-
Notifications
You must be signed in to change notification settings - Fork 0
/
dementiaSubtypeAnalysis.R
400 lines (332 loc) · 15.4 KB
/
dementiaSubtypeAnalysis.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
library(RPostgreSQL)
library(graph)
library(igraph)
library(tidyverse)
library(ggplot2)
library(stringi)
library(compiler)
library(gsubfn)
# htn 'C0020538', 'C0085580'
#########
# DB setup
#########
drv <- dbDriver('PostgreSQL')
con <-
dbConnect(
drv,
dbname = "causalehr",
host = "localhost",
port = 5432,
user = "scottalexandermalec", # "postgres",
password = "mandarin"
)
#########
# manually defined exclusions
#########
manuallyDefinedExclusions <- c('C0450442', 'C0243077', 'C0021083', 'C0013216', 'C0009429', 'C0199176', 'C0085104',
'C0017296', 'C0597357', 'C1519033', 'C1280903', 'C0003289', 'C0279025', 'C0279025',
'C0040808', 'C0009244', 'C0175723', 'C0011581', 'C0026868', 'C0683465', 'C1519595',
'C0042890', 'C0150321', 'C0581602', 'C1874451', 'C0012854', 'C0033972', 'C0279493',
'C0014935', 'C0086132', 'C0344315', 'C0016452', 'C1137094', 'C2718059', 'C0013806',
'C0183185', 'C0678908', 'C2827774', 'C0031765', 'C0003827', 'C0034991', 'C0581601',
'C0282402', 'C0150593', 'C1254359', 'C3858690', 'C1096024', 'C0020431', 'C1293131',
'C1964029', 'C0204727', 'C0005893', 'C0185027', 'C0439775', 'C0418981', 'C0152060',
'C0183210', 'C0870230', 'C0677850', 'C0376626', 'C0677850', 'C0949766', 'C0150133',
'C1515119', 'C0079613', 'C0455142', 'C0585941', 'C0279494', 'C0920425', 'C0033968',
'C0935576', 'C1527374', 'C1269683', 'C0034618', 'C0376547', 'C0020431', 'C0185117',
'C0204514', 'C0439857', 'C0680038', 'C0846672', 'C0015618', 'C0182537', 'C0183115',
'C0220908', 'C0279494', 'C0935576', 'C1268930', 'C0013218', 'C0034764', 'C0183210',
'C0232164', 'C0281481', 'C1139730', 'C1511300', 'C1512629', 'C2986605', 'C3177188',
'C0030567', 'C0497327', 'C0474169', 'C0011269', 'C0599917', 'C0599917')
length(manuallyDefinedExclusions)
#########
# Covariates that are measured for subjects in sample from electronic health database
#########
getGoodCovars <- function(n) {
fNames <- list.files(path = "data/r3dataset/data")
i = 0
goodData <- c()
for (f in fNames) {
if (sum(read.csv(file = paste("data/r3dataset/data/", f, sep = ""), header = FALSE)) >= n)
{ #print(f)
i = i + 1
#print(i)
goodData <- c(goodData, f)
}
}
#print(goodData)
goodData <- gsubfn(goodData, pattern = "\\.csv", replacement = "")
return(goodData)
}
predicates <- " ('CAUSES', 'PREDISPOSES') " #, 'PREVENTS', 'PREDISPOSES') "
# NOTE: 'TREATS' -- excluding because of the ambiguous causal meaning of the verb 'to treat'
#predicates <- " ('CAUSES', 'PREDISPOSES', 'PREVENTS', 'AFFECTS', 'DISRUPTS', 'AUGMENTS') "
#########
# getConfounders for each dementia subtype
#########
#
getConfounders <- function(predicates, DEMsqldef) {
confounders <- dbGetQuery(con, paste(" WITH ZX AS (SELECT cp.subject_cui AS covar, COUNT(*) AS scnt, cp.subject_name
FROM causalPredications cp, citations c
WHERE cp.pmid = c.pmid
AND CAST(c.pyear AS NUMERIC) >= 2010
AND cp.object_cui IN ('C0011581', 'C1269683', 'C0871546', 'C0041696', 'C0588008', 'C0588006', 'C0221480', 'C0154409', 'C0024517')
AND cp.predicate IN ", predicates, "
GROUP BY subject_cui, subject_name),
ZY AS (SELECT cp.subject_cui AS covar , COUNT(*) AS scnt, cp.subject_name
FROM causalPredications cp, citations c
WHERE cp.pmid = c.pmid
AND CAST(pyear AS NUMERIC) >= 2010
AND cp.object_cui IN ", DEMsqldef,
" AND cp.predicate IN ", predicates, "
GROUP BY cp.subject_cui, cp.subject_name)
SELECT DISTINCT ZX.covar, zx.scnt as theta1, zy.scnt AS theta2, lower(zx.subject_name) as covarname FROM ZX AS zx, ZY AS zy WHERE zx.covar = zy.covar ORDER BY theta2 desc, theta1 desc;", sep = ""))
#write.table(file = "covariates/rawConfounders_for_depression2AD.txt", x = confounders, row.names = FALSE, col.names = FALSE, quote = FALSE)
return(confounders)
}
#########
# getColliders for each dementia subtype
#########
#
getColliders <- function(predicates, DEMsqldef) {
colliders <- dbGetQuery(con, paste(" WITH ZX AS (SELECT cp.object_cui AS covar, COUNT(*) AS scnt, cp.object_name
FROM causalPredications cp, citations c
WHERE cp.pmid = c.pmid
AND CAST(c.pyear AS NUMERIC) >= 2010
AND cp.subject_cui IN ('C0011581', 'C1269683', 'C0871546', 'C0041696', 'C0588008', 'C0588006', 'C0221480', 'C0154409', 'C0024517')
AND cp.predicate IN ", predicates, "
GROUP BY object_cui, object_name),
ZY AS (SELECT cp.object_cui AS covar , COUNT(*) AS scnt, cp.object_name
FROM causalPredications cp, citations c
WHERE cp.pmid = c.pmid
AND CAST(pyear AS NUMERIC) >= 2010
AND cp.subject_cui IN ", DEMsqldef,
" AND cp.predicate IN ", predicates, "
GROUP BY cp.object_cui, cp.object_name)
SELECT DISTINCT ZX.covar, zx.scnt as theta1, zy.scnt AS theta2, lower(zx.object_name) as covarname FROM ZX AS zx, ZY AS zy WHERE zx.covar = zy.covar ORDER BY theta2 desc, theta1 desc;", sep = ""))
#write.table(file = "covariates/rawColliders_for_depression2AD.txt", x = colliders, row.names = FALSE, col.names = FALSE, quote = FALSE)
return(colliders)
}
#########
# Define each dementia subtype by corresponding UMLS CUIs
#########
#
#Dementia type Definition in terms of CUIs in the UMLS
Dsql <- " ('C0497327', 'C0011268') " # hypernym of the subtypes
ADsql <- " ('C0002395') " # Alzheimer's disease
EOADsql <- " ('C0750901', 'C0276496', 'C0751071', 'C0276496') " # Early Onset AD
LOADsql <- " ('C0494463') " # Late Onset AD
FTDsql <- " ('C0338451') " # Frontotemporal Dementia
ADCsql <- " ('C0001849') " # AIDS Dementia Complex
VaDsql <- " ('C0011269', 'C0393561') " # Vascular Dementia
#########
# Get confounders and colliders for each dementia subtype and remove any colliders
#########
#
Dconfounders <- getConfounders(predicates, Dsql)
ADconfounders <- getConfounders(predicates, ADsql)
EOADconfounders <- getConfounders(predicates, EOADsql)
LOADconfounders <- getConfounders(predicates, LOADsql)
FTDconfounders <- getConfounders(predicates, FTDsql)
ADCconfounders <- getConfounders(predicates, ADCsql)
VaDconfounders <- getConfounders(predicates, VaDsql)
#Dcolliders <- getColliders(predicates, Dsql)
#ADcolliders <- getColliders(predicates, ADsql)
#EOADcolliders <- getColliders(predicates, EOADsql)
#LOADcolliders <- getColliders(predicates, LOADsql)
#FTDcolliders <- getColliders(predicates, FTDsql)
#ADCcolliders <- getColliders(predicates, ADCsql)
#VaDcolliders <- getColliders(predicates, VaDsql)
# confounders minus colliders
#setdiff(Dconfounders[,4], Dcolliders[,4])
#colliderConfounders <- intersect(confounders, colliders)
#colliderConfounders <- setdiff(colliderConfounders, mediators)
#########
# Set operations to compare confounders
#########
# intersection of all subtypes
intersect(Dconfounders[,4], intersect(ADconfounders[,4], intersect(EOADconfounders[,4], intersect(LOADconfounders[,4], intersect(FTDconfounders[,4], intersect(ADCconfounders[,4], VaDconfounders[,4]))))))
setdiff(Dconfounders[,4], ADconfounders[,4])
setdiff(ADconfounders[,4], Dconfounders[,4])
intersect(ADconfounders[,4], Dconfounders[,4])
intersect(VaDconfounders[,4], ADconfounders[,4])
setdiff(ADconfounders[,4], VaDconfounders[,4])
setdiff(VaDconfounders[,4], ADconfounders[,4])
Dconfounders
ADconfounders
EOADconfounders
LOADconfounders
FTDconfounders
ADCconfounders
VaDconfounders
#########
# Create Venn diagrams
#########
#
library(VennDiagram)
# your data
Dconfs <- Dconfounders[,4] #c('a','b','c','d')
ADconfs <- ADconfounders[,4] #c('a','e','f','g')
EOADconfs <- EOADconfounders[,4]
FTDconfs <- FTDconfounders[,4]
#ADC <- ADCpvars[,3]
VaDconfs <- VaDconfounders[,4]
# Generate plot
v <- venn.diagram(main = "Distribution of confounders for depression and subtypes of dementia", main.cex = 2,
#sub = "",
list(dementia=Dconfs, AD=ADconfs, EOAD=EOADconfs, FTD=FTDconfs, VaD=VaDconfs),
fill = c("orange", "blue", "red", "cyan", "yellow"),
alpha = c(0.5, 0.5, 0.5, 0.5, 0.5), cat.cex = 1, cex=1,
filename=NULL)
# have a look at the default plot
grid.newpage()
grid.draw(v)
# have a look at the names in the plot object v
lapply(v, names)
# We are interested in the labels
lapply(v, function(i) i$label)
# Over-write labels (5 to 7 chosen by manual check of labels)
# in foo only
v[[5]]$label <- paste(setdiff(foo, baa), collapse="\n")
# in baa only
v[[6]]$label <- paste(setdiff(baa, foo) , collapse="\n")
# intesection
v[[7]]$label <- paste(intersect(foo, baa), collapse="\n")
# plot
grid.newpage()
grid.draw(v)
getMediators <- function(predicates, DEMsqldef) {
mediators <- dbGetQuery(con, paste(" WITH ZX AS (SELECT cp.object_cui AS covar, COUNT(*) AS scnt, cp.object_name as covarname
FROM causalPredications cp, citations c
WHERE cp.pmid = c.pmid
AND CAST(c.pyear AS NUMERIC) >= 2010
AND cp.subject_cui IN ('C0011581', 'C1269683', 'C0871546', 'C0041696', 'C0588008', 'C0588006', 'C0221480', 'C0154409', 'C0024517')
AND cp.predicate IN ", predicates, "
GROUP BY object_cui, object_name),
ZY AS (SELECT cp.subject_cui AS covar , COUNT(*) AS scnt, cp.subject_name as covarname
FROM causalPredications cp, citations c
WHERE cp.pmid = c.pmid
AND CAST(pyear AS NUMERIC) >= 2010
AND cp.object_cui IN ", DEMsqldef,
" AND cp.predicate IN ", predicates, "
GROUP BY cp.subject_cui, cp.subject_name)
SELECT DISTINCT ZX.covar, zx.scnt as theta1, zy.scnt AS theta2, lower(zx.covarname) as covarname FROM ZX AS zx, ZY AS zy WHERE zx.covar = zy.covar ORDER BY theta2 desc, theta1 desc;", sep = ""))
# write.table(file = "covariates/rawMediators_for_depression2AD.txt", x = mediators, row.names = FALSE, col.names = FALSE, quote = FALSE)
return(mediators)
}
getPrecisionVars <- function(predicates, DEMsqldef) {
precisionVars <- dbGetQuery(con, paste(" WITH ZX AS (SELECT cp.subject_cui AS covar , COUNT(*) AS scnt, cp.subject_name as covarname
FROM causalPredications cp, citations c
WHERE cp.pmid = c.pmid
AND CAST(pyear AS NUMERIC) >= 2010
AND cp.object_cui IN ", DEMsqldef,
" AND cp.predicate IN ", predicates, "
GROUP BY cp.subject_cui, cp.subject_name)
SELECT DISTINCT ZX.covar, zx.scnt as theta1, lower(zx.covarname) as covarname FROM ZX AS zx ORDER BY theta1 desc;", sep = ""))
# write.table(file = "covariates/rawPrecisionVars_for_AD.txt", x = precisionVars, row.names = FALSE, col.names = FALSE, quote = FALSE)
return(precisionVars)
}
#########
# Get Mediators for each dementia subtype and remove any colliders
#########
#
Dmeds <- getMediators(predicates, Dsql)
ADmeds <- getMediators(predicates, ADsql)
EOADmeds <- getMediators(predicates, EOADsql)
LOADmeds <- getMediators(predicates, LOADsql)
FTDmeds <- getMediators(predicates, FTDsql)
ADCmeds <- getMediators(predicates, ADCsql)
VaDmeds <- getMediators(predicates, VaDsql)
# your data
Dm <- Dmeds[,4] #c('a','b','c','d')
ADm <- ADmeds[,4] #c('a','e','f','g')
EOADm <- EOADmeds[,4]
FTDm <- FTDmeds[,4]
#ADC <- ADCpvars[,3]
VaDm <- VaDmeds[,4]
# Generate plot
v <- venn.diagram(main = "Distribution of mediators for depression and subtypes of dementia", main.cex = 2,
#sub = "",
list(dementia=Dm, AD=ADm, EOAD=EOADm, FTD=FTDm, VaD=VaDm),
fill = c("orange", "blue", "red", "cyan", "yellow"),
alpha = c(0.5, 0.5, 0.5, 0.5, 0.5), cat.cex = 1, cex=1,
filename=NULL, force.unique = TRUE, print.mode = "percent")
# have a look at the default plot
grid.newpage()
grid.draw(v)
setdiff(ADm, c(Dm, VaDm, FTDm, EOADm))
setdiff(Dm, c(ADm, FTDm, VaDm, EOADm))
setdiff(VaDm, c(Dm, ADm, FTDm, EOADm))
setdiff(FTDm, c(Dm, ADm, VaDm, EOADm))
setdiff(intersect(Dm, intersect(FTDm, VaDm)), c(ADm, EOADm))
#########
# Get Precision Variables for each dementia subtype and remove any colliders
#########
#
Dpvars <- getPrecisionVars(predicates, Dsql)
ADpvars <- getPrecisionVars(predicates, ADsql)
EOADpvars <- getPrecisionVars(predicates, EOADsql)
LOADpvars <- getPrecisionVars(predicates, LOADsql)
FTDpvars <- getPrecisionVars(predicates, FTDsql)
ADCpvars <- getPrecisionVars(predicates, ADCsql)
VaDpvars <- getPrecisionVars(predicates, VaDsql)
#####
# your data
Dpv <- Dpvars[,3] #c('a','b','c','d')
ADpv <- ADpvars[,3] #c('a','e','f','g')
EOADpv <- EOADpvars[,3]
FTDpv <- FTDpvars[,3]
#ADC <- ADCpvars[,3]
VaDpv <- VaDpvars[,3]
# Generate plot
v <- venn.diagram(list(dementia=Dpv, AD=ADpv, EOAD=EOADpv, FTD=FTDpv, VaD=VaDpv),
fill = c("orange", "blue", "red", "cyan", "yellow"),
alpha = c(0.5, 0.5, 0.5, 0.5, 0.5), cat.cex = 1, cex=1,
filename=NULL)
# have a look at the default plot
grid.newpage()
grid.draw(v)
intersect(Dpv, intersect(ADpv, intersect(EOADpv, intersect(FTDpv, VaDpv))))
setdiff(FTDpv, intersect(ADpv, intersect(EOADpv, intersect(Dpv, VaDpv))))
setdiff(EOADpv, unique(c(ADpv, EOADpv, Dpv, VaDpv)))
# unique Pvars for each subtype
setdiff(VaDpv, c(Dpv, ADpv, FTDpv, EOADpv))
setdiff(FTDpv, c(Dpv, ADpv, VaDpv, EOADpv))
# have a look at the names in the plot object v
lapply(v, names)
# We are interested in the labels
lapply(v, function(i) i$label)
# Over-write labels (5 to 7 chosen by manual check of labels)
# in foo only
v[[5]]$label <- paste(setdiff(foo, baa), collapse="\n")
# in baa only
v[[6]]$label <- paste(setdiff(baa, foo) , collapse="\n")
# intesection
v[[7]]$label <- paste(intersect(foo, baa), collapse="\n")
# plot
grid.newpage()
grid.draw(v)
v[[6]]$label
nameLookup <- function(cuis) {
covnames <- c()
for (cui in cuis) {
meaning <- dbGetQuery(con, paste("select distinct lower(covname) as covname from causalconcepts cc where cc.cui = '", cui, "'; ", sep = ""))
#print(meaning[,1])
covnames <- c(covnames, meaning[,1])
}
unlist(covnames)
}
#rawConfounders <- cuiLookup(confounders)
######
# look up UMLS CUI given UMLS preferred name as input
######
cuiLookup <- function(covnames) {
covcuis <- c()
for (covname in covnames) {
covcui <- dbGetQuery(con, paste("select distinct cui from causalconcepts cc where cc.covname LIKE '%", covname, "%'; ", sep = ""))
#print(meaning[,1])
covcuis <- c(covcuis, covcui[,1])
}
unlist(covcuis)
}
cuiLookup(rawConfounders)