/
server.R
2558 lines (2218 loc) · 128 KB
/
server.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# This file is part of SMARTsize.
#
# SMARTsize is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# SMARTsize is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with SMARTsize. If not, see <http://www.gnu.org/licenses/>.
##### SERVER.R FOR SMART SAMPLE SIZE CALCULATOR #####
### NICK SEEWALD, 2014
### UNIVERSITY OF MICHIGAN
### DEPARTMENT OF BIOSTATISTICS
options(encoding = 'UTF-8')
# This file is part of SMARTsize.
#
# SMARTsize is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# SMARTsize is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with SMARTsize. If not, see <http://www.gnu.org/licenses/>.
##### SERVER.R FOR SMART SAMPLE SIZE CALCULATOR #####
### NICK SEEWALD, 2014
### UNIVERSITY OF MICHIGAN
### DEPARTMENT OF BIOSTATISTICS
options(encoding = 'UTF-8')
### Start server operation
shinyServer(
function(input, output, session){
session$allowReconnect(TRUE)
premade <- eventReactive(input$premade.startbutton,
list(
p("The three designs below are commonly seen in the field. If your design
matches one, click the button below it to get started. Otherwise, you can
design your own. Click an image to zoom."),
fluidRow(
column(4,
div(a(img(id = "D1.diagram.home", src = "images/SMARTdesignA__.gif",
class = "img-responsive img-zoom", alt = "Design I"),
href = "images/SMARTdesignA__.gif", target = "_blank")),
br(),
actionButton("pickTabA", "Use Design I", class = "center-block"),
p("Both responders and non-responders to first-stage treatment are re-randomized.
There are 8 embedded adaptive interventions: {A,C,E}, {A,C,F}, {A,D,E}, {A,D,F},
{B,G,I}, {B,G,J}, {B,H,I}, {B,H,J}.",
a("Click here for an example from the field.", `data-toggle` = "modal",
`data-target` = "#exampleAmodal", style = "color:#6b6b6b")),
includeHTML("www/html/exampleAmodal.html")
),
column(4,
div(a(img(id = "D2.diagram.home", src = "images/SMARTdesignB__.gif",
class = "img-responsive img-zoom", alt = "Design II"),
href = "images/SMARTdesignB__.gif", target = "_blank")),
br(),
actionButton("pickTabB","Use Design II", class = "center-block"),
p("Only non-responders to either first-stage treatment are re-randomized.
There are 4 embedded adaptive interventions: {A,C,D}, {A,C,E}, {B,F,G}, {B,F,H}.",
a("Click here for an example from the field.", `data-toggle` = "modal",
`data-target` = "#exampleBmodal", style = "color:#6b6b6b")),
includeHTML("www/html/exampleBmodal.html")
),
column(4,
div(a(img(id = "D3.diagram.home", src = "images/SMARTdesignC__.gif",
class = "img-responsive img-zoom", alt = "Design III"),
href = "images/SMARTdesignC__.gif", target = "_blank")),
br(),
bsButton("pickTabC","Use Design III", class = "center-block"),
p("Only non-responders to a particular first-stage treatment are re-randomized.
There are 3 embedded adaptive interventions: {A,C,D}, {A,C,E}, {B,F,G}.",
a("Click here for an example from the field.", `data-toggle` = "modal",
`data-target` = "#exampleCmodal", style = "color:#6b6b6b")),
includeHTML("www/html/exampleCmodal.html")
)
)
# br(),
# fluidRow(
# column(3),
# ,
# column(3)
# )
)
)
output$premadeDesigns <- renderUI(premade())
observeEvent(input$premade.startbutton, shinyjs::hide("premade.startbutton"))
output$design1img.home <- renderImage(list(src = file.path("www", "images", "SMARTdesignA__.gif")), deleteFile = FALSE)
### Watch for clicks on pickTab actionButtons rendered under design diagrams
### On click, redirect to appropriate tab. (More intuitive navigation structure)
# observeEvent(input$pickTabA, updateTabsetPanel(session, "SMARTsize", selected = "Design I"))
# observeEvent(input$pickTabB, updateTabsetPanel(session, "SMARTsize", selected = "Design II"))
# observeEvent(input$pickTabC, updateTabsetPanel(session, "SMARTsize", selected = "Design III"))
observeEvent(input$DYO.startbutton, updateTabsetPanel(session, "SMARTsize", selected = "Design and Size"))
##### DESIGN I #####
##### Design I Header #####
### Render the design image which highlights selected DTRs
### Takes the names of selected DTRs, generates a filepath, and renders the image in the UI
### Note that this requires a very specific naming convention for image assets
### 'SMARTdesignX_DTR1_DTR2.gif'; if DTRX is unselected, DTR name is 0.
output$designAimg <- renderImage(expr = {
filename <-
normalizePath(file.path('./www/images',
paste('SMARTdesignA_', input$firstDTRcompareA,
'_', input$secondDTRcompareA,'.gif', sep = ''))
)
list(src = filename)
}, deleteFile = FALSE)
##### Design I Modules #####
# Backend for outcome selection radioButtons
design1.outcome <- callModule(selectDTROutcome, "design1.outcome")
output$design1outcome <- renderText(design1.outcome())
outputOptions(output, 'design1outcome', suspendWhenHidden = FALSE)
# Backend for result options (alpha, desired power/sample size, etc.)
design1.resultOptions <- callModule(resultOptions, "design1.resultOptions")
output$design1resultOptions <- renderText(unlist(design1.resultOptions()))
outputOptions(output, 'design1resultOptions', suspendWhenHidden = FALSE)
# Backend for selection of primary comparison for sample size calculation
design1.primaryAim <- callModule(primaryAim, "design1.primaryAim",
rerand = reactive({c("responders", "nonresponders")}))
output$design1primaryAim <- renderText(design1.primaryAim())
outputOptions(output, 'design1primaryAim', suspendWhenHidden = FALSE)
### Render 'selectize' dropdown boxes with placeholder text for AI selection.
### Second DTR input populates with all DTRs that are NOT the first DTR
### (eliminates ability to select same DTR twice)
### Placed in server.R rather than ui.R because of dependency on first DTR selection
output$selectAI1A <- renderUI({
AI <- selectizeInput("firstDTRcompareA", label = text.refDTRLabel,
choices = designA.AIs,
options = list(
placeholder = text.refDTRPlaceholder,
onInitialize = I('function() { this.setValue(0); }')
)
)
return(AI)
})
output$selectAI2A <- renderUI({
AI <- selectizeInput("secondDTRcompareA",label = text.compDTRLabel,
choices = designA.AIs[substr(designA.AIs, 1, 1) !=
substringDTR1A()[2]],
options = list(
placeholder = text.compDTRPlaceholder,
onInitialize = I('function() { this.setValue(0); }')
)
)
return(AI)
})
##### DESIGN A PROBABILITY INPUT #####
## Read in DTR names from dropdowns and parse them to determine
# first and second stage treatments
## Outputs full DTR name, first-stage treatment (first character),
# second-stage treatment if response (third character),
# and second-stage treatment if non-response(last character)
## NOTE that these positions are exclusive to design A because responders
# have two second-stage treatment options
substringDTR1A <- reactive({
if (length(input$firstDTRcompareA) > 0) {
DTR1 <- paste(input$firstDTRcompareA)
firstStage1 <- substr(DTR1, 1, 1)
secondStageR1 <- substr(DTR1, 3, 3)
secondStageNR1 <- substr(DTR1, 6, 6)
return(c(DTR1, firstStage1, secondStageR1, secondStageNR1))
}
else return(c(0, 0, 0, 0))
})
substringDTR2A <- reactive({
if (length(input$secondDTRcompareA) > 0) {
DTR2 <- paste(input$secondDTRcompareA)
firstStage1 <- substr(DTR2, 1, 1)
secondStageR1 <- substr(DTR2, 3, 3)
secondStageNR1 <- substr(DTR2, 6, 6)
return(c(DTR2, firstStage1, secondStageR1, secondStageNR1))
}
else return(c(0, 0, 0, 0))
})
## When a first DTR is selected, render an input box corresponding to
# whatever input method is selected.
## For DTR-conditional or cell-specific inputs, first numericInput is
# P(S|DTR1), enabled/disabled depending on cellOrConditionalA
## For target-difference or odds ratio, relevant numericInputs are rendered
## This is the ONLY location in which difference and odds ratio
# numericInputs are built.
generateBinaryInputs1A <- reactive({
validate(
need(input$firstDTRcompareA, text.refDTRPlaceholder)
)
if (input$cellOrConditionalA == TRUE) {
return(disable(numericInput("DTRsuccA1disable",
label = html.refDTRSuccess,
value = NA, min = 0, max = 1, step = 0.01)))
} else {
return(list(numericInput("DTRsuccA1",
label = html.refDTRSuccess,
value = NA, min = 0, max = 1, step = 0.01),
bsTooltip(id = "DTRsuccA1", title = text.tooltip,
placement = "right", trigger = "focus"))
)}
})
generateBinaryInputs2A <- reactive({
validate(
need(input$secondDTRcompareA != "", text.compDTRPlaceholder)
)
# Full DTR success probability
if (input$targetDiffCheckA == FALSE && input$targetOddsCheckA == FALSE) {
if (input$cellOrConditionalA == TRUE) {
return(disable(numericInput("DTRsuccA2disable",
label = html.compDTRSuccess,
value = NA, min = 0, max = 1, step = 0.01)))
}
else{
return(list(numericInput("DTRsuccA2", label = html.compDTRSuccess,
value = NA, min = 0, max = 1, step = 0.01),
bsTooltip(id = "DTRsuccA2", title = text.tooltip,
placement = "right", trigger = "focus"))
)}
}
# Target difference
if (input$targetDiffCheckA == TRUE && input$targetOddsCheckA == FALSE) {
return(list(
numericInput("targetDiffA", label = text.targDiffLabel,
value = NULL, min = 0.01, max = 0.99, step = 0.01),
bsTooltip(id = "targetDiffA", title = text.tooltip,
placement = "right", trigger = "focus"),
radioButtons("diffDirectionA", label = text.diffDirection,
choices = list("Smaller" = -1, "Larger" = 1),
selected = -1)
))
}
# Target Odds Ratio
if (input$targetOddsCheckA == TRUE) {
return(list(
numericInput("targetORA", label = text.targORInputLabel,
value = NULL, min = 0, step = 0.01),
bsTooltip(id = "targetORA", title = text.tooltip,
placement = "right", trigger = "focus")
))
}
})
### Render UI components generated above
output$binaryDTR1probA <- renderUI({
generateBinaryInputs1A()
})
output$binaryDTR2probA <- renderUI({
generateBinaryInputs2A()
})
## For cell-specific probabilities, render a sequence of numericInputs
# labeled by information from DTR substrings
## When DTR1 and DTR2 begin with the same treatment, P(S|stage1trt,r)
# is rendered only once, in output$cellProbsDTR1A
output$cellProbsDTR1A <- renderUI({
return(list(
numericInput("marginalFirstStageA1",
label = paste(text.pathSuccess, substringDTR1A()[2], "r",
substringDTR1A()[3], sep = ""),
value = 0, min = 0, max = 1, step = 0.01),
numericInput("marginalSecondStageNRA1",
label = paste(text.pathSuccess, substringDTR1A()[2], "nr",
substringDTR1A()[4], sep = ""),
value = 0, min = 0, max = 1, step = 0.01),
bsTooltip(id = "marginalFirstStageA1", title = text.tooltip,
placement = "right", trigger = "focus"),
bsTooltip(id = "marginalSecondStageNRA1", title = text.tooltip,
placement = "right", trigger = "focus")
))
})
output$cellProbsDTR2A <- renderUI({
return(list(
numericInput("marginalFirstStageA2",
label = paste(text.pathSuccess, substringDTR2A()[2], "r",
substringDTR2A()[3], sep = ""),
value = 0, min = 0, max = 1, step = 0.01),
numericInput("marginalSecondStageNRA2",
label = paste(text.pathSuccess, substringDTR2A()[2], "nr",
substringDTR2A()[4], sep = ""),
value = 0, min = 0, max = 1, step = 0.01),
bsTooltip(id = "marginalFirstStageA2", title = text.tooltip,
placement = "right", trigger = "focus"),
bsTooltip(id = "marginalSecondStageNRA2", title = text.tooltip,
placement = "right", trigger = "focus")
))
})
### Render enabled/disabled numericInputs when outcome is continuous
generateContinuousInputA <- reactive({
validate(
need(input$firstDTRcompareA, text.refDTRPlaceholder),
need(input$secondDTRcompareA, text.compDTRPlaceholder)
)
return(list(
numericInput("effectSizeA", label = text.effectSize,
value = 0, min = 0, max = 10, step = 0.01),
bsTooltip(id = "effectSizeA", title = text.tooltip,
placement = "right", trigger = "focus"))
)
})
output$continuousProbA <- renderUI({
generateContinuousInputA()
})
##### DESIGN A OBSERVERS #####
### Allow checkboxes to function like radio buttons, while also allowing none to be selected
observe({
if (input$targetDiffCheckA) {
updateCheckboxInput(session, "cellOrConditionalA", value = FALSE)
updateCheckboxInput(session, "targetOddsCheckA", value = FALSE)
}
})
observe({
if (input$targetOddsCheckA) {
updateCheckboxInput(session, "cellOrConditionalA", value = FALSE)
updateCheckboxInput(session, "targetDiffCheckA", value = FALSE)
}
})
### If cell-specific input is selected, deselect target difference and odds-ratio options
### Update disabled full DTR success inputs with computed probabilities
observe({
if (input$cellOrConditionalA == TRUE) {
updateCheckboxInput(session, "targetDiffCheckA", value = FALSE)
updateCheckboxInput(session, "targetOddsCheckA", value = FALSE)
updateNumericInput(session, "DTRsuccA1disable", value = generateProbsA()[1])
updateNumericInput(session, "DTRsuccA2disable", value = generateProbsA()[2])
updateNumericInput(session, "DTRsuccA1", value = generateProbsA()[1])
updateNumericInput(session, "DTRsuccA2", value = generateProbsA()[2])
}
}, priority = 2)
##### DESIGN A RESULT BACKEND #####
### Compute full DTR probabilities or effect size when providing cell-specific probabilities or mean/SD
generateProbsA <- reactive({
if (design1.outcome() == "Binary" && input$cellOrConditionalA == TRUE) {
pDTR1 <- fullDTRprob(input$marginalFirstStageA1, input$respA,
input$marginalSecondStageNRA1)
pDTR2 <- fullDTRprob(input$marginalFirstStageA2, input$respA,
input$marginalSecondStageNRA2)
}
else if (design1.outcome() == "Binary" && input$cellOrConditionalA == FALSE) {
pDTR1 <- input$DTRsuccA1
pDTR2 <- input$DTRsuccA2
}
return(c(pDTR1,pDTR2))
})
### Make sure input probabilities are valid (i.e., check for blank entries)
checkDTRinputsA <- reactive({
return(as.logical((input$DTRsuccA1 > 0) & (input$DTRsuccA1 < 1) &
(input$DTRsuccA2 > 0) & (input$DTRsuccA2 < 1)))
})
### Determine which inputs are being given, check their quality, then pass
### the appropriate arguments for computation
### e.g., if target odds-ratio is given, compute a DTR success probability
### assuming the other is 0.5
dataCompilerA <- reactive({
### Error Check: unselected DTRs, blank/invalid response probability
validate(
need(!(is.na(input$respA)), text.noResponse) %then%
need(0 <= input$respA && input$respA <= 1, text.invalidResponse)
)
### Binary outcome, DTR-specific success probabilities
if (design1.outcome() == "Binary" && input$cellOrConditionalA == FALSE &&
input$targetDiffCheckA == FALSE && input$targetOddsCheckA == FALSE) {
### Error Check: unselected DTRs, blank success probabilities,
### equal success probabilities, invalid success probabilities
validate(
need(!is.null(input$DTRsuccA1), text.refDTRPlaceholder),
need(!is.null(input$DTRsuccA2), text.compDTRPlaceholder) %then%
need(!is.null(input$DTRsuccA1) && !is.null(input$DTRsuccA2),
text.noSuccessProb) %then%
need(input$DTRsuccA1 != input$DTRsuccA2, text.sameSuccessProb) %then%
need(checkDTRinputsA(), text.invalidSuccessProb)
)
return(c(input$DTRsuccA1, input$DTRsuccA2))
}
# Binary outcome, cell-specific success probabilities
if (design1.outcome() == "Binary" && input$cellOrConditionalA == TRUE &&
input$targetDiffCheckA == FALSE && input$targetOddsCheckA == FALSE) {
# Error Check: equal DTR-specific success probabilities
validate(
need(generateProbsA()[1] != generateProbsA()[2], text.invalidCellResult)
)
return(c(generateProbsA()[1], generateProbsA()[2]))
}
### Binary outcome, target difference in success probabilities
if (design1.outcome() == "Binary" && input$cellOrConditionalA == FALSE &&
input$targetDiffCheckA == TRUE && input$targetOddsCheckA == FALSE) {
# Error Check: invalid target difference (must be less than 0.5 since
# we're using a conservative reference probability)
validate(
need(input$DTRsuccA1 + as.numeric(input$diffDirectionA) *
input$targetDiffA >= 0 && input$DTRsuccA1 +
as.numeric(input$diffDirectionA) * input$targetDiffA <= 1,
text.invalidDiffResult),
need(input$targetDiffA > 0, text.invalidDiff)
)
return(c(input$DTRsuccA1, input$DTRsuccA1 +
as.numeric(input$diffDirectionA) * input$targetDiffA))
}
### Binary outcome, target odds ratio of success
if (design1.outcome() == "Binary" && input$cellOrConditionalA == FALSE &&
input$targetDiffCheckA == FALSE && input$targetOddsCheckA == TRUE) {
### Error check: missing/invalid odds ratio
validate(
need(is.numeric(input$targetORA), text.noOddsRatio) %then%
need(input$targetORA != 1, text.invalidOR1) %then%
need(input$targetORA != 0, text.invalidOR0)
)
q <- (input$DTRsuccA1 / (1 - input$DTRsuccA1)) * input$targetORA
return(c(input$DTRsuccA1, q / (1 + q)))
}
### Continuous outcome, standardized effect size
if (design1.outcome() == "Continuous") {
### Error check: nonzero effect size
validate(
need(input$firstDTRcompareA, text.refDTRPlaceholder),
need(input$secondDTRcompareA, text.compDTRPlaceholder),
need(!is.null(input$effectSizeA), " ") %then%
need(!is.na(input$effectSizeA), text.noEffectSize) %then%
need(input$effectSizeA != 0, text.invalidEffectSize)
)
return(input$effectSizeA)
}
})
### Construct a portion of the explainer sentence that appears below the
### result, depending on the provided inputs.
sentenceCompilerA <- reactive({
if (design1.outcome() == "Binary" && input$cellOrConditionalA == FALSE &&
input$targetDiffCheckA == FALSE && input$targetOddsCheckA == FALSE) {
return(paste(text.sentenceOverallSuccess, input$firstDTRcompareA,
" and ", input$secondDTRcompareA, ", are ", input$DTRsuccA1,
" and ", input$DTRsuccA2, ", respectively", sep = ""))
}
if (design1.outcome() == "Binary" && input$cellOrConditionalA == TRUE &&
input$targetDiffCheckA == FALSE && input$targetOddsCheckA == FALSE) {
return(paste(text.sentenceOverallSuccess, input$firstDTRcompareA, " and ",
input$secondDTRcompareA, ", are ", generateProbsA()[1],
" and ", generateProbsA()[2], ", respectively", sep = ""))
}
if (design1.outcome() == "Binary" && input$cellOrConditionalA == FALSE &&
input$targetDiffCheckA == TRUE && input$targetOddsCheckA == FALSE) {
return(paste(text.sentenceDiff, input$firstDTRcompareA, " and ",
input$secondDTRcompareA, ", is ", input$targetDiffA, sep = ""))
}
if (design1.outcome() == "Binary" && input$cellOrConditionalA == FALSE &&
input$targetDiffCheckA == FALSE && input$targetOddsCheckA == TRUE) {
return(paste(text.sentenceOR, input$firstDTRcompareA, " and ",
input$secondDTRcompareA, ", is ",input$targetORA, sep = ""))
}
if (design1.outcome() == "Continuous") {
return(paste(text.sentenceEffectSize, input$firstDTRcompareA, " and ",
input$secondDTRcompareA, ", is ", input$effectSizeA, sep = ""))
}
})
##### DESIGN A RESULTS #####
# Pass arguments from dataCompilerA() to appropriate R function
# Extract and render relevant output
## Sample size calculation and output for binary outcome
output$binarySampleSizeA <- renderUI({
validate(
need(input$inputPowerA > 0, text.power0),
need(input$inputPowerA < 1, text.power100)
)
A <- dataCompilerA()[1] / (1 - dataCompilerA()[1])
B <- (dataCompilerA()[2] / (1 - dataCompilerA()[2])) / A
designEffect <- 2
finalAlpha <- ifelse(input$selectAlternativeA == "two.sided",
input$alphaA / 2, input$alphaA)
rawSampleSize <- 2 * ((qnorm(1 - input$inputPowerA) +
qnorm(finalAlpha)) ^ 2) *
(((1 + A) ^ 2) * B + ((1 + A * B) ^ 2)) / (A * B * (log(B)) ^ 2)
finalSampleSize <- ceiling(designEffect * rawSampleSize)
formatPower <- paste(input$inputPowerA * 100, "%", sep = "")
formatAlpha <- paste(input$alphaA * 100, "%", sep = "")
formatAltHyp <- switch(input$selectAlternativeA,
"one.sided" = "one-sided ",
"two.sided" = "two-sided ")
formatResp <- as.numeric(input$respA)
HTML(paste("<h4 style='color:blue';> N=", paste(finalSampleSize)," </h4>
<p> We wish to find the sample size for a trial with a binary outcome
where the probability of response to first-stage intervention is ",
formatResp, sentenceCompilerA(), ". Given a ", formatAltHyp,
" test with ", formatAlpha, " type-I error, we require a sample
size of at least ", finalSampleSize, " to make this comparison
with ", formatPower, " power. </p>", sep = ""))
})
## Power computation and output for binary outcome
output$binaryPowerA <- renderUI({
validate(
need(input$alphaA != 0 && !is.null(input$alphaA), html.alpha0),
need(input$inputSampleSizeA != 0, text.size0) %then%
need(input$inputSampleSizeA >= 8, text.lowSizeA)
)
A <- dataCompilerA()[1] / (1 - dataCompilerA()[1])
B <- (dataCompilerA()[2] / (1 - dataCompilerA()[2])) / A
designEffect <- 2
finalAlpha <- ifelse(input$selectAlternativeA == "two.sided",
input$alphaA / 2, input$alphaA)
powerQuantile <- sqrt((input$inputSampleSizeA * A * B * log(B) ^ 2) /
(2 * designEffect * ((1 + A) ^ 2 * B +
(1 + A*B) ^ 2))) +
qnorm(finalAlpha)
finalPower <- round(pnorm(powerQuantile), digits = 3)
formatPower <- paste(finalPower * 100, "%", sep = "")
formatAlpha <- paste(input$AlphaA * 100, "%", sep = "")
formatSize <- as.numeric(input$inputSampleSizeA)
formatResp <- as.numeric(input$respA)
HTML(paste("<h4 style='color:blue';> Power=", paste(formatPower), " </h4>
<p> For a trial of size N=", formatSize, " with a binary outcome where
the probability of response to first-stage intervention is ",
formatResp, sentenceCompilerA(),", we have at least ",
formatPower, " power. </p>", sep = ""))
})
## Sample size computation and output for continuous outcome
output$continuousSampleSizeA <- renderUI({
validate(
need(input$inputPowerA > 0, text.power0),
need(input$inputPowerA < 1, text.power100),
need(dataCompilerA() > 0, text.invalidEffectSize) %then%
need(is.numeric(rawSampleSize), text.lowEffectSizeA)
)
alt.hyp <- switch(input$selectAlternativeA,
"one.sided" = "greater")
designEffect <- 4
rawSampleSize <- try(pwr.norm.test(d = dataCompilerA(),
sig.level = input$alphaA,
power = input$inputPowerA,
alternative = alt.hyp)$n,
silent = T)
finalSampleSize <- ceiling(2 * designEffect * rawSampleSize)
formatPower <- paste(input$inputPowerA * 100, "%" , sep = "")
formatAlpha <- paste(input$alphaA * 100, "%" , sep = "")
formatAltHyp <- switch(input$selectAlternativeA,
"one.sided" = "one-sided ",
"two.sided" = "two-sided ")
formatResp <- as.numeric(input$respA)
HTML(paste("<h4 style='color:blue';> N=", paste(finalSampleSize), " </h4>
<p> We wish to find the sample size for a trial with a continuous
outcome where the probability of response to first-stage
intervention is ", formatResp, sentenceCompilerA(),
". Given a ", formatAltHyp, " test with ", formatAlpha,
" type-I error, we require a sample size of at least ",
finalSampleSize, " to make this comparison with ", formatPower,
" power. </p>", sep = ""))
})
output$continuousPowerA <- renderUI({
validate(
need(input$inputSampleSizeA != 0, text.size0) %then%
need(size >= 1, text.lowSizeA)
)
alt.hyp <- switch(input$selectAlternativeA,
'one.sided' = 'greater')
designEffect <- 4
size <- (input$inputSampleSizeA/(2*designEffect))
finalPower <- round(try(pwr.norm.test(d = dataCompilerA(),
sig.level = input$alphaA,
n = size,
alternative = alt.hyp)$power,
silent = T), digits = 3)
formatPower <- paste(finalPower * 100, "%", sep = "")
formatAlpha <- paste(input$alphaA * 100, "%", sep = "")
formatSize <- as.numeric(input$inputSampleSizeA)
formatResp <- as.numeric(input$respA)
HTML(paste("<h4 style='color:blue';> Power=", paste(formatPower)," </h4>
<p> For a trial of size N=", formatSize, " with a continuous outcome
where the probability of response to first-stage intervention
is ", formatResp, sentenceCompilerA(), ", we have ",
formatPower, " power. </p>", sep = ""))
})
##### DESIGN 2 #####
##### Design 2 Header #####
### Render the design image which highlights selected DTRs
### Takes the names of selected DTRs, generates a filepath, and renders the image in the UI
### Note that this requires a specific naming convention for image assets
### 'SMARTdesignX_DTR1_DTR2.gif'; if DTRX is unselected, DTR name is 0.
output$designBimg <- renderImage({
filename <- normalizePath(file.path('./www/images', paste('SMARTdesignB_',input$firstDTRcompareB,'_',input$secondDTRcompareB,'.gif',sep='')))
list(src = filename)
}, deleteFile = FALSE)
##### Design 2 Modules #####
# Backend for outcome selection radioButtons
design2.outcome <- callModule(selectDTROutcome, "design2.outcome")
output$design2outcome <- renderText(design2.outcome())
outputOptions(output, 'design2outcome', suspendWhenHidden = FALSE)
# Backend for result options (alpha, desired power/sample size, etc.)
design2.resultOptions <- callModule(resultOptions, "design2.resultOptions")
output$design2resultOptions <- renderText(unlist(design2.resultOptions()))
outputOptions(output, 'design2resultOptions', suspendWhenHidden = FALSE)
# Backend for selection of primary comparison for sample size calculation
design2.primaryAim <- callModule(primaryAim, "design2.primaryAim",
rerand = reactive("nonresponders"))
output$design2primaryAim <- renderText(design2.primaryAim())
outputOptions(output, 'design2primaryAim', suspendWhenHidden = FALSE)
### Render 'selectize' dropdown boxes with placeholder text for AI selection.
### Second DTR input populates with all DTRs that are NOT the first DTR--eliminates ability to select same DTR twice
### Placed in server.R rather than ui.R because of dependency on first DTR selection
output$selectAI1B <- renderUI({
AI <- selectizeInput("firstDTRcompareB", label = "Reference Adaptive Intevention",
choices = designB.AIs,
options = list(
placeholder = text.refDTRPlaceholder,
onInitialize = I('function() { this.setValue(0); }')
)
)
return(AI)
})
output$selectAI2B <- renderUI({
AI <- selectizeInput("secondDTRcompareB",label = "Comparison Adaptive Intervention",
choices = designB.AIs[substr(designB.AIs, 1, 1) != substr(input$firstDTRcompareB, 1, 1)],
options = list(
placeholder = text.compDTRPlaceholder,
onInitialize = I('function() { this.setValue(""); }')
)
)
return(AI)
})
##### DESIGN B PROBABILITY INPUT #####
### Read in DTR names from dropdowns and parse them to determine first and second stage treatments
### Reactive function allows on-the-fly changes in return values with changes in selection
### Outputs full DTR name, first-stage treatment (first character), second-stage treatment (last character)
### NOTE that these positions are exclusive to design B because responders share the same second-stage treatment
substringDTR1B <- reactive({
if (length(input$firstDTRcompareB) > 0) {
DTR1 <- paste(input$firstDTRcompareB)
firstStage1 <- substr(DTR1, 1, 1)
secondStageR1 <- substr(DTR1, 3, 3)
secondStageNR1 <- substr(DTR1, 6, 6)
return(c(DTR1, firstStage1, secondStageR1, secondStageNR1))
}
else{
return(c(0,0,0))
}
})
substringDTR2B <- reactive({
if (length(input$secondDTRcompareB) > 0) {
DTR2 <- paste(input$secondDTRcompareB)
firstStage2 <- substr(DTR2,1,1)
secondStageR2 <- substr(DTR2,3,3)
secondStageNR2 <- substr(DTR2,6,6)
return(c(DTR2, firstStage2, secondStageR2, secondStageNR2))
}
else{
return(c(0,0,0))
}
})
#When a first DTR is selected, render an input box corresponding to whatever input method is selected.
# For DTR-conditional or cell-specific inputs, first numericInput is P(S|DTR1), enabled/disabled depending on cellOrConditionalB
# For target-difference or OR, relevant numericInputs are rendered
# This is the ONLY location in which difference and OR numericInputs are built.
generateBinaryInputs1B <- reactive({
validate(
need(input$firstDTRcompareB, text.refDTRPlaceholder)
)
if (input$cellOrConditionalB == TRUE) {
return(disable(numericInput("DTRsuccB1disable",
label = HTML("Probability of Success for Reference AI <img src='images/blue_dash.gif'>"),
value = NULL, min = 0, max = 1, step = 0.01)))
} else
return(list(numericInput("DTRsuccB1",
label = HTML("Probability of Success for Reference AI <img src='images/blue_dash.gif'>"),
value = NULL, min = 0, max = 1, step = 0.01),
bsTooltip(id = "DTRsuccB1", title = "Input can range from 0-1 and must be in decimal form, up to two places.",
placement = "right", trigger = "focus"))
)
})
generateBinaryInputs2B <- reactive({
validate(
need(input$secondDTRcompareB != "", "Please select a Comparison AI.")
)
if(input$targetDiffCheckB == FALSE && input$targetOddsCheckB == FALSE){
if(input$cellOrConditionalB == TRUE){
return(disable(numericInput("DTRsuccB2disable",
label = HTML("Probability of Success for Comparison AI <img src='images/red_dash.gif'>"),
value = NULL, min = 0, max = 1, step = 0.01)))
} else
return(list(numericInput("DTRsuccB2",
label = HTML("Probability of Success for Comparison AI <img src='images/red_dash.gif'>"),
value = NA, min = 0, max = 1, step = 0.01),
bsTooltip(id = "DTRsuccB2",
title = "Input can range from 0-1 and must be in decimal form, up to two places.",
placement = "right", trigger = "focus")
))
}
if(input$targetDiffCheckB == TRUE && input$targetOddsCheckB == FALSE){
return(list(numericInput("targetDiffB", label = "Target Difference in Success Probabilities",
value = NULL, min = 0.01, max = 0.99, step = 0.01),
bsTooltip("targetDiffB", title = "Input must be in decimal form, up to two places.",
placement = "right", trigger = "focus"),
radioButtons("diffDirectionB", label = "Is the probability of the Comparison AI smaller or larger than the probability of the Reference AI?",
choices = list("Smaller" = -1, "Larger" = 1), selected = -1)
))
}
if(input$targetOddsCheckB == TRUE){
return(list(numericInput("targetORB", label = "Target Odds Ratio of Success",
value = NULL, min = 0, step = 0.01),
bsTooltip(id = "targetORB", title = "Input must be positive and in decimal form, up to two places.",
placement = "right", trigger = "focus")
))
}
})
### Render UI components generated above
output$binaryDTR1probB <- renderUI({
generateBinaryInputs1B()
})
output$binaryDTR2probB <- renderUI({
generateBinaryInputs2B()
})
### For cell-specific probabilities, render a series of numericInputs labeled by information from DTR substrings
### When DTR1 and DTR2 begin with the same treatment, P(S|stage1trt,r) is rendered only once, in output$cellProbsDTR1B
output$cellProbsDTR1B <- renderUI({
if(substringDTR1B()[1] != substringDTR2B()[1]){
controlInputs <- list(numericInput("marginalFirstStageB1",
label = paste("Probability of success for Path ", substringDTR1B()[2], "r", substringDTR1B()[3], sep = ""),
value = NULL, min = 0, max = 1, step = 0.01),
numericInput("marginalSecondStageNRB1",
label = paste("Probability of success for Path ",substringDTR1B()[2], "nr", substringDTR1B()[4], sep = ""),
value = NULL, min = 0, max = 1, step = 0.01),
bsTooltip(id = "marginalFirstStageB1", title = "Input can range from 0-1 and must be in decimal form, up to two places.",
placement = "right", trigger = "focus"),
bsTooltip(id = "marginalSecondStageNRB1", title = "Input can range from 0-1 and must be in decimal form, up to two places.",
placement = "right", trigger = "focus")
)
}
})
output$cellProbsDTR2B <- renderUI({
list(
numericInput("marginalFirstStageB2", label = paste("Probability of success for Path ", substringDTR2B()[2], "r", substringDTR2B()[3], sep = ""),
value = NULL, min = 0, max = 1, step = 0.01),
numericInput("marginalSecondStageNRB2", label = paste("Probability of success for Path ", substringDTR2B()[2], "nr", substringDTR2B()[4], sep = ""),
value = NULL, min = 0, max = 1, step = 0.01),
bsTooltip(id = "marginalFirstStageB2", title = text.tooltip,
placement = "right", trigger = "focus"),
bsTooltip(id = "marginalSecondStageNRB2", title = text.tooltip,
placement = "right", trigger = "focus"))
})
# Render enabled/disabled numericInputs when outcome is continuous
generateContinuousInputB <- reactive({
validate(
need(input$firstDTRcompareB, text.refDTRPlaceholder),
need(input$secondDTRcompareB, text.compDTRPlaceholder)
)
return(list(numericInput( "effectSizeB", label = "Standardized Effect Size", value = 0, min = 0, max = 10, step = 0.01),
bsTooltip(id = "effectSizeB", title = "Input can range from 0-10 and must be in decimal form, up to two places.",
placement = "right", trigger = "focus"))
)
})
output$continuousProbB <- renderUI({
generateContinuousInputB()
})
##### DESIGN B OBSERVERS #####
### For binary outcome, disallow cell-specific input if target difference input is selected
### If target difference is not selected, also deselect target OR
observe({
if (input$targetDiffCheckB) {
updateCheckboxInput(session, "cellOrConditionalB", value = FALSE)
updateCheckboxInput(session, "targetOddsCheckB", value = FALSE)
}
})
observe({
if (input$targetOddsCheckB) {
updateCheckboxInput(session, "cellOrConditionalB", value = FALSE)
updateCheckboxInput(session, "targetDiffCheckB", value = FALSE)
}
})
### If cell-specific input is selected, deselect target difference and OR options
### Update disabled full DTR success inputs with computed probabilities
observe({
if (input$cellOrConditionalB == TRUE) {
updateCheckboxInput(session, "targetDiffCheckB", value = FALSE)
updateCheckboxInput(session, "targetOddsCheckB", value = FALSE)
updateNumericInput(session, "DTRsuccB1disable", value = generateProbsB()[1])
updateNumericInput(session, "DTRsuccB2disable", value = generateProbsB()[2])
}
}, priority = 2)
### Compute full DTR probabilities when providing cell-specific probabilities
generateProbsB <- reactive({
if (input$selectOutcomeB == 1 && input$cellOrConditionalB == TRUE){
pDTR1 <- fullDTRprob(input$marginalFirstStageB1, input$respB,input$marginalSecondStageNRB1)
pDTR2 <- fullDTRprob(input$marginalFirstStageB2, input$respB,input$marginalSecondStageNRB2)
}
else if (input$selectOutcomeB == 1 && input$cellOrConditionalB == FALSE){
# If not receiving cell-specific probabilities, pass DTR success probabilities
pDTR1 <- input$DTRsuccB1
pDTR2 <- input$DTRsuccB2
}
else {
pDTR1 <- 0
pDTR2 <- 0
}
return(c(pDTR1, pDTR2))
})
##### DESIGN B RESULT BACKEND #####
# Based on provided input probabilities and selected options, compute appropriate arguments to pass to power.prop.test or pwr.norm.test
checkDTRinputsB <- reactive({
return(as.logical((input$DTRsuccB1 > 0) & (input$DTRsuccB1 < 1) & (input$DTRsuccB2 > 0) & (input$DTRsuccB2 < 1)))
})
dataCompilerB <- reactive({
validate(
need(!(is.na(input$respB)), "Please provide a response probability. If unknown, enter 0 for a conservative estimate of power or sample size.") %then%
need(0 <= input$respB && input$respB <= 1, "The provided response probability is not a valid probability. Please enter a value between 0 and 1.")
)
# Binary outcome, provided full-DTR probabilities
if (input$selectOutcomeB == 1 && input$cellOrConditionalB == FALSE && input$targetDiffCheckB == FALSE && input$targetOddsCheckB == FALSE) {
validate(
need(!is.null(input$DTRsuccB1), "Select a Reference AI above."),
need(!is.null(input$DTRsuccB2), "Select a Comparison AI above.") %then%
need(!is.null(input$DTRsuccB1) && !is.null(input$DTRsuccB2), "The success probability is missing for at least one AI. Please provide a numeric input.") %then%
need(input$DTRsuccB1 != input$DTRsuccB2, "Please provide unique success probabilities for each AI. Sample size is indeterminate for equal AI probabilities.") %then%
need(checkDTRinputsB(), "The provided success probability for at least one AI is not a valid probability. Please enter a value between 0 and 1.")
)
return(c(input$DTRsuccB1, input$DTRsuccB2))
}
# Binary outcome, cell-specific probabilities
if (input$selectOutcomeB == 1 && input$cellOrConditionalB == TRUE) {
validate(
need(generateProbsB()[1] != generateProbsB()[2], "The provided cell-specific probabilities yield identical AI-specific probabilities of success.
Sample size is indeterminate for equal AI probabilities. Please adjust your inputs.")
)
return(c(generateProbsB()[1], generateProbsB()[2]))
}
# Binary outcome, target difference in full-DTR probabilities
if (input$selectOutcomeB == 1 && input$targetDiffCheckB == TRUE) {
prob2 <- input$DTRsuccB1 + as.numeric(input$diffDirectionB) * input$targetDiffB
validate(
need(0 < prob2 && prob2 < 1, "Target difference must be less than 0.5 to be valid input"),
need(input$targetDiffB > 0, "Target difference must be greater than 0. Sample size is indeterminate for equal AI probabilities.
Please adjust your inputs.")
)
return(c(input$DTRsuccB1, input$DTRsuccB1 + input$diffDirectionB * input$targetDiffB))
}
# Binary outcome, target odds-ratio relative to reference DTR
if (input$selectOutcomeB == 1 && input$targetOddsCheckB == TRUE) {
validate(
need(is.numeric(input$targetORB), "Please enter an odds ratio.") %then%
need(input$targetORB != 1, "Sample size is indeterminate for an odds ratio of 1. Please enter a different target odds ratio.") %then%
need(input$targetORB != 0, "Sample size is indeterminate for an odds ratio of 0. Please enter a different target odds ratio.")
)
q <- (input$DTRsuccB1 / (1 - input$DTRsuccB1)) * input$targetORB
return(c(input$DTRsuccB1, q / (1 + q)))
}
if(input$selectOutcomeB==2){
validate(
need(input$firstDTRcompareB, "Select a Reference AI above."),
need(input$secondDTRcompareB, "Select a Comparison AI above.") %then%
need(input$effectSizeB != 0, "Sample size is indeterminate for an effect size of 0. Please enter a different target effect size.")
)
return(input$effectSizeB)
}
})