-
Notifications
You must be signed in to change notification settings - Fork 0
/
Proteomics_premodials_Wojciech.Rmd
796 lines (624 loc) · 31.1 KB
/
Proteomics_premodials_Wojciech.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
---
title: "Proteomics Premodials"
author: "Clara Meijs"
date: "2023-11-27"
output:
html_document:
df_print: paged
keep_md: yes
toc: true
toc_float: true
toc_collapsed: true
toc_depth: 5
theme: lumen
---
## Libraries
```{r libraries}
rm(list=ls())
library(pheatmap)
library(ggplot2)
# library(matrixStats)
# library(wesanderson)
# library(clusterProfiler)
# library(enrichplot)
# library(msigdbr)
library(dichromat)
# library(stringr)
library(dplyr)
library(ggrepel)
library(reshape2)
library(umap)
library(ggthemes)
library(cowplot)
#library(MetaboAnalystR)
library(vsn)
library(DEP)
library(readr)
library(naniar)
library(SummarizedExperiment)
library(data.table)
library(readxl)
library(ggpubr)
```
## Set working directories
```{r set-working-directories, message=FALSE, class.source = 'fold-hide'}
# if you are using Rstudio run the following command, otherwise, set the working directory to the folder where this script is in
setwd(dirname(rstudioapi::getActiveDocumentContext()$path))
# create directory for results
dir.create(file.path(getwd(),'results'), showWarnings = FALSE)
# create directory for plots
dir.create(file.path(getwd(),'plots'), showWarnings = FALSE)
```
## Load data
```{r load data}
#load the serum dataset
serum = read_tsv("data/report.pg_matrix.tsv")
#load the plasma dataset
plasma = read_tsv("data/plasma_report.pg_matrix.tsv")
#take redundant information out of the colnames
colnames(serum) = gsub("D:\\\\Przemek\\\\Kuban\\\\2023_10_06\\\\whole samples\\\\sp3\\\\sp3_90min\\\\", "sp3_", colnames(serum))
colnames(serum) = gsub("D:\\\\Przemek\\\\Kuban\\\\2023_10_06\\\\whole samples\\\\SPE\\\\SPE_90min\\\\", "SPE_", colnames(serum))
colnames(plasma) = gsub("D:\\\\Przemek\\\\Kuban\\\\Kuban_plazma\\\\plazma", "plasma", colnames(plasma))
#write duplicates as A and B
for(i in 1:8){
name = paste0("serum_", i)
names = colnames(serum)[grep(name, colnames(serum))]
names = c(paste0("sp3_", name, "_A"),
paste0("sp3_", name, "_B"),
paste0("SPE_", name, "_A"),
paste0("SPE_", name, "_B"))
colnames(serum)[grep(name, colnames(serum))] = names
}
#there is one gene name missing and we replace it with the protein name
serum$Genes[is.na(serum$Genes)] = "A0A0G2JRQ6"
#keep only gene name column of protein identification columns
serum = serum[,c(1, 4, grep("sp3", colnames(serum)), grep("SPE", colnames(serum)))]
plasma = plasma[,c(1, 4, grep("plasma_", colnames(plasma)))]
#make dataframe
serum = as.data.frame(serum)
plasma = as.data.frame(plasma)
#change gene names by removing everything after the ;
f = function(x){
a = unlist(strsplit(x, split=';', fixed=TRUE))[1]
return(a)}
serum$Genes = unlist(lapply(serum$Genes, FUN = f))
plasma$Genes = unlist(lapply(plasma$Genes, FUN = f))
serum$Protein.Group = unlist(lapply(serum$Protein.Group, FUN = f))
plasma$Protein.Group = unlist(lapply(plasma$Protein.Group, FUN = f))
colnames(plasma) = c("Uniprot", "Genes", "plasma_1", "plasma_2", "plasma_3", "plasma_4", "plasma_5", "plasma_6", "plasma_7", "plasma_8")
gene_names_serum = serum[,1:2]
gene_names_plasma = plasma[,1:2]
#make gene name the rownames
serum$Genes = make.unique(serum$Genes)
rownames(serum) = serum$Genes
serum = serum[,colnames(serum)!="Genes"]
serum = serum[,-1]
plasma$Genes = make.unique(plasma$Genes)
rownames(plasma) = plasma$Genes
plasma = plasma[,colnames(plasma)!="Genes"]
plasma = plasma[,-1]
#check if columns have correct identification (if they are numerical)
str(serum)
str(plasma)
#check if all missing are being recognized correctly
sum(is.na(serum))
sum(is.na(plasma))
#separate the different protocols
serum_sp3 = serum[,grepl("sp3", colnames(serum))]
serum_SPE = serum[,grepl("SPE", colnames(serum))]
#make a table with merged duplicates
serum_SPE_merged = as.data.frame(matrix(NA, nrow = nrow(serum_sp3), ncol = ncol(serum_sp3)/2))
rownames(serum_SPE_merged) = rownames(serum_sp3)
serum_sp3_merged = serum_SPE_merged
variance_sp3 = variance_SPE = serum_sp3_merged
missing_sp3 = missing_SPE = serum_sp3_merged
for(i in 1:8){
name = paste0("serum_",i)
serum_sp3[,grep(name, colnames(serum_sp3))]
serum_sp3_merged[,i] = apply(X = serum_sp3[,grep(name, colnames(serum_sp3))], function(x) mean(x, na.rm = TRUE), MARGIN = 1)
serum_SPE_merged[,i] = apply(X = serum_SPE[,grep(name, colnames(serum_SPE))], function(x) mean(x, na.rm = TRUE), MARGIN = 1)
variance_sp3[,i] = apply(X = serum_sp3[,grep(name, colnames(serum_sp3))], function(x) max(x) - min(x), MARGIN = 1)
variance_SPE[,i] = apply(X = serum_SPE[,grep(name, colnames(serum_SPE))], function(x) max(x) - min(x), MARGIN = 1)
missing_sp3[,i] = apply(X = serum_sp3[,grep(name, colnames(serum_sp3))], function(x) 0 + sum(is.na(x)), MARGIN = 1)
missing_SPE[,i] = apply(X = serum_SPE[,grep(name, colnames(serum_SPE))], function(x) 0 + sum(is.na(x)), MARGIN = 1)
colnames(serum_sp3_merged)[i] = name
colnames(serum_SPE_merged)[i] = name
colnames(variance_sp3)[i] = name
colnames(variance_SPE)[i] = name
colnames(missing_sp3)[i] = name
colnames(missing_SPE)[i] = name
}
relative_variance_sp3 = variance_sp3/serum_sp3_merged
relative_variance_SPE = variance_SPE/serum_SPE_merged
serum_sp3_merged[serum_sp3_merged == "NaN"] = NA
serum_SPE_merged[serum_SPE_merged == "NaN"] = NA
#
# #transpose for summarized experiment
# serum_sp3 = as.data.frame(t(serum_sp3))
# serum_sp3_merged = as.data.frame(t(serum_sp3_merged))
# serum_SPE = as.data.frame(t(serum_SPE))
# serum_SPE_merged = as.data.frame(t(serum_SPE_merged))
#make summarized experiments
#serum_sp3
abundance.columns <- 1:ncol(serum_sp3) # get abundance column numbers
clin = data.frame(label = colnames(serum_sp3), #very limited clinical variables
condition = c("control") ,
replicate = 1:ncol(serum_sp3))
serum_sp3$name = rownames(serum_sp3)
serum_sp3$ID = gene_names_serum$Protein.Group
experimental.design = clin
se_serum_sp3 <- make_se(serum_sp3, abundance.columns, experimental.design)
#serum_SPE
abundance.columns <- 1:ncol(serum_SPE) # get abundance column numbers
clin = data.frame(label = colnames(serum_SPE), #very limited clinical variables
condition = c("control") ,
replicate = 1:ncol(serum_SPE))
serum_SPE$name = rownames(serum_SPE)
serum_SPE$ID = gene_names_serum$Protein.Group
experimental.design = clin
se_serum_SPE <- make_se(serum_SPE, abundance.columns, experimental.design)
#serum_sp3_merged
abundance.columns <- 1:ncol(serum_sp3_merged) # get abundance column numbers
clin = data.frame(label = colnames(serum_sp3_merged), #very limited clinical variables
condition = c("control") ,
replicate = 1:ncol(serum_sp3_merged))
serum_sp3_merged$name = rownames(serum_sp3_merged)
serum_sp3_merged$ID = gene_names_serum$Protein.Group
experimental.design = clin
se_serum_sp3_merged <- make_se(serum_sp3_merged, abundance.columns, experimental.design)
#serum_SPE_merged
abundance.columns <- 1:ncol(serum_SPE_merged) # get abundance column numbers
clin = data.frame(label = colnames(serum_SPE_merged), #very limited clinical variables
condition = c("control") ,
replicate = 1:ncol(serum_SPE_merged))
serum_SPE_merged$name = rownames(serum_SPE_merged)
serum_SPE_merged$ID = gene_names_serum$Protein.Group
experimental.design = clin
se_serum_SPE_merged <- make_se(serum_SPE_merged, abundance.columns, experimental.design)
#plasma
abundance.columns <- 1:ncol(plasma) # get abundance column numbers
clin = data.frame(label = colnames(plasma), #very limited clinical variables
condition = c("control") ,
replicate = 1:ncol(plasma))
plasma$name = rownames(plasma)
plasma$ID = gene_names_plasma$Uniprot
experimental.design = clin
se_plasma <- make_se(plasma, abundance.columns, experimental.design)
#save data
write.csv(serum, "results/serum_raw_data.csv", row.names=TRUE)
write.csv(plasma, "results/plasma_raw_data.csv", row.names=TRUE)
write.csv(serum_sp3_merged, "results/serum_sp3_raw_data_no_duplicates.csv", row.names=TRUE)
write.csv(serum_SPE_merged, "results/serum_SPE_raw_data_no_duplicates.csv", row.names=TRUE)
write.csv(variance_sp3, "results/serum_sp3_variance_of_duplicates.csv", row.names=TRUE)
write.csv(variance_SPE, "results/serum_SPE_variance_of_duplicates.csv", row.names=TRUE)
write.csv(relative_variance_sp3, "results/serum_sp3_relative_variance_of_duplicates.csv", row.names=TRUE)
write.csv(relative_variance_SPE, "results/serum_SPE_relative_variance_of_duplicates.csv", row.names=TRUE)
write.csv(missing_sp3, "results/serum_sp3_missing_within_duplicates.csv", row.names=TRUE)
write.csv(missing_SPE, "results/serum_sp3_missing_within_duplicates.csv", row.names=TRUE)
```
## Scatterplots duplicates
```{r scatterplots duplicates}
serum_sp3_A = serum_sp3[,grep("_A",colnames(serum_sp3))]
serum_sp3_B = serum_sp3[,grep("_B",colnames(serum_sp3))]
serum_SPE_A = serum_SPE[,grep("_A",colnames(serum_SPE))]
serum_SPE_B = serum_SPE[,grep("_B",colnames(serum_SPE))]
serum_sp3_A = reshape::melt(as.matrix(serum_sp3_A))
serum_sp3_B = reshape::melt(as.matrix(serum_sp3_B))
serum_SPE_A = reshape::melt(as.matrix(serum_SPE_A))
serum_SPE_B = reshape::melt(as.matrix(serum_SPE_B))
serum_sp3_long = as.data.frame(cbind(serum_sp3_A, serum_sp3_B$value))
colnames(serum_sp3_long)[3:4] = c("duplicate_A", "duplicate_B")
serum_sp3_long = na.omit(serum_sp3_long)
serum_sp3_long$duplicate_A = log2(serum_sp3_long$duplicate_A)
serum_sp3_long$duplicate_B = log2(serum_sp3_long$duplicate_B)
serum_SPE_long = as.data.frame(cbind(serum_SPE_A, serum_SPE_B$value))
colnames(serum_SPE_long)[3:4] = c("duplicate_A", "duplicate_B")
serum_SPE_long = na.omit(serum_SPE_long)
serum_SPE_long$duplicate_A = log2(serum_SPE_long$duplicate_A)
serum_SPE_long$duplicate_B = log2(serum_SPE_long$duplicate_B)
a = ggplot(serum_sp3_long, aes(x=duplicate_A, y=duplicate_B)) +
geom_point( color="darksalmon", alpha = 0.5) +
geom_abline(intercept = 0, slope = 1) +
ggtitle("scatterplot duplicates serum sp3") +
theme_few()
b = ggplot(serum_SPE_long, aes(x=duplicate_A, y=duplicate_B)) +
geom_point( color="yellow4", alpha = 0.5) +
geom_abline(intercept = 0, slope = 1) +
ggtitle("scatterplot duplicates serum SPE") +
theme_few()
ggarrange(a, b, ncol = 2, nrow = 1)
ggsave("plots/scatterplots_duplicates.pdf", width = 11, height = 8/2, units = "in")
```
## Venn diagram proteins
```{r venn diagram proteins}
# install.packages("ggVennDiagram")
library(ggVennDiagram)
proteins_sp3 = rownames(serum_sp3)
proteins_SPE = rownames(serum_SPE)
proteins_plasma = rownames(plasma)
proteins = list(serum_sp3 = proteins_sp3,
serum_SPE = proteins_SPE,
plasma = proteins_plasma)
# 2D Venn diagram
ggVennDiagram(proteins, set_color = c("darksalmon", "yellow4", "mediumpurple1")) +
scale_fill_gradient(low = "white", high = "grey50") +
scale_color_manual(values = c("darksalmon", "yellow4", "mediumpurple1"))
ggsave(file = "plots/venn_diagram.pdf", width = 11/2, height = 8/2, units = "in")
```
## Filtering, normalization, and imputation
```{r filtering, normalization, and imputation}
set.seed(9)
# Filter for proteins that are quantified in at least 2/3 of the samples.
se_serum_sp3_filt <- filter_proteins(se_serum_sp3_merged, "fraction", min = 0.66)
se_serum_SPE_filt <- filter_proteins(se_serum_SPE_merged, "fraction", min = 0.66)
se_plasma_filt <- filter_proteins(se_plasma, "fraction", min = 0.66)
se_serum_sp3_norm = normalize_vsn(se_serum_sp3_filt)
se_serum_SPE_norm = normalize_vsn(se_serum_SPE_filt)
se_plasma_norm = normalize_vsn(se_plasma_filt)
# imputation with several methods: MinProb, MAN, KNN
#sp3 serum
se_serum_sp3_imp_Minprob <- impute(se_serum_sp3_norm, fun = "MinProb", q=0.01)
se_serum_sp3_imp_man <- impute(se_serum_sp3_norm, fun = "man", shift = 1.8, scale = 0.3)
se_serum_sp3_imp_knn <- impute(se_serum_sp3_norm, fun = "knn", rowmax = 0.9)
#SPE serum
se_serum_SPE_imp_Minprob <- impute(se_serum_SPE_norm, fun = "MinProb", q=0.01)
se_serum_SPE_imp_man <- impute(se_serum_SPE_norm, fun = "man", shift = 1.8, scale = 0.3)
se_serum_SPE_imp_knn <- impute(se_serum_SPE_norm, fun = "knn", rowmax = 0.9)
#plasma
se_plasma_imp_Minprob <- impute(se_plasma_norm, fun = "MinProb", q=0.01)
se_plasma_imp_man <- impute(se_plasma_norm, fun = "man", shift = 1.8, scale = 0.3)
se_plasma_imp_knn <- impute(se_plasma_norm, fun = "knn", rowmax = 0.9)
#put se's in a list
se_serum = list(serum_sp3_duplicates = se_serum_sp3,
serum_SPE_duplicates = se_serum_SPE,
serum_sp3_merged = se_serum_sp3_merged,
serum_SPE_merged = se_serum_SPE_merged,
serum_sp3_filt = se_serum_sp3_filt,
serum_SPE_filt = se_serum_SPE_filt,
serum_sp3_norm = se_serum_sp3_norm,
serum_SPE_norm = se_serum_SPE_norm,
serum_sp3_imp_Minprob = se_serum_sp3_imp_Minprob,
serum_sp3_imp_man = se_serum_sp3_imp_man,
serum_sp3_imp_knn = se_serum_sp3_imp_knn,
serum_SPE_imp_Minprob = se_serum_SPE_imp_Minprob,
serum_SPE_imp_man = se_serum_SPE_imp_man,
serum_SPE_imp_knn = se_serum_SPE_imp_knn)
se_plasma = list(plasma_raw = se_plasma,
plasma_filt = se_plasma_filt,
plasma_norm = se_plasma_norm,
plasma_imp_Minprob = se_plasma_imp_Minprob,
plasma_imp_man = se_plasma_imp_man,
plasma_imp_knn = se_plasma_imp_knn)
saveRDS(se_serum, "results/se_serum_list.rds")
saveRDS(se_plasma, "results/se_plasma_list.rds")
write.csv(as.data.frame(assay(se_plasma_norm)), "results/plasma_processed.csv")
write.csv(as.data.frame(assay(se_serum_sp3_norm)), "results/serum_sp3_processed.csv")
write.csv(as.data.frame(assay(se_serum_SPE_norm)), "results/serum_SPE_processed.csv")
```
## Missing inspection
```{r missing inspection}
#serum
vis_miss_plots = list()
frequency_plots = list()
intensity_distributions = list()
for(i in 1:8){
name = names(se_serum)[i]
vis_miss_plots[[i]] = vis_miss(as.data.frame(assay(se_serum[[i]])),show_perc = TRUE, show_perc_col = TRUE, cluster = F) +
ggtitle(name)
frequency_plots[[i]] = plot_frequency(se_serum[[i]]) + ggtitle(name)
intensity_distributions[[i]] = plot_detect(se_serum[[i]])
print(paste0(name, " has ", ncol(se_serum[[i]]), " samples and ", nrow(se_serum[[i]]), " proteins"))
}
#ggarrange
ggarrange(plotlist = vis_miss_plots, nrow = length(vis_miss_plots)/4, ncol = 4)
ggsave("plots/missing_vis_miss_plots_serum.jpeg", width = 11*2, height = 8, units = "in")
ggarrange(plotlist = frequency_plots, nrow = length(frequency_plots)/4, ncol = 4)
ggsave("plots/missing_frequency_plots_serum.pdf", width = 11*2, height = 8, units = "in")
ggarrange(plotlist = intensity_distributions, nrow = length(intensity_distributions)/4, ncol = 4, labels = names(se_serum))
ggsave("plots/missing_intensity_distributions_serum.pdf", width = 11*2, height = 8, units = "in")
imputation_plots = list()
# Plot intensity distributions before and after imputation
imputation_plots[[1]] = plot_imputation(se_serum_sp3_norm, se_serum_sp3_imp_Minprob,
se_serum_sp3_imp_man, se_serum_sp3_imp_knn)
imputation_plots[[2]] = plot_imputation(se_serum_SPE_norm, se_serum_SPE_imp_Minprob,
se_serum_SPE_imp_man, se_serum_SPE_imp_knn)
ggarrange(plotlist = imputation_plots, nrow = 1, ncol = 2)
ggsave("plots/imputation_plots_serum.pdf", width = 11, height = 8, units = "in")
#plasma
vis_miss_plots = list()
frequency_plots = list()
intensity_distributions = list()
for(i in 1:3){
name = names(se_plasma)[i]
vis_miss_plots[[i]] = vis_miss(as.data.frame(assay(se_plasma[[i]])),show_perc = TRUE, show_perc_col = TRUE, cluster = F) +
ggtitle(name)
frequency_plots[[i]] = plot_frequency(se_plasma[[i]]) + ggtitle(name)
intensity_distributions[[i]] = plot_detect(se_plasma[[i]])
print(paste0(name, " has ", ncol(se_plasma[[i]]), " samples and ", nrow(se_plasma[[i]]), " proteins"))
}
#ggarrange
ggarrange(plotlist = vis_miss_plots, nrow = 1, ncol = 3)
ggsave("plots/missing_vis_miss_plots_plasma.jpeg", width = 11*2, height = 4, units = "in")
ggarrange(plotlist = frequency_plots, nrow = 1, ncol = 3)
ggsave("plots/missing_frequency_plots_plasma.pdf", width = 11*2, height = 4, units = "in")
ggarrange(plotlist = intensity_distributions, nrow = 1, ncol = 3, labels = names(se_serum))
ggsave("plots/missing_intensity_distributions_plasma.pdf", width = 11*2, height = 4, units = "in")
imputation_plots = list()
# Plot intensity distributions before and after imputation
plot_imputation(se_plasma_norm, se_plasma_imp_Minprob,
se_plasma_imp_man, se_plasma_imp_knn)
ggsave("plots/imputation_plots_plasma.pdf", width = 11, height = 8, units = "in")
```
## Density plot
```{r Visualization 1b: Density plot}
#figure raw
d = as.data.frame(assay(se_serum_sp3))
d = reshape2::melt(d)
d$technique = rep("serum_sp3", nrow(d))
d2 = as.data.frame(assay(se_serum_SPE))
d2 = reshape2::melt(d2)
d2$technique = rep("serum_SPE", nrow(d2))
d3 = as.data.frame(assay(se_plasma[["plasma_raw"]]))
d3 = reshape2::melt(d3)
d3$technique = rep("plasma", nrow(d3))
d = as.data.frame(rbind(d, d2))
d = as.data.frame(rbind(d, d3))
a = ggplot(d, aes(x=value, color=technique)) +
geom_density() +
theme_few() +
scale_colour_few() +
ggtitle("raw data") +
scale_color_manual(values = c("mediumpurple1", "darksalmon", "yellow4"))
#figure filtered
d = as.data.frame(assay(se_serum_sp3_filt))
d = reshape2::melt(d)
d$technique = rep("sp3", nrow(d))
d2 = as.data.frame(assay(se_serum_SPE_filt))
d2 = reshape2::melt(d2)
d2$technique = rep("SPE", nrow(d2))
d3 = as.data.frame(assay(se_plasma[["plasma_filt"]]))
d3 = reshape2::melt(d3)
d3$technique = rep("plasma", nrow(d3))
d = as.data.frame(rbind(d, d2))
d = as.data.frame(rbind(d, d3))
b = ggplot(d, aes(x=value, color=technique)) +
geom_density() +
theme_few() +
scale_colour_few() +
ggtitle("filtered data")+
scale_color_manual(values = c("mediumpurple1", "darksalmon", "yellow4"))
#figure normalized
d = as.data.frame(assay(se_serum_sp3_norm))
d = reshape2::melt(d)
d$technique = rep("sp3", nrow(d))
d2 = as.data.frame(assay(se_serum_SPE_norm))
d2 = reshape2::melt(d2)
d2$technique = rep("SPE", nrow(d2))
d3 = as.data.frame(assay(se_plasma[["plasma_norm"]]))
d3 = reshape2::melt(d3)
d3$technique = rep("plasma", nrow(d3))
d = as.data.frame(rbind(d, d2))
d = as.data.frame(rbind(d, d3))
c = ggplot(d, aes(x=value, color=technique)) +
geom_density() +
theme_few() +
scale_colour_few() +
ggtitle("filtered and normalized data")+
scale_color_manual(values = c("mediumpurple1", "darksalmon", "yellow4"))
ggarrange(a,b,c, ncol = 3, nrow = 1)
ggsave(file = "plots/density.pdf", width = 11*1.5, height = 3, units = "in")
```
## Make boxplots and histograms data
```{r make boxplots and histograms data}
#visualize every dataset, also raw
mean_expression_plot = function(data, file_sample, file_mass, title){
ggplot(data = reshape2::melt(data), aes(x=Var1, y=value)) +
geom_boxplot(color="darkseagreen4", fill="darkseagreen3") +
theme_set(theme_minimal()) +
theme_few() +
scale_colour_few() +
theme(legend.position = "none") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
theme(axis.text=element_text(size=6)) +
ggtitle(title)
ggsave(file_sample, width = 11, height = 8, units = "in")
ggplot(data = reshape2::melt(data), aes(x=reorder(as.factor(Var2),value), y=value)) +
geom_boxplot(color="darkseagreen4", fill="darkseagreen3") +
theme_set(theme_minimal()) +
theme_few() +
scale_colour_few() +
theme(legend.position = "none") +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
theme(axis.text=element_text(size=6))+
ggtitle(title)
ggsave(file_mass, width = 11*2, height = 8, units = "in")
}
for(i in 1:length(se_serum)){
mean_expression_plot(data = t(assay(se_serum[[i]])),
file_sample = paste0("plots/boxplots_expression_each_sample_",
names(se_serum)[i],
".pdf"),
file_mass = paste0("plots/boxplots_expression_each_mass_",
names(se_serum)[i],
".pdf"),
title = names(se_serum)[i])
}
for(i in 1:length(se_plasma)){
mean_expression_plot(data = t(assay(se_plasma[[i]])),
file_sample = paste0("plots/boxplots_expression_each_sample_",
names(se_plasma)[i],
".pdf"),
file_mass = paste0("plots/boxplots_expression_each_mass_",
names(se_plasma)[i],
".pdf"),
title = names(se_plasma)[i])
}
```
## Heatmap
```{r heatmap}
set.seed(9)
#functions for saving the heatmaps as figures
save_pheatmap_pdf <- function(x, filename, width=11/2, height=8/2) {
stopifnot(!missing(x))
stopifnot(!missing(filename))
pdf(filename, width=width, height=height)
grid::grid.newpage()
grid::grid.draw(x$gtable)
dev.off()
}
make_pheatmap <- function(data, cluster_cols = T, main = "Heatmap", clustering_method = "ward.D", show_rownames = T,
labels_col = clin$label){
p = pheatmap::pheatmap(data, name = "expression", cutree_cols = 1,
show_colnames = T,
show_rownames = show_rownames,
fontsize = 4,
fontsize_col = 4,
fontsize_row = 2,
#annotation_col = annotation,
#annotation_colors = annotation_colours,
#annotation_row = annotation_row,
color = viridis::viridis(100, option="G", direction = -1,),
main = main,
border_color=NA,
cluster_cols = cluster_cols,
cluster_rows = F,
labels_col = labels_col,
#clustering_method = clustering_method,
na_col = "grey80")
return(p)
}
# loop for all datasets and all methods
for(i in 1:length(se_serum)){
title = names(se_serum)[i]
print(title)
#create heatmaps with all patients
#serum
#without grouping, all proteins
p = make_pheatmap(data = assay(se_serum[[i]]), cluster_cols = F, main = paste0("Heatmap all proteins\n",title, "\n not clustered"), show_rownames = F,
labels_col = se_serum[[i]]@colData$label)
save_pheatmap_pdf(p, filename = paste0("plots/heatmap_",title,".pdf"))
# without grouping, 100 most variable proteins
d = assay(se_serum[[i]])
d2 = head(order(rowVars(d),decreasing = T),100)
p = make_pheatmap(data = d[d2,], cluster_cols = F, main = paste0("Heatmap 100 most variable proteins\n",title, "\nnot clustered"),
labels_col = se_serum[[i]]@colData$label)
save_pheatmap_pdf(p, filename = paste0("plots/heatmap_mostvar_",title,".pdf"))
}
for(i in 1:length(se_plasma)){
title = names(se_plasma)[i]
print(title)
#plasma
#without grouping, all proteins
p = make_pheatmap(data = assay(se_plasma[[i]]), cluster_cols = F, main = paste0("Heatmap all proteins\n",title, "\n not clustered"), show_rownames = F,
labels_col = se_plasma[[i]]@colData$label)
save_pheatmap_pdf(p, filename = paste0("plots/heatmap_",title,".pdf"))
# without grouping, 100 most variable proteins
d = assay(se_plasma[[i]])
d2 = head(order(rowVars(d),decreasing = T),100)
p = make_pheatmap(data = d[d2,], cluster_cols = F, main = paste0("Heatmap 100 most variable proteins\n",title, "\nnot clustered"),
labels_col = se_plasma[[i]]@colData$label)
save_pheatmap_pdf(p, filename = paste0("plots/heatmap_mostvar_",title,".pdf"))
}
#heatmap with relative variance
title = "relative_difference_duplicates"
relative_variance_sp3 = relative_variance_sp3 * 100
relative_variance_sp3[relative_variance_sp3 > 300] = 300
relative_variance_sp3[relative_variance_sp3 == 0] = NA
relative_variance_SPE = relative_variance_SPE * 100
relative_variance_SPE[relative_variance_SPE > 300] = 300
relative_variance_SPE[relative_variance_SPE == 0] = NA
#without grouping, all proteins
p = make_pheatmap(data = relative_variance_sp3, cluster_cols = F, main = paste0("Heatmap all proteins sp3\n",title, "\n not clustered"), show_rownames = F, labels_col = se_serum[["serum_sp3_merged"]]@colData$label)
save_pheatmap_pdf(p, filename = paste0("plots/heatmap_sp3_",title,".pdf"))
p = make_pheatmap(data = relative_variance_SPE, cluster_cols = F, main = paste0("Heatmap all proteins SPE\n",title, "\n not clustered"), show_rownames = F, labels_col = se_serum[["serum_SPE_merged"]]@colData$label)
save_pheatmap_pdf(p, filename = paste0("plots/heatmap_SPE_",title,".pdf"))
title = "difference_duplicates"
#variance_sp3[variance_sp3>100] = 100
#without grouping, all proteins
p = make_pheatmap(data = variance_sp3, cluster_cols = F, main = paste0("Heatmap all proteins sp3\n",title, "\n not clustered"), show_rownames = F, labels_col = se_serum[["serum_sp3_merged"]]@colData$label)
save_pheatmap_pdf(p, filename = paste0("plots/heatmap_sp3_",title,".pdf"))
#variance_SPE[variance_SPE>100] = 100
#without grouping, all proteins
p = make_pheatmap(data = variance_SPE, cluster_cols = F, main = paste0("Heatmap all proteins SPE\n",title, "\n not clustered"), show_rownames = F, labels_col = se_serum[["serum_SPE_merged"]]@colData$label)
save_pheatmap_pdf(p, filename = paste0("plots/heatmap_SPE_",title,".pdf"))
#create heatmap of data before merging triplicates
title = "raw_before_merging_duplicates"
#remove ridiculously high values
serum2 = log2(serum)
#without grouping, all proteins
p = make_pheatmap(data = serum2, cluster_cols = F, main = paste0("Heatmap all proteins\n",title, "\n not clustered"), show_rownames = F, labels_col = colnames(serum))
save_pheatmap_pdf(p, filename = paste0("plots/heatmap_sp3_and_SPE",title,".pdf"))
```
## UMAP
```{r UMAP}
# # set seed for reproducible results
set.seed(9)
group = c("mediumpurple1", "darksalmon", "yellow4")
UMAP_density_plot = function(data,
ggtitle = "UMAP with disease status labels",
legend_name = "Disease status",
labels = clin$Condition,
file_location = "plots/UMAP_condition.pdf",
file_location_labels = "plots/UMAP_condition_labels.pdf",
colour_set = c("seagreen4", "slateblue1", "salmon")){
# run umap function
umap_out = umap::umap(data)
umap_plot = as.data.frame(umap_out$layout)
#add condition labels
umap_plot$group = labels
# plot umap
p1 = ggplot(umap_plot) + geom_point(aes(x=V1, y=V2, color = as.factor(group))) +
ggtitle(ggtitle) +
theme_few() +
scale_colour_few() +
scale_color_manual(name = legend_name,
labels = levels(as.factor(umap_plot$group)),
values = colour_set)
xdens <-
axis_canvas(p1, axis = "x") +
geom_density(data = umap_plot, aes(x = V1, fill = group, colour = group), alpha = 0.3) +
scale_fill_manual( values = colour_set) +
scale_colour_manual( values = colour_set)
ydens <-
axis_canvas(p1, axis = "y", coord_flip = TRUE) +
geom_density(data = umap_plot, aes(x = V2, fill = group, colour = group), alpha = 0.3) +
coord_flip() +
scale_fill_manual(values = colour_set) +
scale_colour_manual( values = colour_set)
p1 %>%
insert_xaxis_grob(xdens, grid::unit(1, "in"), position = "top") %>%
insert_yaxis_grob(ydens, grid::unit(1, "in"), position = "right") %>%
ggdraw()
p1
# save umap
ggsave(file_location, width = 11/2, height = 8/2, units = "in")
p1 + geom_text(label = rownames(umap_plot), x = umap_plot$V1, y = umap_plot$V2,
hjust = 0, nudge_x = 1, size = 1.5, colour = "grey")
# save umap with labels
ggsave(file_location_labels, width = 11/2, height = 8/2, units = "in")
}
d1 = t(assay(se_serum[["serum_sp3_imp_Minprob"]]))
d2 = t(assay(se_serum[["serum_SPE_imp_Minprob"]]))
d3 = t(assay(se_plasma[["plasma_imp_Minprob"]]))
proteins_in_both_fluids = colnames(d1)[colnames(d1) %in% colnames(d3)]
proteins_in_both_fluids = proteins_in_both_fluids[proteins_in_both_fluids %in% colnames(d2)]
d1 = d1[,proteins_in_both_fluids]
d2 = d2[,proteins_in_both_fluids]
d3 = d3[,proteins_in_both_fluids]
d = as.data.frame(rbind(d1,d2))
d = as.data.frame(rbind(d,d3))
labels_group = c(rep("serum_sp3", 8), rep("serum_SPE", 8), rep("plasma", 8))
title = "serum_vs_plasma"
#perform plots with function
UMAP_density_plot(data = d,
ggtitle = paste0("UMAP with fluid labels\n", title),
legend_name = "Fluid labels",
labels = labels_group,
file_location = paste0("plots/UMAP_fluid_group_",title,".pdf"),
file_location_labels = paste0("plots/UMAP_fluid_group_labels_",title,".pdf"),
colour_set = group)
```
## SessionInfo
```{r}
sessionInfo()
```