generated from labsyspharm/blank-software-website
-
Notifications
You must be signed in to change notification settings - Fork 0
/
prostate.Rmd
1113 lines (953 loc) · 46.4 KB
/
prostate.Rmd
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
---
title: "prostate"
output: pdf_document
date: "2023-10-03"
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```
```{r setup, include=FALSE}
library(data.table)
library(tidyr) # For gather()
library(tidyverse)
library(ggplot2)
library(survival)
library(forestplot)
library(survminer)
library(umap)
library(ggbeeswarm)
library(ggpubr)
library(igraph)
library(RColorBrewer)
library(scales)
library(cluster)
source("~/git/cyftools/R/helpers.R")
source("~/Dropbox/Sorger/projects/orion/R/orion_helpers.R")
generate_random_string <- function(string_length = 8) {
characters <- c(letters, LETTERS, 0:9) # alphanumeric characters
random_string <- paste0(sample(characters, string_length, replace = TRUE), collapse = "")
return(random_string)
}
save_ppdf <- function(g, filename, width=10, height=10) {
filepath <- paste0("~/Sorger/projects/prostate/figs_raw_paper/", filename)
pdf(filepath, useDingbats = FALSE, width, height)
print(g)
dev.off()
}
```
```{r clinical-data}
crr <- fread("~/Dropbox/Sorger/projects/prostate/230803_clinical_data.csv")
crr <- crr[!duplicated(Slide_ID)]
crr[, Gleason := ifelse(surgery_gleason %in% c("4 + 4","4 + 5","5 + 3","5 + 4"), "High", "Low")]
crr[, Gleason2 := c("4 + 4"="8", "3 + 4"="7", "4 + 5"="5", "5 + 3"="5", "5 + 4"="5")[surgery_gleason]]
setnames(crr, "surgery_lymphatic_vascular_invasion","LVI")
setnames(crr, "surgery_seminal_vesicle","SVI")
setnames(crr, "surgery_perineural", "PNI")
setnames(crr, "surgery_lymph_nodes","LNI")
crr[, any_high_risk := LNI == "Positive" | SVI == "Positive" | LVI == "Positive"]
## Fisher's exact
table_data <- crr[, .N, by = .(Gleason, LVI)]
contingency_table <- as.matrix(dcast(table_data, Gleason ~ LVI, value.var = "N", fill = 0)[, -1])
fisher.test(contingency_table)
```
```{r plot-setup, include=FALSE}
# ggplot theme
theme_jw <- function() {
theme_minimal() +
theme(
panel.grid.major = element_blank(),
panel.grid.minor = element_blank(),
axis.text = element_text(color = "black"),
panel.background = element_rect(fill = "transparent", colour = NA),
plot.background = element_rect(fill = "transparent", colour = NA) ,
text = element_text(family = "Helvetica"),
axis.text.x = element_text(color = "black"),
axis.text.y = element_text(color = "black"),
axis.line = element_line(colour = "black"),
axis.ticks = element_line()
)
}
gleason_colors=c("Low"="#1b9e77","High"="#7570b3")
```
```{r master-file-list}
master_files <- list.files(path = "~/Dropbox/Sorger/projects/prostate/chain_flip/", pattern = "cyf$", full.names = TRUE)
```
```{r cells-per-sample, include=FALSE}
## count the total number of cells in the project
num_cells <- sum(unlist(sapply(master_files, function(f) {
dt <- fread(cmd=paste("cyftools count", f))
})))
## get cell counts by type
cell_counts_by_type <- rbindlist(lapply(master_files, function(f) {
cat(basename(f), "\n")
dt <- fread(cmd=paste("cyftools cellcount", f, "-a 34816 -a 0 -a 3072 -a 6144 -a 2052 -a 20480 -a 18432 -a 17408 - | cyftools view - -R" ))
setnames(dt, "34816_count", "PD1CD3_count")
setnames(dt, "0_count", "unclassified_count")
setnames(dt, "3072_count", "CD3CD4_count")
setnames(dt, "6144_count", "CD3CD8_count")
setnames(dt, "2052_count", "AMACRCD3_count")
setnames(dt, "20480_count", "FOXP3CD8_count")
setnames(dt, "18432_count", "FOXP3CD3_count")
setnames(dt, "17408_count", "FOXP3CD4_count")
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
return(dt)
}))
# Reshape the data to a long format and remove '_count' from Cell_Type
long_data <- cell_counts_by_type %>%
select(Slide_ID, AMCAR_count, HMWCK_count, SMA_count, CD3d_count, CD8a_count, CD4_count, PD1_count, FOXP3_count, TCF1_count, CD68_count, CD163_count, CD20_count, CD11b_count, CD57_count) %>%
gather(key = "Cell_Type", value = "Count", -Slide_ID) %>%
mutate(Cell_Type = gsub("_count", "", Cell_Type))
# Calculate total cells for each Slide_ID
total_cells <- long_data %>%
group_by(Slide_ID) %>%
summarize(Total_Count = sum(Count))
## merge in the clinical data
long_data <- merge(data.table::copy(rr)[,.(Slide_ID, Gleason)], long_data, by="Slide_ID", all.x=TRUE)
# Merge total cells with long data
long_data <- merge(long_data, total_cells, by = "Slide_ID")
## order the slides, first by Gleason then by total count
ordered_slide_ids <- long_data %>%
arrange(Gleason, Total_Count) %>%
.$Slide_ID %>%
unique()
long_data$Slide_ID <- factor(long_data$Slide_ID, levels = ordered_slide_ids)
# Define the order of cell types within each color group
cell_order <- c("AMCAR", "HMWCK", "SMA", # Group A: Greens
"CD3d", "CD8a", "CD4", "PD1", "FOXP3", "TCF1", # Group B: Blues
"CD68", "CD163", "CD20", "CD11b","CD57") # Group C: Reds
# Ensure long_data Cell_Type is in the specified order
long_data$Cell_Type <- factor(long_data$Cell_Type, levels = cell_order)
long_data <- as.data.table(long_data)
# Define color palettes for each group
colors <- c(rev(brewer.pal(3, "Greens")[1:3]),
brewer.pal(6, "Blues")[1:6],
brewer.pal(5, "Reds")[1:5])
# Mapping of cell types to colors
color_map <- setNames(colors, c("AMCAR", "HMWCK", "SMA",
"CD3d", "CD8a", "CD4", "PD1", "FOXP3", "TCF1",
"CD68", "CD163", "CD20", "CD11b","CD57"))
# Create the plot
g_cellcounts <- ggplot(long_data, aes(x = Slide_ID, y = Count/1e6, fill = Cell_Type)) +
geom_bar(stat = "identity", position = "stack") +
scale_fill_manual(values = color_map) +
coord_flip() +
#facet_wrap(~Gleason, scales="free_y") +
labs(x = "Slide ID", y = "Cell count (millions)") +
theme_jw() +
theme(#legend.position = c(0.65, 0.3),
#legend.justification = c(-0.2, -0.3), # Coordinates relative to the plot area
#legend.box.margin = margin(6, 6, 6, 6), # Adjust spacing around the legend box if necessary
#legend.margin = margin(-10, -10, -10, -10),
text = element_text(family = "Arial"),
axis.text.x = element_text(color = "black"),
axis.text.y = element_text(color = "black"),
axis.title.y = element_text(size = 9, margin = margin(r = 5)),
axis.title.x = element_text(size = 9, margin = margin(t = 2)),
plot.title = element_text(size = 9),
legend.title = element_blank(),
#legend.spacing.y = unit(1, "cm"),
legend.text = element_text(size = 9),
legend.background = element_blank(),
legend.key = element_blank(),
axis.line = element_line(colour = "black"), # Add axis lines
panel.background = element_rect(fill = "transparent", colour = NA), # remove panel background
plot.background = element_rect(fill = "transparent", colour = NA)
) +
#scale_y_log10(breaks = 10^seq(4), labels = trans_format("log10", scales::math_format(10^.x)))
scale_y_continuous(expand = c(0, 0))
save_ppdf(g, "cell_counts_color_bar.pdf", width=5, height=4)
# Calculate the fractions
fraction_data <- long_data %>%
group_by(Slide_ID, Gleason) %>%
mutate(Total = sum(Count)) %>%
ungroup() %>%
mutate(Fraction = Count / Total)
# Your ggplot code with fractions
g <- ggplot(fraction_data, aes(x = Slide_ID, y = Fraction, fill = Cell_Type)) +
geom_bar(stat = "identity", position = "stack") +
scale_fill_manual(values = color_map) +
coord_flip() +
facet_wrap(~Gleason, scales = "free_y") +
labs(title = "Cell Type Fractions by Slide ID and Cell Type", x = "Slide ID", y = "Fraction") +
theme_jw()
dtt <- lapply(files, function(f) {
dt <- fread(cmd=paste("cyftools count",f))
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
return(dt)
})
dtr <- rbindlist(dtt)
dtr[, Slide_ID := factor(Slide_ID, levels = Slide_ID[order(V1)])]
g <- ggplot(dtr, aes(x = Slide_ID, y = V1 / 1000)) +
geom_bar(stat = "identity") +
coord_flip() + # Makes the bar chart horizontal
labs(x = "Slide_ID",
y = "Total cell count (thousands)") + theme_jw()
save_ppdf(g, "cell_counts_bar.pdf", width=5, height=5)
```
```{r cys-read, include=FALSE}
# read all the data as the means within the tumor compartment
nn <- c("tumor"="-A 8", "stroma"="-N 8","all"="", "scramble_tumor"="-A 8", "scramble_stroma"="-N 8","tumor_margin"="-A 40","stroma_margin"="-A 32 -N 8","scramble_tumor_margin"="-A 40", "scramble_stroma_margin"="-A 32 -N 8","stroma_core"="-N 12", "tumor_core"="-A 8 -N 4")
nn <- c("tumor"="-A 8", "stroma"="-N 8","all"="","tumor_margin"="-A 40", "tumor_core"="-A 8 -N 4")
if (FALSE)
lapply(seq_along(nn), function(n) {
cat("working on", names(nn)[n],"\n")
dtt <- parallel::mclapply(master_files, function(f) {
#cat(basename(f), " - ", names(nn)[n], "\n")
tmpfile <- paste0("/tmp/", generate_random_string(8))
if (grepl("scramble", names(nn)[n])) {
ccmd=paste("cyftools scramble",f, "- | cyftools filter -", nn[n], tmpfile)
} else {
ccmd=paste("cyftools filter",f, nn[n],tmpfile)
}
system(ccmd)
dt <- fread(cmd=paste("cyftools mean", tmpfile, " - | cyftools view - -R"))
dt[ , allcount := fread(cmd=paste("cyftools count", tmpfile))$V1]
# indivdual cell counts
drr = fread(cmd=paste("cyftools cellcount", tmpfile," - -a 34816 -a 33792 -a 65540 -a 49152 -a 0 -a 20480 -a 18432 -a 17408 -a 6144 | cyftools view - -R"))
dt[, amcar_count := drr$AMCAR_count]
dt[, hmwck_count := drr$HMWCK_count]
dt[, cd11c_count := drr$CD11c_count]
dt[, cd68count := drr$CD68_count]
dt[ ,cd4count. := drr$CD4_count]
dt[, foxp3count:= drr$FOXP3_count]
dt[, cd8count := drr$CD8a_count]
dt[, cd20count := drr$CD20_count]
dt[, cd3count := drr$CD3d_count]
dt[, cd163count := drr$CD163_count]
dt[, pd1count := drr$`PD1_count`]
dt[, smacount := drr$SMA_count]
dt[ , cd3pd1count := drr$`34816_count`]
dt[ , cd4pd1count := drr$`33792_count`]
dt[ , uncounted := drr$`0_count`]
dt[ , AMACRCD57 := drr$`65540_count`]
dt[ , FOXP3PD1 := drr$`49152_count`]
dt[, FOXP3CD8 := drr$`20480_count`]
dt[, FOXP3CD3 := drr$`18432_count`]
dt[, FOXP3CD4 := drr$`17408_count`]
dt[, cd3cd8count := drr$`6144_count`]
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
dt[, margin := names(nn)[n]]
file.remove(tmpfile)
return(dt)
}, mc.cores=8)
dtr <- rbindlist(dtt, fill=TRUE)
dtr <- dtr[!duplicated(paste(margin, Slide_ID))]
saveRDS(dtr, paste0("~/Dropbox/Sorger/cache/240929_prostate_",names(nn)[n], ".rds"), compress=FALSE)
})
dtr <- rbindlist(lapply(paste0("~/Dropbox/Sorger/cache/240929_prostate_",names(nn),".rds"), readRDS), fill=TRUE)
dtr <- dtr[!duplicated(paste(margin, Slide_ID))]
rr <- merge(data.table::copy(crr), dtr, by = "Slide_ID", all.x = TRUE)
#fwrite(rr, "~/Dropbox/Sorger/projects/prostate/240806_prostate_data.csv", sep=",", quote=FALSE)
rr[, CD8_n := cd8count / allcount]
rr[, CD3_n := cd3count / allcount]
rr[, FOXP3PD1_n := FOXP3PD1 / allcount]
rr[, PD1_n := pd1count / allcount]
rr[, FOXP3_n := foxp3count / allcount]
rr[, FOXP3CD8_n := FOXP3CD8 / allcount]
rr[, FOXP3CD3_n := FOXP3CD3 / allcount]
rr[, FOXP3CD4_n := FOXP3CD4 / allcount]
rr[, CD163_n := cd163count / allcount]
rr[, AMCAR_n := amcar_count / allcount]
rr[, AMACRCD57_n := AMACRCD57 / allcount]
```
```{r QC}
dt <- fread(cmd=paste("cyftools filter -a 2048 -o -a 4096 ~/Dropbox/Sorger/projects/prostate/chain_flip_cat.cyf - | cyftools clean -M - - | cyftools clean -m - - | cyftools view - -R"))
dt[, CD8 := ifelse(bitops::bitAnd(pflag, 4096)==4096, 1, 0)]
dt[, CD3 := ifelse(bitops::bitAnd(pflag, 2048)==2048, 1, 0)]
dt[, SMA := ifelse(bitops::bitAnd(pflag, 32)==32, 1, 0)]
dt[, CD4 := ifelse(bitops::bitAnd(pflag, 1024)==1024, 1, 0)]
cd8_only <- nrow(dt[CD4 == 1 & CD3 == 0])
cd3_only <- nrow(dt[CD4 == 0 & CD3 == 1])
both_cd8_cd3 <- nrow(dt[CD4 == 1 & CD3 == 1])
# Create the Venn diagram
venn.plot <- VennDiagram::draw.pairwise.venn(
area1 = cd8_only + both_cd8_cd3,
area2 = cd3_only + both_cd8_cd3,
cross.area = both_cd8_cd3,
category = c("CD8", "CD3"),
scaled = TRUE,
fill = c("skyblue", "pink"),
lty = "blank",
cex = 2,
cat.cex = 2,
cat.pos = c(-20, 20),
cat.dist = c(0.05, 0.05),
cat.just = list(c(-1, -1), c(1, 1))
)
grid.draw(venn.plot)
# Count the number of occurrences for each category
cd8_only <- nrow(dt[CD8 == 1 & CD3 == 0 & CD4 == 0])
cd3_only <- nrow(dt[CD8 == 0 & CD3 == 1 & CD4 == 0])
cd4_only <- nrow(dt[CD8 == 0 & CD3 == 0 & CD4 == 1])
cd8_cd3 <- nrow(dt[CD8 == 1 & CD3 == 1 & CD4 == 0])
cd8_cd4 <- nrow(dt[CD8 == 1 & CD3 == 0 & CD4 == 1])
cd3_cd4 <- nrow(dt[CD8 == 0 & CD3 == 1 & CD4 == 1])
cd8_cd3_cd4 <- nrow(dt[CD8 == 1 & CD3 == 1 & CD4 == 1])
# Create the Venn diagram
venn.plot <- draw.triple.venn(
area1 = cd8_only + cd8_cd3 + cd8_cd4 + cd8_cd3_cd4,
area2 = cd3_only + cd8_cd3 + cd3_cd4 + cd8_cd3_cd4,
area3 = cd4_only + cd8_cd4 + cd3_cd4 + cd8_cd3_cd4,
n12 = cd8_cd3 + cd8_cd3_cd4,
n13 = cd8_cd4 + cd8_cd3_cd4,
n23 = cd3_cd4 + cd8_cd3_cd4,
n123 = cd8_cd3_cd4,
category = c("CD8", "CD3", "CD4"),
fill = c("skyblue", "pink", "lightgreen"),
lty = "blank",
cex = 2,
cat.cex = 2,
cat.col = c("skyblue", "pink", "lightgreen"),
cat.pos = c(-20, 20, 160),
cat.dist = c(0.05, 0.05, 0.05),
cat.just = list(c(-1, -1), c(1, 1), c(1, 1))
)
```
```{r whole-slide-bees}
rrt = rr[margin=="all"]
rrm <- melt(rrt, id.vars=c("Slide_ID","Gleason"), measure.vars=c("CD8a_50r","CD3d_50r", "CD4_50r","CD20_50r","CD68_50r","CD163_50r"))
g <- create_beeswarm_plot(rrm, x_var="Gleason", y_var="value", withbox=TRUE) +
facet_wrap(~ variable, scale="free") +
scale_fill_manual(values=gleason_colors)
```
```{r tls-counts}
mfiles <- list.files(path = "~/Sorger/projects/prostate/tls/", pattern = "cyf$", full.names = TRUE)
drr <- parallel::mclapply(mfiles, function(f) {
dt <- fread(cmd=paste("cyftools view",f,"-x tls_id | cut -d',' -f 7 | sort -n | uniq | wc -l"))
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
}, mc.cores=8)
tls <- rbindlist(drr)
setnames(tls, "V1","TLS_count")
tls[, TLS_count := TLS_count - 1] ## 0 is not a TLS
rr <- merge(rr, tls, by="Slide_ID", all.x=TRUE)
```
```{r cd-waterfall}
rrt = rr[margin=="all"] ## & !Slide_ID %in% c("LSP12645","LSP12643","LSP12631")]
# "ratio" is not good term, but it's the field we'll be looking at for this plot
rrt[, ratio := CD3d_50r]
setkeyv(rrt, "ratio")
clevels <- rrt$Slide_ID
rrt[, Slide_ID := factor(Slide_ID, levels=clevels)]
rrt[, samplecount := seq(.N)]
g <- ggplot(rrt, aes(x=Slide_ID, y=ratio, fill=Gleason)) +
geom_bar(stat="identity", color="black") + ylab("Cells / mm2") + xlab("Sample ID") +
coord_flip() + scale_fill_manual(values=gleason_colors) + theme_jw() +
theme(axis.line = element_line(), axis.ticks = element_line()) +
theme(legend.position = c(0.65, 0.3),
legend.justification = c(-0.2, -0.3), # Coordinates relative to the plot area
legend.box.margin = margin(6, 6, 6, 6), # Adjust spacing around the legend box if necessary
legend.margin = margin(-10, -10, -10, -10),
text = element_text(family = "Helvetica"),
axis.text.x = element_text(color = "black"),
axis.text.y = element_text(color = "black"),
axis.title.y = element_text(size = 9, margin = margin(r = 5)),
axis.title.x = element_text(size = 9, margin = margin(t = 2)),
plot.title = element_text(size = 9),
legend.title = element_blank(),
#legend.spacing.y = unit(1, "cm"),
legend.text = element_text(size = 9),
legend.background = element_blank(),
legend.key = element_blank(),
axis.line = element_line(colour = "black"), # Add axis lines
panel.background = element_rect(fill = "transparent", colour = NA), # remove panel background
plot.background = element_rect(fill = "transparent", colour = NA)
) +
#scale_x_continuous(expand=c(0,0), breaks=c(0,10,20,30,40,50,60,70)) +
scale_y_continuous(expand=c(0,0))
save_ppdf(g, "cd8_waterfall_whole_slide.pdf", width=5, height=4)
```
```{r compare-compartments-counts}
tumorn="tumor_core"
stroman="stroma_core"
tumorns="stroma_core"
stromans="tumor_core"
# Reshape the data to long format for ggplot
long_data <- as.data.table(rr %>%
gather(key = "Cell_Type", value = "Value", CD8a_50r, CD3d_50r, PD1_50r, FOXP3_50r, AMCAR_50r, CD163_50r,
CD8_n, CD3_n, FOXP3PD1_n,PD1_n, FOXP3_n, AMCAR_n, CD163_n,AMACRCD57_n) %>%
mutate(Gleason = factor(Gleason, levels = c("Low", "High"))))
# Calculate the ratio (tumor)
ratio_data <- long_data[margin %in% c(tumorn, tumorns), .(
tumor_normalized = Value[margin == tumorn] / Value[margin == tumorns]
), by = .(Cell_Type, Slide_ID)
]
# Join ratio_data back to long_data for rows where margin is "A8"
new_rows <- long_data[margin == tumorn, on = .(Cell_Type, Slide_ID)]
# Merge the calculated ratios into new_rows
new_rows <- merge(new_rows, ratio_data, by = c("Cell_Type", "Slide_ID"))
# Update the margin and Value columns in new_rows
new_rows[, `:=`(margin = "tumor_normalized", Value = tumor_normalized)]
# Calculate the ratio (stroma)
ratio_data <- long_data[margin %in% c(stroman, stromans), .(
stroma_normalized = Value[margin == stroman] / Value[margin == stromans]
), by = .(Cell_Type, Slide_ID)
]
# Join ratio_data back to long_data for rows where margin is "A8"
new_rows_s <- long_data[margin == stroman, on = .(Cell_Type, Slide_ID)]
# Merge the calculated ratios into new_rows
new_rows_s <- merge(new_rows, ratio_data, by = c("Cell_Type", "Slide_ID"))
# Update the margin and Value columns in new_rows
new_rows_s[, `:=`(margin = "stroma_normalized", Value = stroma_normalized)]
# Add the new rows to the original data.table
long_data <- rbind(long_data, new_rows, new_rows_s, fill = TRUE)
long_data <- long_data[margin %in% c(tumorn,"all",stroman,"tumor_normalized","stroma_normalized")]
# Create an interaction variable for margin and Gleason
long_data$Margin_Gleason <- with(long_data, factor(interaction(margin, Gleason, sep = "-"),
levels = c("all-Low", "all-High",
paste0(tumorn, c("-Low","-High")),
paste0(stroman,c("-Low","-High")),
paste0("tumor_normalized",c("-Low","-High")),
paste0("stroma_normalized",c("-Low","-High")))))
long_data <- long_data[!(Slide_ID == "LSP12657" & Cell_Type =="CD8a_20r") &
!(Slide_ID == "LSP12629" & Cell_Type =="CD20_20r") &
!(Slide_ID == "LSP12629" & Cell_Type =="CD163_20r")]
# Define colors for each margin and Gleason level
# Define the main categories and their low and high variants
categories <- c(tumorn, "all",stroman, "tumor_normalized","stroma_normalized")
variants <- c("Low", "High")
# Create a diverging color palette for each main category
# Adjust the number of colors based on your specific needs
palette_A8 <- brewer.pal(3, "Reds")[c(1,3)]
palette_all <- brewer.pal(3, "Greens")[c(1,3)]
palette_N8 <- brewer.pal(3, "Blues")[c(1,3)]
palette_scramble <- brewer.pal(3, "Purples")[c(1,3)]
palette_stroma_scramble <- brewer.pal(3, "Greys")[c(1,3)]
# Combine the palettes
colors <- c(palette_A8, palette_all, palette_N8, palette_scramble, palette_stroma_scramble)
# Construct the final color mapping
margin_colors <- setNames(colors, paste0(rep(categories, each = length(variants)), "-", variants))
# Extract the corresponding names, which are the LSP numbers
lsp_numbers = names(master_files)
g <- ggplot(long_data[Cell_Type %in%
c("CD3_n","CD8_n","FOXP3_n","PD1_n","AMCAR_n","CD163_n","AMACRCD57_n") &
#c("CD8a_50r", "CD3d_50r", "PD1_50r", "FOXP3_50r", "AMCAR_50r", "CD163_50r") &
!Slide_ID %in% c("LSP12631","LSP12643","LSP12645") &
grepl("normalized", Margin_Gleason)], aes(x = Margin_Gleason, y = Value, fill = Margin_Gleason)) +
geom_boxplot(position = position_dodge2(width = 0.8, preserve = "single")) +
scale_fill_manual(values = margin_colors) +
facet_wrap(~ Cell_Type, nrow=1) +
scale_y_log10() +
labs(title = "Boxplots by Cell Type, Margin, and Gleason", x = "Margin - Gleason", y = "Fold-increase") +
# coord_cartesian(ylim = c(0, 3.5)) + # Set y-axis limits
theme_jw() +
theme(axis.text.x = element_text(angle = 90, vjust = 1, hjust = 1))+
scale_x_discrete(labels = function(x) {
ifelse(x == "tumor_normalized−Low", "Tumor-Low",
ifelse(x == "tumor_normalized−High", "Tumor-High",
ifelse(x == "stroma_normalized−Low", "Stroma-Low",
ifelse(x == "stroma_normalized−High", "Stroma-High", x))))
}) +
geom_hline(aes(yintercept=1))
dtw <- rbindlist(lapply(c("tumor_normalized","stroma_normalized","all","tumor_margin","stroma_margin"), function(mn) {
rbindlist(lapply(c("CD8a_50r", "CD3d_50r", "PD1_50r", "FOXP3_50r", "AMCAR_50r", "CD163_50r"), function(cell) {
a = long_data[Cell_Type == cell & !Slide_ID %in% c("LSP12631","LSP12643","LSP12645") & Gleason == "High" & margin==mn, Value]
b = long_data[Cell_Type == cell & !Slide_ID %in% c("LSP12631","LSP12643","LSP12645") & Gleason == "Low" & margin==mn, Value]
data.table(p=wilcox.test(a,b)$p.value, margin=mn, cell=cell)
}))
}))
dtw <- rbindlist(lapply(c("tumor_normalized","stroma_normalized","all","tumor_core","stroma_core"), function(mn) {
rbindlist(lapply(c("AMACRCD57_n","AMCAR_n","CD163_n","CD3_n","CD8_n","FOXP3_n","PD1_n"), function(cell) {
a = long_data[Cell_Type == cell & !Slide_ID %in% c("LSP12631","LSP12643","LSP12645") & Gleason == "High" & margin==mn, Value]
b = long_data[Cell_Type == cell & !Slide_ID %in% c("LSP12631","LSP12643","LSP12645") & Gleason == "Low" & margin==mn, Value]
data.table(p=t.test(a,b)$p.value, margin=mn, cell=cell)
}))
}))
save_ppdf(g, "tumor_v_stroma_boxplots_tumor.pdf", width=8, height=4)
```
```{r adaptive-beeswarm}
rrt = rr[margin=="all"]
tcol <- c("CD4","CD3d","CD8a","FOXP3","CD20")
rrp <- rrt[, .SD, .SDcols = c(paste0(tcol,"_50r"), "Gleason","Slide_ID")]
rrm <- data.table::melt(rrp, id.vars = c("Gleason","Slide_ID"))
rrm <- rrm[!(Slide_ID == "LSP12657" & variable=="CD8a_50r")][!(Slide_ID == "LSP12629" & variable=="CD20_50r")][!(Slide_ID == "LSP12629" & variable=="CD163_50r")]
g <- create_beeswarm_plot(rrm, x="Gleason", y="value", withbox=TRUE) +
facet_wrap(rrm$variable, scales="free", nrow=1) +
scale_y_continuous(expand = expansion(mult = c(0.05, 0.1))) +
theme(
strip.background = element_blank(),
strip.text = element_text(face = "bold", size = 10)
) + ylab("Cell Density (cells/mm2)") +
theme(text = element_text(family = "Helvetica"),
axis.text.x = element_text(color = "black"),
axis.text.y = element_text(color = "black"),
axis.title.y = element_text(size = 9, margin = margin(r = 5)),
axis.title.x = element_text(size = 9, margin = margin(t = 2)),
plot.title = element_text(size = 9),
panel.background = element_rect(fill = "transparent", colour = NA), # remove panel background
plot.background = element_rect(fill = "transparent", colour = NA)
)
save_ppdf(g, "innate_beeswarm.pdf", width=6, height=1.8)
```
```{r cd57-beeswarm}
rrt = rr[margin=="stroma",]
tcol <- c("FOXP3CD3_n","FOXP3CD8_n","FOXP3CD4_n")
#rrp <- rrt[, .SD, .SDcols = c(paste0(tcol,"_50r"), "Gleason","Slide_ID")]
rrp <- rrt[, .SD, .SDcols = c(tcol, "Gleason","Slide_ID")]
rrm <- data.table::melt(rrp, id.vars = c("Gleason","Slide_ID"))
rrm <- rrm[!(Slide_ID == "LSP12657" & variable=="CD8a_50r")][!(Slide_ID == "LSP12629" & variable=="CD20_50r")][!(Slide_ID == "LSP12629" & variable=="CD163_50r")]
g <- create_beeswarm_plot(rrm, x="Gleason", y="value", withbox=TRUE) +
facet_wrap(rrm$variable, scales="free", nrow=1) +
scale_y_continuous(expand = expansion(mult = c(0.05, 0.1))) +
theme(
strip.background = element_blank(),
strip.text = element_text(face = "bold", size = 10)
) + ylab("Cell Density (cells/mm2)") +
theme(text = element_text(family = "Helvetica"),
axis.text.x = element_text(color = "black"),
axis.text.y = element_text(color = "black"),
axis.title.y = element_text(size = 9, margin = margin(r = 5)),
axis.title.x = element_text(size = 9, margin = margin(t = 2)),
plot.title = element_text(size = 9),
panel.background = element_rect(fill = "transparent", colour = NA), # remove panel background
plot.background = element_rect(fill = "transparent", colour = NA)
)
save_ppdf(g, "cd57amcar_beeswarm.pdf", width=2, height=1.8)
```
```{r foxp3-beeswarm}
rrt = rr[margin=="all",]
tcol <- c("FOXP3","PD1","CD163")
rrp <- rrt[, .SD, .SDcols = c(paste0(tcol,"_50r"), "Gleason","Slide_ID")]
#rrp <- rrt[, .SD, .SDcols = c(tcol, "Gleason","Slide_ID")]
rrm <- data.table::melt(rrp, id.vars = c("Gleason","Slide_ID"))
rrm <- rrm[!(Slide_ID == "LSP12657" & variable=="CD8a_50r")][!(Slide_ID == "LSP12629" & variable=="CD20_50r")][!(Slide_ID == "LSP12629" & variable=="CD163_50r")]
g <- create_beeswarm_plot(rrm, x="Gleason", y="value", withbox=TRUE) +
facet_wrap(rrm$variable, scales="free", nrow=1) +
scale_y_continuous(expand = expansion(mult = c(0.05, 0.1))) +
theme(
strip.background = element_blank(),
strip.text = element_text(face = "bold", size = 10)
) + ylab("Cell Density (cells/mm2)") +
theme(text = element_text(family = "Helvetica"),
axis.text.x = element_text(color = "black"),
axis.text.y = element_text(color = "black"),
axis.title.y = element_text(size = 9, margin = margin(r = 5)),
axis.title.x = element_text(size = 9, margin = margin(t = 2)),
plot.title = element_text(size = 9),
panel.background = element_rect(fill = "transparent", colour = NA), # remove panel background
plot.background = element_rect(fill = "transparent", colour = NA)
)
save_ppdf(g, "innate_beeswarm_stroma.pdf", width=4, height=1.8)
```
```{r tls-pd1}
rrt = rr[margin=="all",]
g <- ggplot(rrt, aes(x=SMA_50r, y=TLS_count, color=Gleason, fill=Gleason)) +
theme_jw() + geom_point() +
geom_smooth(method = "lm", se = TRUE) + # Add linear regression line for each group
scale_color_manual(values=gleason_colors) +
scale_fill_manual(values=gleason_colors) +
xlab("CD3+PD1+ cells/mm2") + ylab("Tertiary Lymphoid Structures (count per sample)")
save_ppdf(g, "tls_pd1_correlation.pdf", width=3, height=3)
```
```{r compare-compartments}
# Reshape the data to long format for ggplot
long_data <- as.data.table(rr[!is.na(margin) & margin %in% c("all","A8","N8","tumor_margin","tumor_stroma")] %>%
gather(key = "Cell_Type", value = "Value", CD3d_20r, CD20_20r, CD8a_20r,
AMCAR_20r, CD57_20r, CD163_20r, PD1_20r, CD8a_20r, FOXP3_20r, HLAA_20r, HLADR_20r,
CD8_n, CD3_n, FOXP3PD1_n) %>%
mutate(Gleason = factor(Gleason, levels = c("Low", "High"))))
# Create an interaction variable for margin and Gleason
long_data$Margin_Gleason <- with(long_data, factor(interaction(margin, Gleason, sep = "-"),
levels = c("all-Low", "all-High", "A8-Low", "A8-High","N8-Low", "N8-High","tumor_stroma-Low","tumor_stroma-High","tumor_margin-Low","tumor_margin-High")
))
long_data <- long_data[!(Slide_ID == "LSP12657" & Cell_Type =="CD8a_20r") &
!(Slide_ID == "LSP12629" & Cell_Type =="CD20_20r") &
!(Slide_ID == "LSP12629" & Cell_Type =="CD163_20r")]
# Define colors for each margin and Gleason level
# Define the main categories and their low and high variants
categories <- c("A8", "all", "N8","tumor_stroma","tumor_margin")
variants <- c("Low", "High")
# Create a diverging color palette for each main category
# Adjust the number of colors based on your specific needs
palette_A8 <- brewer.pal(3, "Reds")[c(1,3)]
palette_all <- brewer.pal(3, "Greens")[c(1,3)]
palette_N8 <- brewer.pal(3, "Blues")[c(1,3)]
palette_ts <- brewer.pal(3, "Purples")[c(1,3)]
palette_tm <- brewer.pal(3, "Greys")[c(1,3)]
# Combine the palettes
colors <- c(palette_A8, palette_all, palette_N8, palette_ts, palette_tm)
# Construct the final color mapping
margin_colors <- setNames(colors, paste0(rep(categories, each = length(variants)), "-", variants))
g <- ggplot(long_data[Cell_Type %in% c("CD8a_20r")], aes(x = Margin_Gleason, y = Value, fill = Margin_Gleason)) +
geom_boxplot(position = position_dodge2(width = 0.8, preserve = "single")) +
scale_fill_manual(values = margin_colors) +
facet_wrap(~ Cell_Type, nrow=1) +
labs(title = "Boxplots by Cell Type, Margin, and Gleason", x = "Margin - Gleason", y = "Cells / mm2") +
coord_cartesian(ylim = c(0, 2300)) + # Set y-axis limits
theme_jw() +
theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1))
# calculate the wilcox p values
variables <- c("FOXP3_20r", "CD8a_20r", "PD1_20r", "CD163_20r","CD4_20r","CD103_20r",
"CD20_20r")
margins <- c("all", "N8", "A8")
rr_subset <- rr[!is.na(margin) & Slide_ID != "LSP12629",]
results <- rbindlist(lapply(variables, function(A) {
rbindlist(lapply(margins, function(B) {
data.table(
Cell = sub("_20r", "", A),
Compartment = c("all"="Whole slide", "N8"="Stroma","A8"="Tumor")[B],
p_value = t.test(
rr_subset[margin == B & Gleason == "Low", get(A)],
rr_subset[margin == B & Gleason == "High", get(A)]
)$p.value,
LG_median = round(rr_subset[margin == B & Gleason == "Low", mean(get(A), na.rm = TRUE)])/1000,
HG_median = round(rr_subset[margin == B & Gleason == "High", mean(get(A), na.rm = TRUE)])/1000
)
}))
}))
results <- results[, c("Cell", "Compartment", "LG_median", "HG_median", "p_value")]
digits <- rep(2, ncol(results)+1)
digits[6] <- 4
latex_table <- xtable(results, digits = digits)
print(latex_table, include.rownames = FALSE, hline.after = c(-1, 0, nrow(data)))
# Write to CSV
fwrite(results, file = "~/Sorger/projects/prostate/formatted_results.csv")
save_ppdf(g, "boxplot_region_compar.pdf", width=8, height=8)
```
```{r 20r-boxplots}
rr_long <- melt(rr, id.vars = "Gleason", measure.vars = c("FOXP3_20r","HLAA_20r"), variable.name = "Metric", value.name = "Value")
rr_long[, cells_per_mm2 := Value * 1e6 / (1000^2 * 0.325^2)]
# Assuming the data has two Gleason groups "A" and "B"
get_pvalue <- function(data){
wilcox.test(data[data$Gleason == "High",]$cells_per_mm2,
data[data$Gleason == "Low",]$cells_per_mm2)$p.value
}
p_values <- by(rr_long, rr_long$Metric, get_pvalue)
labels_with_pvalues <- sapply(names(p_values), function(name) {
sprintf("%s (p = %.3f)", name, p_values[[name]])
})
g <- ggplot(rr_long, aes(x = Gleason, y = cells_per_mm2 / 1000, fill = Gleason)) +
geom_boxplot(alpha = 0.5, width=0.8, outlier.shape=NA) +
geom_jitter(position = position_jitter(width = 0.2), size = 1, alpha = 0.7, shape=20) +
labs(y = "Thousands of cells / mm2") +
#scale_fill_manual(values = pMMR_colors) +
facet_wrap(~ Metric, labeller = as_labeller(labels_with_pvalues)) +
theme_jw() + scale_y_log10()
save_ppdf(g, "whole_slide_immune_boxplot2.pdf", width=6, height=4)
```
```{r clinical_plots}
g <- ggplot(rr, aes(x = Gleason, y = age_surgery_years, fill = Gleason)) +
geom_boxplot(alpha = 0.5, width=0.8, outlier.shape=NA) +
geom_jitter(position = position_jitter(width = 0.2), size = 1, alpha = 0.7, shape=20) +
labs(y = "Age at surgery (years)") +
theme_jw()
wilcox.test(rr[Gleason=="High",time_to_progression_days], rr[Gleason=="Low",time_to_progression_days])
save_ppdf(g, "age_boxplot.pdf", width=3, height=3)
```
```{r cohort-statistic}
system(paste("cyftools cellcount ~/Sorger/projects/prostate/allp.cys /tmp/tmp.cys -a 34816 -a 0"), intern=TRUE)
headr <- header_read("/tmp/tmp.cys")
dt <- fread(cmd=paste("cyftools view /tmp/tmp.cys"))
colnames(dt) <- c("sid","cid","cflag","pflag","x","y",headr[tag_type %in% c("CA","MA"), id])
# Gather the count fields to long format
dt_long <- dt[, .(Variable = names(.SD), Count = unlist(.SD)),
.SDcols = patterns("_count")]
dt_long$Variable <- gsub("_count$", "", dt_long$Variable)
dt_long$Variable[dt_long$Variable=="0"] <- "Uncalled"
dt_long$Variable[dt_long$Variable=="34816"] <- "CD3+PD1+"
# Plot using ggplot2
g <- ggplot(dt_long, aes(x = Variable, y = Count / 29 / 1000)) +
geom_bar(stat = "identity") +
coord_flip() + # Flip to make horizontal bars
theme_minimal() +
labs(x = "", y = "Cell counts / tumor (thousands)") +
scale_y_continuous(labels = scales::comma) + # Use comma format for y-axis labels
theme_jw()
save_ppdf(g, "cell_counts_hist.pdf", width=6, height=6)
```
```{r umap}
dt <- fread("cyftools clean -M ~/Sorger/projects/prostate/allp.02.cys - | cyftools rescale - - | cyftools view - -R")[,.(Ki67, AMCAR, HMWCK, CD19, SMA, CD20, CD68, CD163, CD4, CD3d, CD8a, FOXP3, PD1, CD57, CD11c, GranzymeB, CD15, HLADR, CD103, CD31, pTBK1, HLAA, CD44, CD206)]
expm <- t(as.matrix(dt[sample(.N, 99000)]))
colnames(expm) <- seq(ncol(expm))
seurat_obj <- CreateSeuratObject(counts = expm)
# pre-process
seurat_obj <- NormalizeData(seurat_obj)
#seurat_obj <- FindVariableFeatures(seurat_obj, selection.method = "vst", nfeatures = 2000)
seurat_obj <- ScaleData(seurat_obj)
seurat_obj <- RunPCA(seurat_obj, verbose = FALSE, npcs = min(10, ncol(seurat_obj) - 1), features = rownames(seurat_obj))
ElbowPlot(seurat_obj)
# cluster using Louvain
seurat_obj <- FindNeighbors(seurat_obj, dims = 1:10) # you can choose an appropriate number of dims based on ElbowPlot
seurat_obj <- FindClusters(seurat_obj, resolution = 0.5) # adjust resolution based on the desired granularity
saveRDS(seurat_obj, "~/Sorger/cache/seurat.rds", compress=FALSE)
# run umap
seurat_obj <- RunUMAP(seurat_obj, dims = 1:10, n_neighbors = 30, min_dist = 0.5)
DimPlot(seurat_obj, group.by = "seurat_clusters")
```
```{r macrophage-tcell}
files <- list.files(path = "~/Sorger/projects/prostate/chain/", full.names = TRUE)
files <- files[grepl("cys$", files)]
dtt <- lapply(files, function(f) {
cat(basename(f), "\n")
dt <- fread(cmd=paste("cyftools filter -a 512 -a 256 -T",f,"/tmp/tmp.cys && cyftools mean /tmp/tmp.cys - | cyftools view - -R"))
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
dt[, margin := "cd163"]
return(dt)
})
dtr <- rbindlist(dtt)
dtt <- lapply(files, function(f) {
cat(basename(f), "\n")
dt <- fread(cmd=paste("cyftools filter -n 768 -T",f,"/tmp/tmp.cys && cyftools mean /tmp/tmp.cys - | cyftools view - -R"))
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
dt[, margin := "not_cd163"]
return(dt)
})
dtr <- rbind(dtr, rbindlist(dtt))
rrm <- fread("~/Sorger/projects/prostate/230803_clinical_data.csv")
rrm <- rrm[!duplicated(Slide_ID)]
rrm[, Gleason := ifelse(surgery_gleason %in% c("4 + 4","4 + 5","5 + 3","5 + 4"), "High", "Low")]
rrm <- merge(data.table::copy(rrm), dtr, by = "Slide_ID", all.x = TRUE)
rr_long <- melt(rrm, id.vars = c("Gleason","margin"), measure.vars = c("CD3d_20r","CD4_20r","CD8a_20r","PD1_20r"), variable.name = "Metric", value.name = "Value")
rr_long[, cells_per_mm2 := Value * 1e6 / (1000^2 * 0.325^2)]
p_values <- by(rr_long, rr_long$Metric, get_pvalue)
labels_with_pvalues <- sapply(names(p_values), function(name) {
sprintf("%s (p = %.3f)", name, p_values[[name]])
})
g <- ggplot(rr_long, aes(x = margin, y = cells_per_mm2 / 1000, fill = Gleason)) +
geom_boxplot(alpha = 0.5, width=0.8, outlier.shape=NA) +
geom_jitter(position = position_jitter(width = 0.2), size = 1, alpha = 0.7, shape=20) +
labs(y = "Thousands of cells / mm2") +
#scale_fill_manual(values = pMMR_colors) +
facet_wrap(~ Metric, labeller = as_labeller(labels_with_pvalues)) +
theme_jw() + scale_y_log10()
```
```{r lda}
files <- list.files(path = "~/Sorger/projects/prostate/lda", full.names = TRUE)
files <- files[grepl("cys$", files)]
dtt <- lapply(files, function(f) {
cat(basename(f), "\n")
dt <- fread(cmd=paste("cyftools filter",f,"/tmp/tmp.cys && cyftools mean /tmp/tmp.cys - | cyftools view - -R"))
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
#dt[, margin := "cd163"]
return(dt)
})
dtr <- rbindlist(dtt)
rr <- fread("~/Sorger/projects/prostate/230803_clinical_data.csv")
rr <- rr[!duplicated(Slide_ID)]
rr[, Gleason := ifelse(surgery_gleason %in% c("4 + 4","4 + 5","5 + 3","5 + 4"), "High", "Low")]
rr <- merge(data.table::copy(rr), dtr, by = "Slide_ID", all.x = TRUE)
rr_long <- melt(rr, id.vars = "Gleason", measure.vars = paste0("topic",seq(10)), variable.name = "Metric", value.name = "Value")
#rr_long[, cells_per_mm2 := Value * 1e6 / (1000^2 * 0.325^2)]
# Assuming the data has two Gleason groups "A" and "B"
get_pvalue <- function(data) {
wilcox.test(data[data$Gleason == "High",]$Value,
data[data$Gleason == "Low",]$Value)$p.value
}
p_values <- by(rr_long, rr_long$Metric, get_pvalue)
labels_with_pvalues <- sapply(names(p_values), function(name) {
sprintf("%s (p = %.3f)", name, p_values[[name]])
})
g <- ggplot(rr_long, aes(x = Gleason, y = Value, fill = Gleason)) +
geom_boxplot(alpha = 0.5, width=0.8, outlier.shape=NA) +
geom_jitter(position = position_jitter(width = 0.2), size = 1, alpha = 0.7, shape=20) +
labs(y = "Thousands of cells / mm2") +
#scale_fill_manual(values = pMMR_colors) +
facet_wrap(~ Metric, labeller = as_labeller(labels_with_pvalues)) +
theme_jw() + scale_y_log10()
save_ppdf(g, "lda_boxplots.pdf", width=6, height=6)
```
```{r scramble-lda}
files <- list.files(path = "~/Sorger/projects/prostate/scramble", full.names = TRUE)
files <- files[grepl("scramble.ptrdl.cys$", files)]
dtt <- lapply(files, function(f) {
cat(basename(f), "\n")
dt <- fread(cmd=paste("cyftools filter",f,"/tmp/tmp.cys && cyftools mean /tmp/tmp.cys - | cyftools view - -R"))
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
#dt[, margin := "cd163"]
return(dt)
})
dtr <- rbindlist(dtt)
rr <- fread("~/Sorger/projects/prostate/230803_clinical_data.csv")
rr <- rr[!duplicated(Slide_ID)]
rr[, Gleason := ifelse(surgery_gleason %in% c("4 + 4","4 + 5","5 + 3","5 + 4"), "High", "Low")]
rr <- merge(data.table::copy(rr), dtr, by = "Slide_ID", all.x = TRUE)
rr_long <- melt(rr, id.vars = "Gleason", measure.vars = paste0("topic",seq(10)), variable.name = "Metric", value.name = "Value")
#rr_long[, cells_per_mm2 := Value * 1e6 / (1000^2 * 0.325^2)]
# Assuming the data has two Gleason groups "A" and "B"
get_pvalue <- function(data) {
wilcox.test(data[data$Gleason == "High",]$Value,
data[data$Gleason == "Low",]$Value)$p.value
}
p_values <- by(rr_long, rr_long$Metric, get_pvalue)
labels_with_pvalues <- sapply(names(p_values), function(name) {
sprintf("%s (p = %.3f)", name, p_values[[name]])
})
g <- ggplot(rr_long, aes(x = Gleason, y = Value, fill = Gleason)) +
geom_boxplot(alpha = 0.5, width=0.8, outlier.shape=NA) +
geom_jitter(position = position_jitter(width = 0.2), size = 1, alpha = 0.7, shape=20) +
labs(y = "Thousands of cells / mm2") +
#scale_fill_manual(values = pMMR_colors) +
facet_wrap(~ Metric, labeller = as_labeller(labels_with_pvalues)) +
theme_jw() + scale_y_log10()
```
```{r jaccard}
ff <- c("/Users/jeremiahwala/Sorger/projects/prostate/chain/low.cys","/Users/jeremiahwala/Sorger/projects/prostate/chain/high.cys")
#dtt <- lapply(master_files, function(f) {
dtt <- lapply(ff, function(f) {
cat(basename(f), "\n")
dt <- fread(cmd=paste("cyftools jaccard",f,"-j"))
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
#dt[, margin := "cd163"]
return(dt)
})
dtr <- rbindlist(dtt)
# scrambled
files <- list.files(path = "~/Sorger/projects/prostate/clean/", full.names = TRUE)
files <- files[grepl("cys$", files)]
#dtt <- lapply(files, function(f) {
dtt <- lapply(ff, function(f) {
cat(basename(f),"\n")
dt <- rbindlist(mclapply(seq(5), function(s) {
dt <- fread(cmd=paste("cyftools scramble -s",s, f,"- | cyftools jaccard - -j"))
dt[, Slide_ID := sub("^(LSP[0-9]+).*", "\\1", basename(f))]
dt[, seed := s]
return(dt)
}, mc.cores=5))
dt[, V3_scramble := mean(V3, na.rm = TRUE), by = c("V1","V2","Slide_ID")]
return(dt[!duplicated(paste(Slide_ID,V1,V2)), .(V1,V2,Slide_ID,V3_scramble)])
})
dtr_s <- rbindlist(dtt)
saveRDS(dtr_s, "~/cache/dtr_s_high_low_jaccard.rds", compress=FALSE)
# merged dtr
dtr <- merge(dtr, dtr_s[, .(V1, V2, Slide_ID, V3_scramble)], by = c("V1", "V2", "Slide_ID"))
dtr[, Jaccard_score_normalized := V3 / V3_scramble]
# Create a mirrored dataset by switching V1 and V2
mirrored_dtr <- dtr[, .(V1 = V2, V2 = V1, Slide_ID = Slide_ID, V3 = V3, Jaccard_score_normalized = Jaccard_score_normalized)]
# Combine the original and mirrored datasets
combined_dtr <- rbind(dtr[,.(V1, V2, V3, Slide_ID, Jaccard_score_normalized)], mirrored_dtr[,.(V1, V2, V3, Slide_ID, Jaccard_score_normalized)])
# Convert data.table to wide format
wide_dtr <- dcast(combined_dtr, V1 + Slide_ID ~ V2, value.var = "Jaccard_score_normalized")
# Convert wide format to long format
long_dtr <- melt(wide_dtr, id.vars = c("V1", "Slide_ID"))
# Convert V1 and variable to character
long_dtr[, V1 := as.character(V1)]
long_dtr[, variable := as.character(variable)]
# Filter out the lower diagonal
long_dtr <- long_dtr[V1 < variable]
# if jaccard normalized
long_dtr[, value_maxed := ifelse(value > 50, 50, value)]
# Plot heatmap using ggplot2
tcell <- c("CD3d","CD4","CD8a","PD1","FOXP3","AMCAR","SMA","CD20","CD57")
g <- ggplot(long_dtr[V1 %in% tcell & variable %in% tcell], aes(x = V1, y = variable)) +
geom_tile(aes(fill = log10(value_maxed)), color = "black") +
scale_fill_gradient2(low = "blue", mid = "white", high = "red", midpoint = 0) +
theme_minimal() +
labs(title = "Jaccard Correlation Enrichment", x="",y="",fill = "Correlation") +
theme_jw() +
theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
facet_wrap(~ Slide_ID) +
coord_fixed(clip = "off") +
theme(plot.margin = margin(1, 1, 1, 1, "cm"))
save_ppdf(g, "jaccard_tcell.pdf", width=5, height=4)
```
```{r jaccard}
files <- list.files(path = "~/Sorger/projects/orion/orion_1_74/pearson/", full.names = TRUE)
files <- files[grepl("pearson$", files)]
col_range <- colorRampPalette(c("darkblue", "white", "darkred"))(200)
dtt <- lapply(files, function(f) {
cat(basename(f), "\n")
dt <- fread(f)
bl <- c("Hoechst", "Argo550","AF1")
dt <- dt[!V1 %in% bl & !V2 %in% bl]
# Visualize using ggplot2
g <- ggplot(data = dt, aes(x = V1, y = V2)) +
geom_tile(aes(fill = V3), color = "white") +
scale_fill_gradient2(low = "blue", high = "red", mid = "white",
midpoint = 0, limit = c(-1,1), name="Correlation") +
#geom_text(aes(label = sprintf("%.2f", value)), vjust = 1) +
theme_jw() + ggtitle(sub("\\..*$", "", basename(f))) +
theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1),