-
Notifications
You must be signed in to change notification settings - Fork 0
/
app.R
996 lines (853 loc) · 41.5 KB
/
app.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
library(shiny)
library(httr)
library(rjson)
library(DT)
library(plotly)
options(repos = BiocManager::repositories())
library(limma)
library(DESeq2)
DEBUG <- Sys.getenv('DEBUG') == 'TRUE'
# ----------------------------------- App UI --------------------------------- #
# Check whether user has auth'd
has_auth_code <- function(params) {
# params is a list object containing the parsed URL parameters. Return TRUE if
# based on these parameters, it looks like auth code is present that we can
# use to get an access token. If not, it means we need to go through the OAuth
# flow.
return(!is.null(params$code))
}
# UI will change depending on whether the user has logged in
uiFunc <- function(req) {
if (!has_auth_code(parseQueryString(req$QUERY_STRING))) {
# Login button
AnonymousUI
} else {
# App UI
AuthenticatedUI
}
}
# Import UI to be shown after user before and after auth'd
source('app_ui.R')
if (!dir.exists('data')){
dir.create('data')
}
# ------------------------ Virtualenv setup -------------------------- #
if (Sys.info()[['sysname']] != 'Darwin'){
# When running on shinyapps.io, create a virtualenv
reticulate::virtualenv_create(envname = 'python35_txn_env',
python = '/usr/bin/python3')
reticulate::virtualenv_install('python35_txn_env',
packages = c('synapseclient', 'requests'))
}
reticulate::use_virtualenv('python35_txn_env', required = T)
reticulate::source_python('connect_to_synapse.py')
# ---------------------------- OAuth --------------------------------- #
# Initialize Synapse client
login_to_synapse(username = Sys.getenv('SYN_USERNAME'),
api_key = Sys.getenv('SYN_API_KEY'))
logged_in <- reactiveVal(FALSE)
source('oauth.R')
# ----------------------------------- Server --------------------------------- #
server <- function(input, output, session) {
# Click on the 'Log in' button to kick off the OAuth round trip
observeEvent(input$action, {
session$sendCustomMessage("customredirect", oauth2.0_authorize_url(API, APP, scope = SCOPE))
return()
})
params <- parseQueryString(isolate(session$clientData$url_search))
if (!has_auth_code(params)) {
return()
}
url <- paste0(API$access, '?', 'redirect_uri=', APP_URL, '&grant_type=',
'authorization_code', '&code=', params$code)
# Get the access_token and userinfo token
token_request <- POST(url,
encode = 'form',
body = '',
authenticate(APP$key, APP$secret, type = 'basic'),
config = list()
)
stop_for_status(token_request, task = 'Get an access token')
token_response <- httr::content(token_request, type = NULL)
access_token <- token_response$access_token
id_token <- token_response$id_token
if (token_request$status_code == 201){
logged_in(T)
}
# ------------------------------ App --------------------------------- #
# Get information about the user
user_response = get_synapse_userinfo(access_token)
user_id = user_response$userid
user_content_formatted = paste(lapply(names(user_response),
function(n) paste(n, user_response[n])), collapse="\n")
# Get user profile
profile_response <- get_synapse_user_profile()
observeEvent(get_synapse_user_profile(), {
query_string = paste0(getQueryString(), '&uid=', user_id)
updateQueryString(query_string, mode = 'replace')
})
# Get the user's teams
teams_response <- get_synapse_teams(user_id)
teams = unlist(lapply(teams_response$results, function(l) paste0(l$name, ' (', l$id, ')')))
team_ids = unlist(lapply(teams_response$results, function(l) paste0('team_', l$id)))
teams_content_formatted = paste(teams, collapse = '\n')
# Format team(s) and project(s) enabled for this app that the user can access
enabled_teams = team_ids[team_ids %in% names(PROJECT_CONFIG)]
PROJECT_DROPDOWN_LIST = list()
ALL_ANALYSES = list()
for (team_id in enabled_teams){
projects = PROJECT_CONFIG[[team_id]]
for (project in projects){
PROJECT_DROPDOWN_LIST[project$project_name] = list(names(project$analyses))
for (analysis in names(project$analyses)){
ALL_ANALYSES[[analysis]] = c(PROJECT_CONFIG[[team_id]][[project$project_name]][['analyses']][[analysis]])
}
}
}
# Get projects associated with that team - why is this empty?
projects_response <- get_synapse_projects(access_token)
# Cache responses
if (DEBUG){
saveRDS(token_response, 'cache/token_response.rds')
saveRDS(user_response, 'cache/user_response.rds')
saveRDS(teams_response, 'cache/teams_response.rds')
saveRDS(projects_response, 'cache/projects_response.rds')
saveRDS(profile_response, 'cache/profile_response.rds')
}
output$userInfo <- renderText(user_content_formatted)
output$teamInfo <- renderText(teams_content_formatted)
# See in app_ui.R with verbatimTextOutput("userInfo")
# ---------------------------- Menus --------------------------------- #
# Logout modal
observeEvent(input$user_account_modal, {
showModal(
modalDialog(title = tagList(
img(src = 'www/synapse_logo.png', width = 200)
),
h4(paste0(profile_response$firstName, ' ', profile_response$lastName)),
p(profile_response$company),
p(user_response$email, style = 'color: #27adde;'),
easyClose = T,
footer = tagList(
actionButton("button_view_syn_profile", "View Profile on Synapse",
style = 'color: #ffffff; background-color: #27adde; border-color: #1ea0cf;',
onclick = paste0("window.open('https://www.synapse.org/#!Profile:", profile_response$ownerId, "', '_blank')")),
modalButton("Back to Analysis")
#actionButton("button_logout", "Log Out")
)
)
)
})
# Project info modal
observeEvent(input$info_projects_modal, {
showModal(modalDialog(
title = 'Selecting a Synapse Project',
p("The Projects listed in this dropdown menu are associated with your Synapse account. You must be granted access to a Project in Synapse in order to view it here. Note that some Projects may not be enabled for this app."),
p("If you belong to multiple Teams on Synapse, the Projects you see in this menu will be grouped underneath the Team name."),
p("Contact the Predictive BioAnalytics group (",
a('[email protected]', href='mailto:[email protected]',
style = 'color: #27adde;'),
") if you have any questions!"),
easyClose = T,
footer = NULL
))
})
output$logged_user <- renderText({
if(logged_in()){
return(paste0('Welcome, ', profile_response$firstName, '!'))
}
})
# ---------------------- Methods writing functions ------------------- #
generate_umap_methods <- function(n_genes, color_by, shape_by){
in_file = 'data/methods/umap.txt'
out_file = 'data/methods/umap_filled.txt'
txt = read.table(in_file, sep = '|', stringsAsFactors = F)
editable_txt = gsub('<GENE>', n_genes, txt[1,])
editable_txt = gsub('<COLOR>', color_by, editable_txt)
editable_txt = gsub('<SHAPE>', shape_by, editable_txt)
txt[1,1] = editable_txt
write.table(txt, out_file, sep = '\t', row.names = F, quote = F, col.names = F)
}
generate_volcano_methods <- function(a, numa, b, numb, method){
in_file = 'data/methods/deg.txt'
out_file = 'data/methods/deg_filled.txt'
txt = read.table(in_file, sep = '|', stringsAsFactors = F)
editable_txt = gsub('<A>', a, txt[1,])
editable_txt = gsub('<NUMA>', numa, editable_txt)
editable_txt = gsub('<B>', b, editable_txt)
editable_txt = gsub('<NUMB>', numb, editable_txt)
txt[1,1] = editable_txt
write.table(txt, out_file, sep = '\t', row.names = F, quote = F, col.names = F)
}
# -------------------------- Tab 1: Samples -------------------------- #
experimentData <- reactiveValues(selected_project = NULL,
umap_color_by = NULL,
umap_shape_by = NULL,
sample_metadata_df = NULL,
sample_color_columns = NULL,
sample_shape_columns = NULL,
gene_counts_df = NULL,
umap_df = NULL,
platform = NULL,
data_loaded = F)
# Load the projects the user has access to
observeEvent(logged_in(), {
updateSelectInput(session, 'project_select',
choices = PROJECT_DROPDOWN_LIST)
experimentData$selected_project <- PROJECT_DROPDOWN_LIST[[1]][1]
})
observeEvent(input$project_select, {
if (input$project_select != 'Loading...'){
experimentData$selected_project <- input$project_select
}
})
selected_project <- reactive({ experimentData$selected_project })
# Load the sample data, gene counts, and UMAP for the analysis
observeEvent(selected_project(), {
proj = selected_project()
if (!is.null(proj)){
ANALYSIS = ALL_ANALYSES[[proj]]
sample_metadata_csv = fetch_synapse_filepath(ANALYSIS$metadata)
experimentData$sample_metadata_df <- read.csv(sample_metadata_csv,
stringsAsFactors = F)
gene_counts_csv = fetch_synapse_filepath(ANALYSIS$counts)
experimentData$gene_counts_df <- read.csv(gene_counts_csv,
stringsAsFactors = F)
umap_csv = fetch_synapse_filepath(ANALYSIS$umap)
experimentData$umap_df <- read.csv(umap_csv,
row.names = 1,
stringsAsFactors = F)
experimentData$platform <- ANALYSIS$platform
experimentData$data_loaded <- T
# Extract colorable columns from column names
dat = experimentData$sample_metadata_df
experimentData$sample_color_columns <- sort(names(dat)[sapply(names(dat), function(x) {
length(unique(dat[,x])) <= 11
})])
# Extract shapeable columns from column names
experimentData$sample_shape_columns <- sort(names(dat)[sapply(names(dat), function(x) {
length(unique(dat[,x])) <= 5
})])
}
})
sample_dat <- reactive({ experimentData$sample_metadata_df })
counts_dat <- reactive({ experimentData$gene_counts_df })
umap_dat <- reactive({ experimentData$umap_df })
platform <- reactive({ experimentData$platform })
loaded <- reactive({ experimentData$data_loaded })
color_columns <- reactive({ experimentData$sample_color_columns })
shape_columns <- reactive({ experimentData$sample_shape_columns })
observeEvent(selected_project(), {
proj = selected_project()
if (!is.null(proj)){
color_choices = color_columns()
shape_choices = shape_columns()
sample_metadata = sample_dat()
updateRadioButtons(session, 'umap_color_by',
choices = color_choices, selected = color_choices[1])
updateRadioButtons(session, 'umap_shape_by',
choices = shape_choices, selected = shape_choices[length(shape_choices)])
updateSelectInput(session, 'select_volcano_column',
choices = names(sample_metadata), selected = names(sample_metadata)[2])
}
})
selected_color_column <- reactive({ input$umap_color_by })
selected_shape_column <- reactive({ input$umap_shape_by })
observeEvent(input$info_umap_modal, {
showModal(
modalDialog(title = "Visualizing sample similarity with UMAP",
p('This plot helps us visualize how similar or different samples are, based on their gene expression.
Similar to a principal components analysis (PCA), the Uniform Manifold Approximation and
Projection (UMAP) algorithm is used here for dimensionality reduction. Briefly, UMAP takes the
expression profile (all genes measured in the experiment, typically 15,000 - 30,000 genes) for every sample
and condenses that information down so that it can be displayed in 2 dimensions. The output is a
plot that summarizes sample-to-sample similarity.'),
p("Let's look at the following example plot: "),
div(style = 'padding: 30px;',
img(src = 'www/umap_example.png', width = 500)),
p('A couple of insights can be gleaned from the plot above. First, we see that samples from Group 1 (
red), Group 2 (green), and Group 3 (blue) tend to cluster near other samples from the same group. This
means that samples within each Group have similar expression profiles.'),
p('Secondly, we see that Group 1, 2, and 3 samples that have been treated with drug (triangle shape)
still cluster by group (color), but all drug-treated samples cluster together too, showing that the
drug treatment likely had a consistent effect on samples from all Groups.'),
p("Finally, there is one outlier green triangle point at the very bottom of the plot. Given the consistency
of the other sample clustering, this sample appears to be a clear outlier. It's possible that
the sample was mislabeled or that something went wrong during the RNA-seq/microarray measurement. When this happens,
it's worth performing some additional quality checks to investigate the cause."),
strong('References:'),
p("Leland McInnes, John Healy, James Melville. UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. 2018. ",
a('Read the paper', href='https://arxiv.org/pdf/1802.03426.pdf',
target = '_blank', style = 'color: #27adde;'),
" (opens in a new window)."),
easyClose = T,
footer = NULL)
)
})
output$umap_plot <- renderPlotly({
# User input plot params
color_column = selected_color_column()
shape_column = selected_shape_column()
# Experimental data
plot_mat = umap_dat()
sample_metadata = sample_dat()
if (loaded() & !is.null(color_column) & !is.null(shape_column)){
if (color_column %in% names(sample_metadata) &
shape_column %in% names(sample_metadata)){
# Format dataframe for plotting
row.names(sample_metadata) = sample_metadata$Sample_ID
plot_df = cbind(sample_metadata[row.names(plot_mat),], plot_mat)
names(plot_df)[(ncol(plot_df)-1):ncol(plot_df)] = c('V1', 'V2')
# Make sure the color & shape columns are factors
plot_df[,color_column] = as.factor(plot_df[,color_column])
plot_df[,shape_column] = as.factor(plot_df[,shape_column])
n_colors = length(unique(plot_df[,color_column]))
n_symbols = length(unique(plot_df[,shape_column]))
plot_cols = c('#27adde', sample(PLOT_COLORS[-10], n_colors-1))
plot_symbols = sample(PLOT_SHAPES, n_symbols)
# Plot
p <- plot_ly(data = plot_df, x = ~V1, y = ~V2,
color = ~get(color_column),
colors = plot_cols,
symbol = ~get(shape_column),
symbols = plot_symbols,
text = ~Sample_ID,
hovertemplate = '<b>Sample ID:</b> %{text}',
type = 'scatter', mode = 'markers',
marker = list(size = 10,
line = list(
color = '#212D32',
width = 1
)))
# Save methods section
generate_umap_methods(nrow(counts_dat()), color_column, shape_column)
# Save to PDF
pdf('data/sample_umap.pdf', height = 6, width = 8)
x_max = max(plot_df$V1)
x_min = min(plot_df$V1)
x_range = x_max - x_min
new_x_max = x_max + .5*x_range
pdf_colors = plot_cols
names(pdf_colors) = unique(plot_df[,color_column])
pdf_symbols = plot_symbols
names(pdf_symbols) = unique(plot_df[,shape_column])
plot(plot_df$V1, plot_df$V2,
main = 'Sample Similarity (UMAP algorithm)',
xlab = 'UMAP dimension 1',
ylab = 'UMAP dimension 2',
xlim = c(x_min, new_x_max),
pch = pdf_symbols[plot_df[,shape_column]],
bg = pdf_colors[plot_df[,color_column]],
las = 1)
unique_colors = as.character(unique(plot_df[,color_column]))
unique_symbols = as.character(unique(plot_df[,shape_column]))
all_samples_combined = paste0(plot_df[,color_column], '__', plot_df[,shape_column])
unique_samples = unique(all_samples_combined)
plot_legend = titlify(unique_samples)
plot_legend_colors = pdf_colors[as.character(sapply(unique_samples, function(x) strsplit(x, '__')[[1]][1]))]
plot_legend_symbols = pdf_symbols[as.character(sapply(unique_samples, function(x) strsplit(x, '__')[[1]][2]))]
legend('right', legend = plot_legend,
pt.bg = plot_legend_colors,
pch = plot_legend_symbols)
dev.off()
# Display Plotly plot in UI
return(p)
} else{
return(NULL)
}
} else{
return(NULL)
}
})
# Download UMAP plot as a PDF
output$download_umap_pdf <- downloadHandler(
filename = "Sample_similarity_UMAP.pdf",
content = function(file) {
file.copy("data/sample_umap.pdf", file)
}
)
# Download UMAP methods as a .txt file
output$download_umap_methods <- downloadHandler(
filename = "Sample_similarity_UMAP_methods.txt",
content = function(file) {
file.copy('data/methods/umap_filled.txt', file)
}
)
# Output the table of all sample metadata
output$table_sample_metadata <- DT::renderDT({
df = sample_dat()
if (!is.null(df)){
return(datatable(df, rownames = F, selection = 'none',
style = 'bootstrap'))
} else{
return(NULL)
}
})
observeEvent(input$info_sample_metadata_modal, {
showModal(
modalDialog(title = "Sample metadata table",
p('This table contains annotations describing each sample used in the experiment. The data was
uploaded to Synapse and can be modified on Synapse if corrections are needed. The color and shape
parameters for the UMAP plot above are dynamically generated from the column names of this file,
so column names can vary as needed by experiment as long as the sample ID is in the first column.'),
a('View the original file on Synapse', href = paste0('https://www.synapse.org/#!Synapse:', ANALYSIS$metadata),
target = '_blank', style = 'color: #27adde;'),
easyClose = T,
footer = NULL)
)
})
# ------------------------- Tab 2: Diff Expr ------------------------- #
diffExprData <- reactiveValues(message = NULL,
group1_samples = NULL,
group2_samples = NULL,
diff_expr_result = NULL,
diff_expr_csv = NULL,
boxplot_df = NULL)
volcano_column <- reactive({ input$select_volcano_column })
group1_criteria <- reactive({ input$select_group1_criteria })
group2_criteria <- reactive({ input$select_group2_criteria })
group1 <- reactive({ diffExprData$group1_samples })
group2 <- reactive({ diffExprData$group2_samples })
most_recent_result <- reactive({ diffExprData$diff_expr_result })
most_recent_boxplot <- reactive({ diffExprData$boxplot_df })
observeEvent(input$info_preproc_modal, {
showModal(
modalDialog(title = paste0(selected_project(), " - gene expression data preprocessing"),
p('This table contains annotations describing each sample used in the experiment. The data was
uploaded to Synapse and can be modified on Synapse if corrections are needed. The color and shape
parameters for the UMAP plot above are dynamically generated from the column names of this file,
so column names can vary as needed by experiment as long as the sample ID is in the first column.'),
easyClose = T,
footer = NULL)
)
})
output$message_diff_expr <- renderUI({
msg = 'The comparison is performed as "Group A vs Group B," so we recommend setting Group B as your control.'
if (!is.null(diffExprData$message)){
msg = diffExprData$message
}
return(helpText(msg))
})
observeEvent(selected_project(), {
proj = selected_project()
if (!is.null(proj)){
dat = sample_dat()
all_columns = names(dat)
initial_column = all_columns[2]
# Set volano column choices, and initial column selected
updateSelectInput(session, 'select_volcano_column',
choices = all_columns, selected = initial_column)
}
})
observeEvent(volcano_column(), {
selected_column = volcano_column()
if (!is.null(selected_column) & selected_column != 'Loading...'){
dat = sample_dat()
# Allow the unique values in that column to be used as filter options
filter_options = unique(as.character(dat[,selected_column]))
updateSelectInput(session, 'select_group1_criteria',
label = paste0('Group A ', selected_column, ' ='),
choices = filter_options, selected = filter_options[1])
updateSelectInput(session, 'select_group2_criteria',
label = paste0('Group B ', selected_column, ' ='),
choices = filter_options, selected = filter_options[length(filter_options)])
}
})
observeEvent(group1_criteria(), {
column = volcano_column()
dat = sample_dat()
group1_filter = group1_criteria()
if (loaded() & column %in% names(dat)){
diffExprData$group1_samples <- dat[dat[,column] == group1_filter, 'Sample_ID']
if (length(diffExprData$group1_samples) < 2 | length(diffExprData$group2_samples) < 2){
diffExprData$message <- 'Too few samples to perform analysis. Please select another group with at least 2 samples.'
} else{
diffExprData$message <- NULL
}
}
})
observeEvent(group2_criteria(), {
column = volcano_column()
dat = sample_dat()
group2_filter = group2_criteria()
if (loaded() & column %in% names(dat)){
diffExprData$group2_samples <- dat[dat[,column] == group2_filter, 'Sample_ID']
if (length(diffExprData$group1_samples) < 2 | length(diffExprData$group2_samples) < 2){
diffExprData$message <- 'Too few samples to perform analysis. Please select another comparison with at least 2 samples per group.'
} else{
diffExprData$message <- NULL
}
}
})
output$group1_group2_selected <- renderText({
a = group1()
b = group2()
group1_filter = group1_criteria()
group2_filter = group2_criteria()
str = '<A> (n = <NUMA> samples) vs <B> (n = <NUMB> samples)'
str = gsub('<A>', group1_filter, str)
str = gsub('<B>', group2_filter, str)
str = gsub('<NUMA>', length(a), str)
str = gsub('<NUMB>', length(b), str)
return(str)
})
observeEvent(input$button_run_volcano, {
withProgress(message = 'Running differential expression analysis', value = 0, {
a = group1()
b = group2()
diff_expr_method = 'DESeq2'
# Increment the progress bar, and update the detail text
incProgress(0.2, detail = 'Formatting data')
group1_filter = group1_criteria()
group2_filter = group2_criteria()
column = volcano_column()
count_data = counts_dat()
sample_metadata = sample_dat()
# Add a _# to gene symbols that are duplicated
count_data$Gene_Symbol = make.unique(as.character(count_data$Gene_Symbol), sep = "_")
# Format for DESeg2
row.names(sample_metadata) = sample_metadata$Sample_ID
sample_metadata$Sample_ID = NULL
row.names(count_data) = count_data$Gene_Symbol
count_data$Gene_Symbol = NULL
if (!is.null(count_data) & !is.null(sample_metadata)
& !is.null(a) & !is.null(b) & !is.null(column)
& !is.null(group1_filter) & !is.null(group2_filter)){
out_filename = paste0('data/diff_expr_', group1_filter, '_vs_', group2_filter, '.csv')
out_filename_counts = paste0('data/diff_expr_', group1_filter, '_vs_', group2_filter, '_counts.csv')
diffExprData$diff_expr_csv <- out_filename
out_rnk = paste0('data/diff_expr_', group1_filter, '_vs_', group2_filter, '.rnk')
count_data_tmp = count_data[,c(a, b)]
sample_data_tmp = sample_metadata[c(a, b),]
if (length(a) > 1 & length(b) > 1){
incProgress(0.2, detail = 'Setting up comparisons (<30 sec)')
if (file.exists(out_filename)){
deg_tab = read.csv(out_filename, stringsAsFactors = F)
boxplot_tab = read.csv(out_filename_counts, stringsAsFactors = F)
} else{
incProgress(0.3, detail = 'Performing analysis (~1 min)')
# If there's only one column, need to transform back into a 2D df
if (is.null(dim(sample_data_tmp))){
sample_data_tmp = data.frame(sample_data_tmp, stringsAsFactors = F)
names(sample_data_tmp) = column
}
plat = platform()
if (plat == 'rnaseq'){
# DESeq2 METHOD
dds <- DESeqDataSetFromMatrix(countData = count_data_tmp,
colData = sample_data_tmp,
design = as.formula(paste0('~ ', column)))
dds <- DESeq(object = dds)
cont = c(column, group1_filter, group2_filter)
res = results(dds,
contrast = cont,
pAdjustMethod = "fdr",
cooksCutoff = FALSE)
deg_tab = data.frame(res[,c('baseMean', 'log2FoldChange', 'padj')])
deg_tab = cbind(Gene = row.names(deg_tab), deg_tab)
deg_tab$Gene = as.character(deg_tab$Gene)
names(deg_tab)[-1] = c('AvgExpr', 'log2FC', 'adjPVal')
deg_tab = deg_tab[order(deg_tab$adjPVal, decreasing = F),]
# Trigger new boxplot (with normalized counts)
norm_counts <- counts(dds, normalized = TRUE)
boxplot_tab = as.data.frame(cbind(sample_data_tmp[,column], t(norm_counts)))
names(boxplot_tab)[1] = column
} else if (plat == 'microarray'){
# Limma
model_matrix = model.matrix(~0 + as.factor(sample_data_tmp[,column]))
colnames(model_matrix) = c(group1_filter, group2_filter)
# Contrasts
contrasts_cmd = paste0('makeContrasts(', group1_filter,
'-', group2_filter, ', levels = model_matrix)')
contrasts = eval(parse(text=contrasts_cmd))
# Differential expression
fit = lmFit(count_data_tmp, model_matrix)
fit2 = contrasts.fit(fit, contrasts = contrasts)
fit2 = eBayes(fit2)
deg_tab = topTable(fit2, number = 100000, adjust = 'fdr')
deg_tab = data.frame(Gene = row.names(deg_tab), deg_tab)
deg_tab$Gene = as.character(deg_tab$Gene)
deg_tab = deg_tab[,c(1,3,2,6)]
names(deg_tab)[-1] = c('AvgExpr', 'log2FC', 'adjPVal')
deg_tab = deg_tab[order(deg_tab$adjPVal, decreasing = F),]
# Trigger new boxplot (with normalized counts)
boxplot_tab = as.data.frame(cbind(sample_data_tmp[,column], t(count_data_tmp)))
names(boxplot_tab)[1] = column
} else{
cat(paste0('Unrecognized platform: ', plat))
}
# Save table to prevent re-creating it in the same session
write.table(deg_tab, out_filename, sep = ',', row.names = F, quote = F)
write.table(boxplot_tab, out_filename_counts, sep = ',', row.names = F, quote = F)
# Save methods file
generate_volcano_methods(group1_filter, length(a),
group2_filter, length(b),
diff_expr_method)
}
incProgress(0.2, detail = 'Finalizing (<30 sec)')
diffExprData$diff_expr_result <- deg_tab
diffExprData$boxplot_df <- boxplot_tab
}
}
})
})
observeEvent(input$info_volcano_modal, {
showModal(
modalDialog(title = "Visualizing differential gene expression with volcano plots",
p('A volcano plot summarizes both the magnitude of expression change (x axis) and the statistical
significance (y axis). Each point represents a gene.'),
p("Let's look at the following example plot: "),
div(style = 'padding: 30px;',
img(src = 'www/volcano_example.png', width = 500)),
p('In the plot above, the genes represented by green points had significantly lower expression in
the Experimental samples compared to the samples in the Control group. Likewise, the genes represented
by red points had significantly higher expression in the Experimental group. A "significant" change
means an adjusted p-value of < 0.05. Genes represented by grey points did not show a significant difference
in expression in the Experimental samples compared to the Control.'),
strong('References:'),
p("Matthew E. Ritchie, Belinda Phipson, Di Wu, Yifang Hu, Charity W. Law, Wei Shi, Gordon K. Smyth. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research. 2015.",
a('View the paper', href='https://academic.oup.com/nar/article/43/7/e47/2414268',
target = '_blank', style = 'color: #27adde;'),
" (opens in a new window)."),
p("Leland McInnes, John Healy, James Melville. UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. 2018. ",
a('View the limma R package vignette', href='https://bioconductor.org/packages/release/bioc/html/limma.html',
target = '_blank', style = 'color: #27adde;'),
" (opens in a new window)."),
easyClose = T,
footer = NULL)
)
})
output$volcano_message <- renderUI({
if (is.null(diffExprData$diff_expr_result)){
return(p(style='padding-left: 20px; color: #D2D6DD;',
'Use the section to the left to set analysis parameters and click "Run analysis" to view results'))
} else{
return(p(style='padding-left: 20px; color: #D2D6DD;',
'Hover over a point to view the gene name'))
}
})
output$volcano_plot <- renderPlotly({
group1_filter = isolate(group1_criteria())
group2_filter = isolate(group2_criteria())
deg_results = most_recent_result()
p_cutoff = 0.05
upper_fc_cutoff = 1
lower_fc_cutoff = -1
if (!is.null(deg_results)){
deg_results$log10pval = -log10(deg_results$adjPVal)
deg_results$sig = ifelse(deg_results$adjPVal < p_cutoff & deg_results$log2FC > upper_fc_cutoff, 'increased',
ifelse(deg_results$adjPVal < p_cutoff & deg_results$log2FC < lower_fc_cutoff, 'decreased',
'not significant'))
plot_cols = c(PLOT_COLORS[2], '#3aa4a9', '#C3C5C7')
names(plot_cols) = c('increased', 'decreased', 'not significant')
# Plot
p <- plot_ly(data = deg_results, x = ~log2FC, y = ~log10pval,
color = ~sig,
colors = plot_cols,
text = ~Gene,
hovertemplate = '<b>Gene:</b> %{text}',
type = 'scatter', mode = 'markers',
marker = list(size = 10,
line = list(
color = '#212D32',
width = 1
))) %>%
layout(title = paste0(group1_filter, ' vs ', group2_filter),
xaxis = list(title = 'log2 fold change'),
yaxis = list(title = '-log10 adjusted p-value')
)
# Save plot as PDF
pdf('data/diffexpr_volcano.pdf', height = 8, width = 6)
volcano_plot(deg_results$log2FC, deg_results$adjPVal,
p_adj_threshold = p_cutoff,
plot_main = paste0(group1_filter, ' vs ', group2_filter))
legend('bottomright', bg = 'white',
pch = 21, pt.bg = plot_cols,
legend = names(plot_cols))
dev.off()
return(p)
} else{
return(NULL)
}
})
# Download volcano plot as a PDF
output$download_volcano_pdf <- downloadHandler(
filename = "DEG_volcano.pdf",
content = function(file) {
file.copy("data/diffexpr_volcano.pdf", file)
}
)
# Download differential expression analysis methods as a .txt file
output$download_volcano_methods <- downloadHandler(
filename = "DEG_methods.txt",
content = function(file) {
if (!is.null(diffExprData$diff_expr_result)){
file.copy('data/methods/deg_filled.txt', file)
} else{
file.copy('data/methods/deg.txt', file)
}
}
)
output$table_differential_expression <- DT::renderDT({
df = most_recent_result()
if (!is.null(df)){
# Replace gene names with hyperlink
gene_names = df$Gene
df$Gene = sapply(gene_names, function(s){
HTML(paste0("<a href='https://www.genecards.org/cgi-bin/carddisp.pl?gene=", s, "' target='_blank'>", s,"</a>"))
})
df = df[,c('Gene', 'adjPVal', 'log2FC', 'AvgExpr')]
# Add protein links
df$Protein = sapply(gene_names, function(s){
# Human Protein Atlas
paste0(HTML(paste0("<a href='https://www.proteinatlas.org/search/", s, "' target='_blank'>", 'Human Protein Atlas',"</a>")), ' | ',
HTML(paste0("<a href='https://www.uniprot.org/uniprot/?fil=organism%3A%22Homo+sapiens+%28Human%29+%5B9606%5D%22&sort=score&query=", s, "' target='_blank'>", 'UniProt',"</a>")))
})
# Round
df$log2FC = round(as.numeric(df$log2FC), 3)
df$AvgExpr = round(as.numeric(df$AvgExpr), 3)
df$adjPVal = formatC(df$adjPVal, format = "e", digits = 3)
# Display
return(datatable(df, rownames = F,
selection = 'single',
style = 'bootstrap', escape = F))
} else{
return(NULL)
}
})
output$download_diff_expr_table <- downloadHandler(
filename = function(){
if (!is.null(diffExprData$diff_expr_csv)){
gsub('data/', '', diffExprData$diff_expr_csv)
}
},
content = function(file) {
if (!is.null(diffExprData$diff_expr_csv)){
file.copy(diffExprData$diff_expr_csv, file)
}
}
)
selected_gene <- reactive({
row_selected = input$table_differential_expression_rows_selected
df = most_recent_result()
if (!is.null(row_selected)){
return(df[row_selected,1])
} else{
# Show the top hit
return(df$Gene[1])
}
})
output$gene_boxplot <- renderPlotly({
boxplot_df = most_recent_boxplot()
boxplot_df$Sample_ID = row.names(boxplot_df)
# To allow coloring of points
group_column = names(boxplot_df)[1]
boxplot_df$Color_ID = sapply(boxplot_df[,group_column], function(s) paste0('X', s))
gene = selected_gene()
boxplot_df[,gene] = log2(as.numeric(boxplot_df[,gene]))
# TODO check these numbers
#boxplot_df[,gene] = as.numeric(boxplot_df[,gene])
boxplot_cols = c('#CCCCCC', '#CCCCCC', PLOT_COLORS[5], '#27adde')
names(boxplot_cols) = c(sort(as.character(unique(boxplot_df[,group_column]))),
sort(unique(boxplot_df$Color_ID)))
if (!is.null(boxplot_df)){
p <- plot_ly(type = 'box', data = boxplot_df,
x = ~get(names(boxplot_df)[1]), y = ~get(gene),
hoverinfo='none',
color = ~get(names(boxplot_df)[1]),
colors = boxplot_cols) %>%
add_markers(~get(names(boxplot_df)[1]), y = ~get(gene),
type = 'scatter', mode = 'markers',
hoverinfo = 'text',
symbol = 21,
text = ~Sample_ID,
color = ~Color_ID,
marker = list(
size = 10,
line = list(
color = '#212D32',
width = 1
))) %>%
layout(showlegend = FALSE,
title = gene,
xaxis = list(title = 'Sample Group'),
yaxis = list(title = paste0(gene, ' (log2 normalized counts)')))
# Save plot as PDF
pdf(paste0('data/gene_boxplot_', gene, '.pdf'),
height = 6, width = 4)
boxplot(get(gene)~get(group_column), data = boxplot_df,
col = 'lightgrey', las = 1,
main = gene, xlab = '',
ylab = paste0(gene, ' (log2 normalized counts)'))
points(get(gene)~get(group_column), data = boxplot_df,
pch = 21, bg = boxplot_cols[boxplot_df$Color_ID])
dev.off()
return(p)
} else{
return(NULL)
}
})
# Download gene boxplot plot as a PDF
output$download_boxplot_pdf <- downloadHandler(
filename = function(){
paste0('gene_boxplot_', selected_gene(), '.pdf')
},
content = function(file) {
file.copy(paste0('data/gene_boxplot_', selected_gene(), '.pdf'), file)
}
)
observeEvent(input$info_diffexpr_modal, {
showModal(
modalDialog(title = "Differentially expressed gene table",
p("This table shows each gene and it's log fold change (logFC) calculated from comparing
Group A - Group B according to the criteria you selected."),
strong('References:'),
p("Matthew E. Ritchie, Belinda Phipson, Di Wu, Yifang Hu, Charity W. Law, Wei Shi, Gordon K. Smyth. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research. 2015.",
a('View the paper', href='https://academic.oup.com/nar/article/43/7/e47/2414268',
target = '_blank', style = 'color: #27adde;'),
" (opens in a new window)."),
easyClose = T,
footer = NULL)
)
})
output$row_diff_expr_results <- renderUI({
boxplot_df = most_recent_boxplot()
if (!is.null(boxplot_df)){
return(tagList(
box(title = tagList("Differentially expressed genes",
HTML(' '),
tags$i(
class = "fa fa-info-circle",
style = "color: #27adde; font-size: 8pt;"
),
actionLink('info_diffexpr_modal', label = 'What is this?',
style = 'font-size: 8pt; color: #27adde;')),
width = 6,
div(style = 'padding-left: 20px;',
p(style='color: #D2D6DD;', "Click on a gene's row to display its boxplot to the right")),
div(withSpinner(dataTableOutput('table_differential_expression'),
type = 4, color = '#27adde')),
div(style = 'padding-left: 20px; padding-top: 52px; padding-bottom: 20px;',
downloadButton('download_diff_expr_table', 'Download full table (.csv)',
style = 'color: #ffffff; background-color: #27adde; border-color: #1ea0cf;
border-radius: 5px;')
)
),
box(title = "Box and whisker plot",
width = 6,
div(style = 'padding-left: 20px;',
p(style='color: #D2D6DD;', "Hover over a point to see the sample name")),
div(withSpinner(plotlyOutput('gene_boxplot'),
type = 4, color = '#27adde')),
div(style = 'padding-left: 20px; padding-top: 65px; padding-bottom: 20px;',
downloadButton('download_boxplot_pdf', 'Download boxplot (.pdf)',
style = 'color: #ffffff; background-color: #27adde; border-color: #1ea0cf;
border-radius: 5px;')
)
))
)
}
})
}
# uiFunc instead of ui
shinyApp(uiFunc, server)