-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathCHD_premetas.r
1014 lines (790 loc) · 39.3 KB
/
CHD_premetas.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
#Transcribed script from main CHD script for the generation of the premeta files going into the final analysis
#3/12/18: So far we have the following studies:
#UKBB - HAIL adjusted
#EPIC
#Cardiogram - 1KG
#Cardiogram - Exome
#Cardiogram - Metabo
#GerMIFs1-3,4-7
#deCODE
#Greek
#HUNT
#Partners
#TIMI
#This is the scripts to making each of the premeta files from the raw input files
rm(list=ls())
options(stringsAsFactors = FALSE)
library("data.table", lib.loc="/home/tj241/R_Packages")
source("/scratch/tj241/demo/Test/Manhattan_source.R")
source("/scratch/tj241/demo/Test/Manhattan/turboman_source.R")
####################
#UKBB - HAIL
####################
input=fread(paste0("/scratch/asb38/ukbiobank/CAD.INTERMEDIATE.allvariants.birthYear.sex.multiethnic.HRC.UK10K.1000G.no.varfilter.rel.excl.sel.3rd.deg.logreg.052318_chr1_22.tsv"), colClasses=c("character"))
input=as.data.frame(input)
temp=strsplit(input[,1], "[:]")
temp2=do.call(rbind, temp)
#Second allele is effect Allele
baseline=temp2[,3]
effect=temp2[,4]
CHR=as.numeric(temp2[,1])
POS=temp2[,2]
#MAF column is actually EAF
EAF=as.numeric(input[,5])
INFO=as.numeric(input[,4])
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
output=cbind(SNPID, CHR,POS, baseline, effect, input[,c(8,9,11)], EAF, INFO)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "INFO")
output=output[order(as.numeric(output[,2]), as.numeric(output[,3])),]
#There are perfect dupes with different RSIDs, fine to just drop one copy
output=unique(output)
#We have NA's, remove them
output=output[-which(is.na(output[,10])==T),]
#93095623 down to 93068690
#Filter INFO 0.4
output=output[as.numeric(output[,10])>0.4,]
#down to 68576130
#Filter MAF 0.00005
output=output[which(as.numeric(output[,9])>0.00005&as.numeric(output[,9])<0.99995),]
#down to 47902336
#Stitch on cases, 33206
Cases=rep(33206, length(output[,1]))
Effective_Cases=as.numeric(output[,10])*Cases
output=cbind(output, Cases, Effective_Cases)
#Resolve duplicates
dupes=output[output[,1]%in%output[duplicated(output[,1]),1],]
#These are all indels with flipped alleles which represent different variants. However, the only way to resolve them is to come up with a new naming system and apply it to every other study
#Decision: just drop all of these variants
output=output[-which(output[,1]%in%dupes[,1]),]
#down to 47900528
#We have 1044 with NA for betas/se/pval
output=output[-which(is.na(output[,6])==T|is.na(output[,7])==T|is.na(output[,8])==T),]
#47899484 left
#output[as.numeric(output[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(output[,6]))
#summary(as.numeric(output[,7]))
#summary(as.numeric(output[,8]))
write.table(output, paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/HAIL_adjusted_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
####################
#UKBB - SAIGE
####################
options(scipen=999)
#Note, no scientific notation as they messed up their positions in the raw files
#Don't convert p-values, etc to numeric as they won't be in scientific notation
input=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/CAD_intermediate_GWAS_UK10K_allchr.SAIGE.txt", colClasses=c("character"))
input=as.data.frame(input)
#summary(pmin(nchar(input[,4]), nchar(input[,5])))
#temp=which(pmin(nchar(input[,4]), nchar(input[,5]))>1)
#Somehow we have complex indels
#ALLELE2 is the effect allele, the allele2af and allele2ac both refer to it so I can use as is
#Just leave complex indels in as is
baseline=toupper(input[,4])
effect=toupper(input[,5])
CHR=as.numeric(input[,1])
POS=as.numeric(input[,2])
#Position is numeric. However, no scientific notation so it's OK
#MAF column is actually EAF
EAF=as.numeric(input[,7])
#We need to get INFO from HAIL results, let's filter first
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
Cases=rep(33941, length(baseline))
N=rep(472335, length(baseline))
output=cbind(SNPID, CHR,POS, baseline, effect, input[,c(9,10,12)], EAF, Cases, N)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "Cases", "N")
output=output[order(as.numeric(output[,2]), as.numeric(output[,3])),]
#Must sort out duplicates before filters
print(dim(output))
output=unique(output)
print(dim(output))
#Now the flipped indels dupes
dupes=output[output[,1]%in%output[duplicated(output[,1]),1],]
print(dim(dupes))
output=output[-which(output[,1]%in%dupes[,1]),]
print(dim(output))
#Now I can apply filters
#Start with MAF
output=output[which(as.numeric(output[,9])>0.00005&as.numeric(output[,9])<0.99995),]
print(dim(output))
write.table(output, "/scratch/tj241/demo/Test/UKBB500K/CHD/SAIGE_500k_temporary.tsv", quote=F, row.names=F, col.names=T, sep="\t")
#dupes sorted and MAF filtered
#Time to append INFO
ref=fread(paste0("awk '{print $1,$4;}' /scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/CAD.INTERMEDIATE.allvariants.birthYear.sex.multiethnic.HRC.UK10K.1000G.no.varfilter.rel.excl.sel.3rd.deg.logreg.052318_chr1_22.tsv"), colClasses=c("character"))
ref=as.data.frame(ref)
temp=strsplit(ref[,1], "[:]")
temp2=do.call(rbind, temp)
baseline=temp2[,3]
effect=temp2[,4]
CHR=as.numeric(temp2[,1])
POS=as.numeric(temp2[,2])
#Again, pos is numeric since we set up no scientific notation earlier.
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
ref[,1]=SNPID
colnames(ref)[1]="ALT_ID"
ref=unique(ref)
infopass=ref[which(as.numeric(ref[,2])>0.4),]
#infopass has duplicated IDs but these should be indels with flipped alleles, which have been removed entirely from output
output=output[output[,1]%in%infopass[,1],]
print(dim(output))
output2=merge(output, infopass, by="ALT_ID")
print(dim(output2))
finaloutput=output2[which(as.numeric(output2[,12])>0.4),]
Effective_Cases=as.numeric(finaloutput[,12])*as.numeric(finaloutput[,10])
finaloutput=cbind(finaloutput, Effective_Cases)
finaloutput=finaloutput[,c(1:9,12,10,13,11)]
print(head(finaloutput[as.numeric(finaloutput[,3])%%1000000==0,]))
#Check for NA beta/se/pval
print(summary(as.numeric(finaloutput[,6])))
print(summary(as.numeric(finaloutput[,7])))
print(summary(as.numeric(finaloutput[,8])))
write.table(finaloutput, "/scratch/tj241/demo/Test/UKBB500K/CHD/UKBB_SAIGE_premeta.tsv", quote=F, row.names=F, col.names=T, sep="\t")
#This takes an age to run, must submit it as job and leave it
#system(paste0("sbatch -N 1 -n 1 -c 10 -p long -J SAIGE -t 48:0:0 -o /scratch/tj241/demo/Test/Epic/UKBB_CHD_meta/Temp/saige.slurm /scratch/tj241/demo/Test/Epic/UKBB_CHD_meta/Temp/saige.sh"))
#####################
#Epic
####################
input=fread("/scratch/tj241/demo/Test/Epic/UKBB_CHD_meta/EPIC_CHD_gwas.tsv", colClasses=c("character"))
input=as.data.frame(input)
#input[as.numeric(input[,4])%%1000000==0,]
#this raw input is fine
files=c("/scratch/tj241/demo/Test/Epic/Incident_CHD/hce_cases", "/scratch/tj241/demo/Test/Epic/Incident_CHD/hce_noncases", "/scratch/tj241/demo/Test/Epic/Incident_CHD/quad_cases", "/scratch/tj241/demo/Test/Epic/Incident_CHD/quad_noncases")
temp1=fread(paste0(files[2], ".snpstats"))
temp1=as.data.frame(temp1)
temp2=fread(paste0(files[4], ".snpstats"))
temp2=as.data.frame(temp2)
discordance=cbind(temp1[,2], abs(as.numeric(temp1[,15])-as.numeric(temp2[,15])), temp1[,15], temp2[,15])
colnames(discordance)=c("rsid", "discordance","HCE_MAF", "QUAD_MAF")
output=merge(input, discordance, by="rsid", all.x=T, sort=F)
relative_discordance=as.numeric(output[,47])/as.numeric(output[,29])
output=cbind(output, relative_discordance)
ratio=pmin(as.numeric(output[,48])/as.numeric(output[,49]), as.numeric(output[,49])/as.numeric(output[,48]))
output=cbind(output, ratio)
#write.table(output, "/scratch/tj241/demo/Test/Epic/UKBB_CHD_meta/EPIC_CHD_gwas_full.tsv", quote=F, row.names=F, col.names=T, sep="\t")
#output=fread("/scratch/tj241/demo/Test/Epic/UKBB_CHD_meta/EPIC_CHD_gwas_full.tsv", colClasses=c("character"))
#output=as.data.frame(output)
#output[as.numeric(output[,4])%%1000000==0,]
#This is also correct
#In SNPtest, ALLELEB is the effect allele
baseline=output[,5]
effect=output[,6]
#no indels
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(output[,3], ":", output[,4], "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(output[which(flip==1),3], ":", output[which(flip==1),4], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
EAF=(2*as.numeric(output[,16])+as.numeric(output[,15]))/2/(as.numeric(output[,14])+as.numeric(output[,15])+as.numeric(output[,16]))
output2=cbind(SNPID, output[,3],output[,4], baseline, effect, output[,c(44,45,42,9,29,47,50,51)], EAF)
colnames(output2)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "INFO", "MAF", "Discordance", "Rel_Discordance", "Ratio", "EAF")
output3=output2[which(as.numeric(output2[,9])>0.7&as.numeric(output2[,10])>0.01&as.numeric(output2[,11])<0.05&as.numeric(output2[,12])<0.07&as.numeric(output2[,13])>0.9),]
#Final filters we actually applied
#INFO>0.7
#MAF>0.01
#Discordance<0.05
#Rel Discordance<0.07
#Ratio>0.9
#5.55M Variants remaining of the original 11.77M
#5703735 Variants actually, corrected now
#stitch on number of cases, 10365 for all SNPs
Cases=rep(10365, length(output3[,1]))
Effective_Cases=as.numeric(output3[,9])*Cases
N=rep(22430, length(output3[,1]))
output3=cbind(output3, Cases, Effective_Cases, N)
#output3[as.numeric(output3[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(output3[,6]))
#summary(as.numeric(output3[,7]))
#summary(as.numeric(output3[,8]))
write.table(output3[,c(1:8,14, 9, 15, 16, 17)], "/scratch/tj241/demo/Test/UKBB500K/CHD/EPIC_premeta.tsv", quote=F, row.names=F, col.names=T, sep="\t")
####################
#Cardiogram - 1KG
####################
cardiogram=fread("/scratch/asb38/ukbiobank/results/cad_1000g_results_for_metal.txt", colClasses=c("character"))
cardiogram=as.data.frame(cardiogram)
cardiogram=cardiogram[order(as.numeric(cardiogram[,2]), as.numeric(cardiogram[,3])),]
allelemapping=fread("/scratch/asb38/ukbiobank/ALL_1000G_phase1integrated_v3.legend", colClasses=c("character"))
allelemapping=as.data.frame(allelemapping)
#Pretty sure first allele is "other allele"
colnames(allelemapping)=c("CHR", "SNPID", "POS", "other_allele", "reference_allele")
snps=c("A","T","G","C")
nonindels=cardiogram[-which(cardiogram[,4]=="D"|cardiogram[,5]=="D"),]
indels=cardiogram[which(cardiogram[,4]=="D"|cardiogram[,5]=="D"),]
baseline=nonindels[,5]
effect=nonindels[,4]
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
CHR=nonindels[,2]
POS=nonindels[,3]
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
nonindels=cbind(SNPID, nonindels)
baseline=allelemapping[,4]
effect=allelemapping[,5]
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
CHR=allelemapping[,1]
POS=allelemapping[,3]
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
allelemapping=cbind(SNPID, allelemapping)
temp=substr(allelemapping[,3],1,5)
fails=allelemapping[which(temp=="MERGE"),]
allelemapping3=allelemapping[-which(allelemapping[,3]%in%fails[,3]),]
colnames(allelemapping3)[3]="RSID"
#all cardiogram IDs are unique
#all allelemapping3 IDs are unique
finalnonindels=merge(nonindels, allelemapping3, by="SNPID")
#all matched and found
#We only need to fix the indels
#So we fix indels and just stick it at the end of nonindels then reorder
newid=paste0(indels[,2], ":", indels[,3])
indels=cbind(newid, indels)
newid=paste0(allelemapping[,2], ":", allelemapping[,4])
allelemapping=cbind(newid, allelemapping)
#trim allelemapping down to indels as well
allelemapping2=allelemapping[-which(allelemapping[,6]%in%snps&allelemapping[,7]%in%snps),]
#write.table(allelemapping2, "/scratch/tj241/demo/Test/UKBB500K/CHD/mapping_indels.tsv", quote=F, row.names=F, col.names=T, sep="\t")
#remove variants which begins with MERGED_DEL_...
#Plan is to remove all of them and then merge whatever I can
temp=substr(allelemapping2[,4],1,5)
fails=allelemapping2[which(temp=="MERGE"),]
allelemapping3=allelemapping2[-which(allelemapping2[,4]%in%fails[,4]),]
#Just merge it in
finalindel=merge(indels, allelemapping3, by="newid")
#lose 1534 indels
#Miraculously there are no "-" notation SNPs anymore!! Let's keep them this way for now then.
finalnonindels=finalnonindels[,2:18]
finalindel=finalindel[,c(2:13,15:19)]
colnames(finalindel)[14]="RSID"
#Process nonindels first
baseline=finalnonindels[,5]
effect=finalnonindels[,4]
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
CHR=finalnonindels[,2]
POS=finalnonindels[,3]
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
finalnonindels=cbind(SNPID, finalnonindels)
#write.table(finalnonindels, "/scratch/tj241/demo/Test/UKBB500K/CHD/Cardiogram_nonindels.tsv", quote=F, row.names=F, col.names=T, sep="\t")
#Now to process the indels
#all of them have del as 1 char
baseline=finalindel[,5]
effect=finalindel[,4]
flip=nchar(finalindel[,16])==1
SNPID=paste0(finalindel[,2], ":", finalindel[,3], "_", finalindel[,17], "_", finalindel[,16])
SNPID[which(flip==T)]=paste0(finalindel[which(flip==T),2], ":", finalindel[which(flip==T),3], "_", finalindel[which(flip==T),16], "_", finalindel[which(flip==T),17])
finalindel=cbind(SNPID, finalindel)
deletion=pmin(finalindel[,17], finalindel[,18])
insertion=pmax(finalindel[,17], finalindel[,18])
newbaseline=deletion
newbaseline[which(finalindel[,6]=="I")]=insertion[which(finalindel[,6]=="I")]
neweffect=deletion
neweffect[which(finalindel[,5]=="I")]=insertion[which(finalindel[,5]=="I")]
finalindel[,6]=newbaseline
finalindel[,5]=neweffect
final=rbind(finalnonindels, finalindel)
final=final[order(final[,3], final[,4]),]
#write.table(final, "/scratch/tj241/demo/Test/UKBB500K/CHD/Cardiogram_full.tsv", quote=F, row.names=F, col.names=T, sep="\t")
#final=fread("/scratch/tj241/demo/Test/UKBB500K/CHD/Cardiogram_full.tsv", colClasses=c("character"))
#final=as.data.frame(final)
mapping=fread("/scratch/asb38/ukbiobank/snp.counts.txt1")
mapping=as.data.frame(mapping)
#apparantly all variants' RSIDs were found in the mapping file
colnames(mapping)[1]="RSID"
finaloutput=merge(final, mapping, by="RSID")
finaloutput2=finaloutput[,c(2,4,5,7,6,10:12,8,9,19)]
Effective_Cases=as.numeric(finaloutput[,9])*as.numeric(finaloutput[,19])
finaloutput2=cbind(finaloutput2, Effective_Cases)
colnames(finaloutput2)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "INFO", "Cases", "Effective_Cases")
finaloutput2=finaloutput2[order(as.numeric(finaloutput2[,2]), as.numeric(finaloutput2[,3])),]
write.table(finaloutput2, "/scratch/tj241/demo/Test/UKBB500K/CHD/Cardiogram_premeta.tsv", quote=F, row.names=F, col.names=T, sep="\t")
####################
#Cardiogram - Exome
####################
input=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/MICAD.NA.EUR.ExA.Consortium.logistic.201404231.FINAL.txt", colClasses=c("character"))
input=as.data.frame(input)
missing=input[is.na(input[,23])==T|is.na(input[,22])==T,]
praveen1=fread("/scratch/backup/bioinformatics/illumina_manifests/exome_plus/version_2/5_other_output/version_1-0_restore/3_summary_files/v1_0_restore_info.csv", colClasses=c("character"))
praveen1=as.data.frame(praveen1)
colnames(praveen1)[4]="MarkerName"
test1=merge(missing, praveen1, by="MarkerName")
#perfect mapping file, but not for indel alleles
notmissing=input[-which(is.na(input[,23])==T|is.na(input[,22])==T),]
temp=test1[,c(1,26,27)]
tempoutput=merge(missing, temp,by="MarkerName")
tempoutput[,22]=tempoutput[,25]
tempoutput[,23]=tempoutput[,26]
#chr/pos sorted
output=rbind(notmissing, tempoutput[,1:24])
#write.table(output, "/scratch/tj241/demo/Test/UKBB500K/CHD/exome_temp_chrpos.tsv", quote=F, row.names=F, col.names=T, sep="\t")
indels=output[output[,2]=="d",]
#Pull chr/pos from 1kg files
nonindels=output[-which(output[,2]=="d"),]
test2=merge(indels, praveen1, by="MarkerName")
#Attempt to extract indel info from the strands
temp=gsub("(.*)(.)(\\[)(\\-)(\\/)(.+)(])(.*)", "\\2\\3\\4\\5\\6\\7", test2[,35])
#This is the correct one to use as I checked all 4 and they only differ for the last indel which was manually checked vs ukbb
test2[,2]=gsub("(.*)(.)(\\[)(\\-)(\\/)(.+)(])(.*)", "\\2", test2[,35])
test2[,3]=gsub("(.*)(.)(\\[)(\\-)(\\/)(.+)(])(.*)", "\\2\\6", test2[,35])
test2[,23]=as.numeric(test2[,23])-1
#Must make numeric!!
#Check perfect round positions at the end!!
exomefull=rbind(nonindels, test2[,1:24])
#This had better work
input=exomefull
#drop nonautosomal
input=input[-which(input[,22]%in%c("M", "X", "XY")),]
CHR=input[,22]
POS=input[,23]
baseline=toupper(input[,3])
effect=toupper(input[,2])
#Allele1 is assumed to be the effect allele until I get confirmation from Adam
#Confirmed
EAF=as.numeric(input[,4])
Cases=as.numeric(input[,20])
N=as.numeric(input[,21])+Cases
INFO=rep(1, length(Cases))
Effective_Cases=Cases
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
output=cbind(SNPID, CHR,POS, baseline, effect, input[,c(8:10)], EAF,INFO, Cases,Effective_Cases, N)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "INFO","Cases","Effective_Cases", "N")
output=output[order(as.numeric(output[,2]), as.numeric(output[,3])),]
#There are perfect dupes with different RSIDs, fine to just drop one copy
output=unique(output)
#184249
#Hold up, there's duplicated variants here, same variant but different sample size/beta/se/pval/MAF. Need to filter down to unique set.
#we drop one with the fewest cases
dupes=output[duplicated(output[,1])==T, 1]
nondupes=output[-which(output[,1]%in%dupes),]
duplicates=output[which(output[,1]%in%dupes),]
dupesoutput=NULL
removal=NULL
for(i in 1:length(unique(duplicates[,1])))
{
temp=duplicates[duplicates[,1]==unique(duplicates[,1])[i],]
if(length(temp[,1])!=2)print(paste0("Different Number: ",length(temp[,1])))
#all are pairs
if(temp[1,10]==temp[2,10]&temp[1,11]==temp[2,11])
{
print(paste0("Same cases and controls: ",i))
removal=c(removal, i)
}
temp=temp[order(-as.numeric(temp[,10]), -as.numeric(temp[,11])),]
tempoutput=temp[1,]
dupesoutput=rbind(dupesoutput, tempoutput)
}
#This still doesnt work, we have duplicated variants with the same chr/pos/alleles/cases/controls but somehow different MAF/effect/pval
#Just drop them
dupesoutput=dupesoutput[-removal,]
#29 lost
finaloutput=rbind(nondupes, dupesoutput)
finaloutput=finaloutput[order(as.numeric(finaloutput[,2]), as.numeric(finaloutput[,3])),]
#183644
#finaloutput[as.numeric(finaloutput[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(finaloutput[,6]))
#summary(as.numeric(finaloutput[,7]))
#summary(as.numeric(finaloutput[,8]))
write.table(finaloutput, paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/Exome_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
####################
#Cardiogram - Metabo
####################
input=fread("/scratch/curated_genetic_data/association_results/chd/EUR_ancestry_metabochip_79K_phenoscanner_april2017.tab", colClasses=c("character"))
input=as.data.frame(input)
CHR=input[,1]
POS=input[,3]
baseline=toupper(input[,6])
effect=toupper(input[,5])
#No indels
#Allele1 is effect allele
EAF=as.numeric(input[,7])
INFO=rep(1,length(effect))
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
totalsamples=as.numeric(input[,19])
#Getting cases too hard, we never got it for a couple of studies
#We will apply constant ratio of case:control given total sample size
#case:control = 73856:147712 from spreadsheet
Cases=totalsamples*73856/(147712+73856)
Effective_Cases=Cases
output=cbind(SNPID, CHR,POS, baseline, effect, input[,c(11:13)], EAF, INFO, Cases, Effective_Cases, totalsamples)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "INFO","Cases","Effective_Cases", "N")
output=output[order(as.numeric(output[,2]), as.numeric(output[,3])),]
#There are perfect dupes with different RSIDs, fine to just drop one copy
output=unique(output)
#79037
#output[as.numeric(output[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(output[,6]))
#summary(as.numeric(output[,7]))
#summary(as.numeric(output[,8]))
write.table(output, paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/Metabo_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
####################
#GerMIFs1-7 - not 3
####################
input1=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/GWAS_HRC.GerMIFSI.EUR.CAD.06072017.LZ.tab", colClasses=c("character"))
input1=as.data.frame(input1)
input2=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/GWAS_HRC.GerMIFSII.EUR.CAD.06072017.LZ.tab", colClasses=c("character"))
input2=as.data.frame(input2)
input3=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/GWAS_HRC.GerMIFSIV.EUR.CAD.06072017.LZ.tab", colClasses=c("character"))
input3=as.data.frame(input3)
input4=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/GWAS_HRC.GerMIFSV.EUR.CAD.06072017.LZ.tab", colClasses=c("character"))
input4=as.data.frame(input4)
input5=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/GWAS_HRC.GerMIFSVI.EUR.CAD.12122017.LZ.tab", colClasses=c("character"))
input5=as.data.frame(input5)
input6=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/GWAS_HRC.GerMIFSVII.EUR.CAD.13062018.LL.tab", colClasses=c("character"))
input6=as.data.frame(input6)
#GerMIFS6 has problems, need an incl list
#incllist=read.table("/scratch/tj241/demo/Test/UKBB500K/CHD/GerMIFS6_incl_list.txt", header=F)
#keeplist is 2830433
#only found 2796168 of the 5108393 total
#input2=input2[which(input2[,1]%in%incllist[,1]),]
#Leave GerMIFs6 as is and just flag all gwsig vars into a watch list, also add the chr1 false positive from EPIC to that watch list
summary(as.factor(input1[,6]))
summary(as.factor(input1[,7]))
summary(as.factor(input2[,6]))
summary(as.factor(input2[,7]))
summary(as.factor(input3[,6]))
summary(as.factor(input3[,7]))
summary(as.factor(input4[,6]))
summary(as.factor(input4[,7]))
summary(as.factor(input5[,6]))
summary(as.factor(input5[,7]))
summary(as.factor(input6[,6]))
summary(as.factor(input6[,7]))
#No indels on any of them
input=NULL
input[[1]]=input1
input[[2]]=input2
input[[3]]=NULL
input[[4]]=input3
input[[5]]=input4
input[[6]]=input5
input[[7]]=input6
numbers=c(622,1188,"NA",940,2392, 1639, 3062)
totals=c(2143,2426,"NA",2068,3929, 2825, 6524)
for(i in c(1,2,4:7))
{
baseline=input[[i]][,6]
effect=input[[i]][,7]
#no indels
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(input[[i]][,4], ":", input[[i]][,5], "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(input[[i]][which(flip==1),4], ":", input[[i]][which(flip==1),5], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
#B is the effect allele
EAF=(2*as.numeric(input[[i]][,10])+as.numeric(input[[i]][,9]))/2/(as.numeric(input[[i]][,8])+as.numeric(input[[i]][,9])+as.numeric(input[[i]][,10]))
output=cbind(SNPID, input[[i]][,4],input[[i]][,5], baseline, effect, input[[i]][,c(14:17)], EAF)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "INFO", "EAF")
output2=output[which(output[,9]>0.4),]
#INFO filter 0.4
Cases=rep(numbers[i], length(output2[,1]))
Effective_Cases=as.numeric(output2[,9])*as.numeric(Cases)
N=rep(totals[i], length(output2[,1]))
output2=cbind(output2, Cases, Effective_Cases, N)
output2=output2[order(as.numeric(output2[,2]), as.numeric(output2[,3])),]
#output2[as.numeric(output2[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(output2[,6]))
#summary(as.numeric(output2[,7]))
#summary(as.numeric(output2[,8]))
#summary(as.numeric(output2[,10]))
#summary(as.numeric(output2[,9]))
write.table(output2[,c(1:8,10,9,11,12,13)], paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/GerMIFS",i,"_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
}
#####################
#deCODE
#####################
#Now for deCODE
#There are complex indels, cannot do anything
#Just process everything for now unless decode gets back to us
#WARNING! COMPLEX INDELS ARE CODED ALPHABETICALLY! Consequently they are not necessarily shorter indel first.
input=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/DECODE.Coronary_Artery_Disease_All_06112017.18122017.txt", colClasses=c("character"))
input=as.data.frame(input)
input[,4]=substr(input[,4],4,nchar(input[,4]))
#Get rid of x chromosome
input=input[-which(input[,4]=="X"),]
baseline=input[,6]
effect=input[,7]
#Allele B is effect allele
EAF=as.numeric(input[,8])
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(input[,4], ":", input[,5], "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(input[which(flip==1),4], ":", input[which(flip==1),5], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
output=cbind(SNPID, input[,4],input[,5], baseline, effect, input[,c(12:15)], EAF)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "INFO", "EAF")
finaloutput=output[which(output[,9]>0.4),]
#Stitch on cases, 38918
Cases=rep(38918, length(finaloutput[,1]))
Effective_Cases=as.numeric(finaloutput[,9])*Cases
N=rep(356875, length(finaloutput[,1]))
finaloutput=cbind(finaloutput, Cases, Effective_Cases, N)
#write.table(finaloutput, paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/deCODE_unfiltered_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
#Remove imperfect duplicates completely
finaloutput=finaloutput[-which(finaloutput[,1]%in%finaloutput[duplicated(finaloutput[,1]),1]),]
#3486 dropped
#They prefiltered by INFO score to 0.8
#Apply MAF filter 0.00005
#finaloutput=temp
finaloutput=finaloutput[which(as.numeric(finaloutput[,10])>0.00005&as.numeric(finaloutput[,10])<0.99995),]
#from 18398244 to 16595653
#finaloutput[as.numeric(finaloutput[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(finaloutput[,6]))
#summary(as.numeric(finaloutput[,7]))
#summary(as.numeric(finaloutput[,8]))
write.table(finaloutput[,c(1:8,10,9,11,12,13)], paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/deCODE_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
#Note, complex indels not sorted but no other study have them so they're fine as is
#Note, 1k variants are duplicated perfectly but different EAF. This suggests different population but METAL automatically drops second one systematically. Must get confirmation from decode.
#Decision: just drop them since no response from deCODE
#####################
#Greek
#####################
#Now for Greek Dataset
input=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/GWAS_HRC.GREEK.DATASET.CAD.EUR.12122017.OG.txt", colClasses=c("character"))
input=as.data.frame(input)
#MAF=pmin((2*input[,6]+input[,7])/2/(input[,6]+input[,7]+input[,8]),(2*input[,8]+input[,7])/2/(input[,6]+input[,7]+input[,8]))
#Needs filtering. I have MAF calculated and it comes with INFO score
#Just INFO>0.4 no MAF filter
#No indels
baseline=input[,4]
effect=input[,5]
#B is effect allele
EAF=(2*as.numeric(input[,8])+as.numeric(input[,7]))/2/(as.numeric(input[,6])+as.numeric(input[,7])+as.numeric(input[,8]))
#no indels
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(input[,2], ":", input[,3], "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(input[which(flip==1),2], ":", input[which(flip==1),3], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
output=cbind(SNPID, input[,2],input[,3], baseline, effect, input[,c(12:15)], EAF)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "INFO", "EAF")
output2=output[which(output[,9]>0.4&as.numeric(output[,10])<=0.95&as.numeric(output[,10])>=0.05),]
#Apply 0.4 INFO filter and 0.05 MAF filter
#stitch on number of cases, 335 for all SNPs
Cases=rep(335, length(output2[,1]))
Effective_Cases=as.numeric(output2[,9])*Cases
N=rep(1485, length(output2[,1]))
output2=cbind(output2, Cases, Effective_Cases, N)
#output2[as.numeric(output2[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(output2[,6]))
#summary(as.numeric(output2[,7]))
#summary(as.numeric(output2[,8]))
write.table(output2[,c(1:8,10,9,11,12)], paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/Greek_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
#########################
#HUNT - SAIGE
#########################
#We're happy with HUNT as well
input=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/CAD_HUNT.allchr.SAIGE.sorted.MAC.R2.MAF.hwe.dat", colClasses=c("character"))
input=as.data.frame(input)
colnames(input)=c("SNPID","CHR","POS","Allele0","Allele1","AC","AF","N","BETA","SE","Tstat","p.value","p.value.NA","Is.SPA.converge","varT","varTstar","MAC","R2","MAF","HWE_PVAL")
INFO=as.numeric(input[,18])
#Prefiltered to 0.3 INFO
#14614 have no INFO, presumably because they were genotyped like last time. Set their INFO to 1
INFO[which(is.na(INFO)==T)]=1
CHR=as.numeric(input[,2])
POS=input[,3]
baseline=toupper(input[,4])
effect=toupper(input[,5])
#ALLELE1 is taken to be the effect allele confirmed. AC and AF are all for Allele1.
#There are indels
#All indels are good, single variant deletion and no "-" so can process together
EAF=as.numeric(input[,7])
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
output=cbind(SNPID, CHR,POS, baseline, effect, input[,c(9,10,12)], EAF, INFO)
#output=output[order(as.numeric(output[,2]), as.numeric(output[,3]), output[,1]),]
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "INFO")
#Stitch on cases, 7710
Cases=rep(7710, length(output[,1]))
Effective_Cases=as.numeric(output[,10])*Cases
N=input[,8]
output2=cbind(output, Cases, Effective_Cases, N)
#INFO is prefiltered to 0.3
#We filter MAF to 0.0005
output2=output2[which(as.numeric(output2[,9])<=0.9995&as.numeric(output2[,9])>=0.0005),]
#Lose 8M variants, 16143079 left
#output2[as.numeric(output2[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(output2[,6]))
#summary(as.numeric(output2[,7]))
#summary(as.numeric(output2[,8]))
output2=output2[order(as.numeric(output2[,2]), as.numeric(output2[,3]), output2[,1]),]
write.table(output2, paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/HUNT_SAIGE_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
#########################
#PARTNERS
#########################
#Process Partners
input=fread("/scratch/asb38/ukbiobank/CAD_meta_analysis_study_results/PHB_CAD_INTERMEDIATE_relevant_chips_ancestries_infoMAFcases_filtered_N_cases_N_total_meta_logORs_with_AVGINFO_0102191.tbl", colClasses=c("character"))
input=as.data.frame(input)
temp=strsplit(input[,1], "[:]")
temp2=do.call(rbind, temp)
CHR=temp2[,1]
POS=temp2[,2]
baseline=toupper(input[,3])
effect=toupper(input[,2])
#No indels
#Allele1 is effect allele
EAF=as.numeric(input[,4])
INFO=as.numeric(input[,20])
Cases=as.numeric(input[,17])
Effective_Cases=INFO*Cases
N=as.numeric(input[,18])
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
output=cbind(SNPID, CHR,POS, baseline, effect, input[,c(8:10)], EAF, INFO, Cases, Effective_Cases, N)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "INFO", "Cases", "Effective_Cases", "N")
output=output[order(as.numeric(output[,2]), as.numeric(output[,3])),]
#output[as.numeric(output[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(output[,6]))
#summary(as.numeric(output[,7]))
#summary(as.numeric(output[,8]))
write.table(output, paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/PARTNERS_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
################
#TIMI
################
#Process TIMI
input=fread("/scratch/asb38/TIMI/TIMI.EUR.CAD.age_sex_hxaf_hxt2d_batch_pc1_5.chr1_22.CR.FILT.20181102.txt", colClasses=c("character"))
input=as.data.frame(input)
#filter out non autosomals (bad chrs) and ones where chrs dont match
input=input[which(input[,22]%in%c(1:22)&input[,22]==input[,2]),]
#lost 35556
CHR=input[,22]
POS=input[,23]
baseline=toupper(input[,4])
effect=toupper(input[,6])
#We have indels but they have front allele. I can just proceed as expected.
#summary(pmin(nchar(baseline), nchar(effect)))
EAF=as.numeric(input[,25])
EAF[which(effect!=input[,26])]=1-as.numeric(input[which(effect!=input[,26]),25])
INFO=as.numeric(input[,17])
Cases=as.numeric(input[,27])
Effective_Cases=INFO*Cases
N=as.numeric(input[,8])
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
output=cbind(SNPID, CHR,POS, baseline, effect, input[,c(24, 10, 12)], EAF, INFO, Cases, Effective_Cases, N)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "INFO", "Cases", "Effective_Cases", "N")
output=output[order(as.numeric(output[,2]), as.numeric(output[,3])),]
#drop all duplicate variants
output=output[-which(output[,1]%in%output[duplicated(output[,1]),1]),]
#3204 dupes, mostly indels but some are different positions in build 38 but same in hg19, e.g.chr1:121605507:C:T and chr1:121853089:C:T
#output[as.numeric(output[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(output[,6]))
#summary(as.numeric(output[,7]))
#summary(as.numeric(output[,8]))
write.table(output, paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/TIMI_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
#########################
#HUNT - OLD
#########################
#We're happy with HUNT as well
input=fread("/scratch/asb38/ukbiobank/HUNT/CAD.intermediate.all.imputed.results_Cook.txt", colClasses=c("character"))
input=as.data.frame(input)
input=input[-which(input[,2]==23),]
#drop chr23, no other dataset has it
#info mapping file
info=fread("/scratch/asb38/ukbiobank/HUNT/CAD.intermediate.all.imputed.results.info.bed", colClasses=c("character"))
info=as.data.frame(info)
#same number of rows after removal of chr23 from input
INFO=as.numeric(info[,15])
CHR=info[,2]
POS=info[,3]
baseline=toupper(info[,5])
effect=toupper(info[,6])
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
temp=cbind(SNPID, INFO)
#Problem, many variants don't have INFO score. Many are very rare but some are quite common
#Those are genotyped hence have info score of 1
temp[which(is.na(INFO)==T),2]=1
temp[,2]=as.numeric(temp[,2])
CHR=input[,2]
POS=input[,3]
baseline=toupper(input[,5])
effect=toupper(input[,6])
#ALLELE0 is taken to be the effect allele contrary to BOLT output manual. Adam gave the go ahead.
#This has been confirmed by the people who made the dataset
#There are indels
#All indels are good, single variant deletion and no "-" so can process together
EAF=as.numeric(input[,7])
#Actually EAF is just A1_freq despite the fact we want A0 Freq as A0 is our effect allele. Turns out, the freq column was also mislabelled.
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
output=cbind(SNPID, CHR,POS, baseline, effect, input[,c(16,17,14)], EAF)
#output=output[order(as.numeric(output[,2]), as.numeric(output[,3]), output[,1]),]
output2=merge(output, temp, by="SNPID")
colnames(output2)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "INFO")
#Stitch on cases, 7710
Cases=rep(7710, length(output2[,1]))
Effective_Cases=as.numeric(output2[,10])*Cases
output2=cbind(output2, Cases, Effective_Cases)
#INFO is prefiltered to 0.3
#We filter MAF to 0.0005
output2=output2[which(as.numeric(output2[,9])<=0.9995&as.numeric(output2[,9])>=0.0005),]
#Lose 8M variants, 16143117 left
#output2[as.numeric(output2[,3])%%1000000==0,]
#Check for NA beta/se/pval
#summary(as.numeric(output2[,6]))
#summary(as.numeric(output2[,7]))
#summary(as.numeric(output2[,8]))
output2=output2[order(as.numeric(output2[,2]), as.numeric(output2[,3]), output2[,1]),]
write.table(output2, paste0("/scratch/tj241/demo/Test/UKBB500K/CHD/HUNT_premeta.tsv"), quote=F, row.names=F, col.names=T, sep="\t")
#########################
#Biobank Japan
#########################
#Process Biobank Japan
input=fread("zcat /rds/project/jmmh2/rds-jmmh2-projects/coronary_genetics/million_hearts_project/Temp/bbj/CAD.auto.rsq07.mac10.txt.gz", colClasses=c("character"))
input=as.data.frame(input)
CHR=input[,1]
POS=input[,2]
baseline=toupper(input[,4])
effect=toupper(input[,5])
#Indels present but coded correctly, e.g. A/AAT
#Allele1 is effect allele
EAF=as.numeric(input[,7])
INFO=as.numeric(input[,19])
Cases=rep(29319 , length(input[,19]))
Effective_Cases=INFO*Cases
N=as.numeric(input[,8])
flip=rep(0, length(baseline))
flip[which(effect<baseline)]=1
SNPID=paste0(CHR, ":", POS, "_", baseline, "_", effect)
SNPID[which(flip==1)]=paste0(CHR[which(flip==1)], ":", POS[which(flip==1)], "_", effect[which(flip==1)], "_", baseline[which(flip==1)])
output=cbind(SNPID, CHR,POS, baseline, effect, input[,c(9,10,12)], EAF, INFO, Cases, Effective_Cases, N)
colnames(output)=c("ALT_ID", "CHR", "POS", "BASELINE", "EFFECT", "BETA", "SE", "PVAL", "EAF", "INFO", "Cases", "Effective_Cases", "N")