sample)) #14 samples
#n samples by group
mtd<-data.table(cbps_f@meta.data,keep.rownames = "bc")
mts<-unique(mtd,by=c("sample_hto"))
table(mts$hto,mts$group)
# ctrl lga
# TRUE 8 6
#n of cells
table(mtd$hto,mtd$group)
# ctrl lga
# TRUE 5823 6861
#n of HSC
table(mtd[lineage_hmap=="HSC"]$hto,mtd[lineage_hmap=="HSC"]$group)
# ctrl lga
# TRUE 2075 1903
saveRDS(cbps_f,fp(out,"cbps_filtered.rds"))
#check good assignmenet
VlnPlot(cbps_f,"predicted.lineage.score",group.by = "predicted.lineage",pt.size = 0)
#attention a HSC2, 3 et 4, GMP cycle, et MPP Ery