-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathanal.shtml
1724 lines (1510 loc) · 60.9 KB
/
anal.shtml
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
<html>
<body>
<head>
<link rel="stylesheet" href="plink.css" type="text/css">
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=utf-8">
<title>PLINK: Whole genome data analysis toolset</title>
</head>
<!--<html>-->
<!--<title>PLINK</title>-->
<!--<body>-->
<font size="6" color="darkgreen"><b>plink...</b></font>
<div style="position:absolute;right:10px;top:10px;font-size:
75%"><em>Last original <tt>PLINK</tt> release is <b>v1.07</b>
(10-Oct-2009); <b>PLINK 1.9</b> is now <a href="plink2.shtml"> available</a> for beta-testing</em></div>
<h1>Whole genome association analysis toolset</h1>
<font size="1" color="darkgreen">
<em>
<a href="index.shtml">Introduction</a> |
<a href="contact.shtml">Basics</a> |
<a href="download.shtml">Download</a> |
<a href="reference.shtml">Reference</a> |
<a href="data.shtml">Formats</a> |
<a href="dataman.shtml">Data management</a> |
<a href="summary.shtml">Summary stats</a> |
<a href="thresh.shtml">Filters</a> |
<a href="strat.shtml">Stratification</a> |
<a href="ibdibs.shtml">IBS/IBD</a> |
<a href="anal.shtml">Association</a> |
<a href="fanal.shtml">Family-based</a> |
<a href="perm.shtml">Permutation</a> |
<a href="ld.shtml">LD calcualtions</a> |
<a href="haplo.shtml">Haplotypes</a> |
<a href="whap.shtml">Conditional tests</a> |
<a href="proxy.shtml">Proxy association</a> |
<a href="pimputation.shtml">Imputation</a> |
<a href="dosage.shtml">Dosage data</a> |
<a href="metaanal.shtml">Meta-analysis</a> |
<a href="annot.shtml">Result annotation</a> |
<a href="clump.shtml">Clumping</a> |
<a href="grep.shtml">Gene Report</a> |
<a href="epi.shtml">Epistasis</a> |
<a href="cnv.shtml">Rare CNVs</a> |
<a href="gvar.shtml">Common CNPs</a> |
<a href="rfunc.shtml">R-plugins</a> |
<a href="psnp.shtml">SNP annotation</a> |
<a href="simulate.shtml">Simulation</a> |
<a href="profile.shtml">Profiles</a> |
<a href="ids.shtml">ID helper</a> |
<a href="res.shtml">Resources</a> |
<a href="flow.shtml">Flow chart</a> |
<a href="misc.shtml">Misc.</a> |
<a href="faq.shtml">FAQ</a> |
<a href="gplink.shtml">gPLINK</a>
</em></font>
</p>
<table border=0>
<tr>
<td bgcolor="lightblue" valign="top" width=20%>
<font size="1">
<a href="index.shtml">1. Introduction</a> </p>
<a href="contact.shtml">2. Basic information</a> </p>
<ul>
<li> <a href="contact.shtml#cite">Citing PLINK</a>
<li> <a href="contact.shtml#probs">Reporting problems</a>
<li> <a href="news.shtml">What's new?</a>
<li> <a href="pdf.shtml">PDF documentation</a>
</ul>
<a href="download.shtml">3. Download and general notes</a> </p>
<ul>
<li> <a href="download.shtml#download">Stable download</a>
<li> <a href="download.shtml#latest">Development code</a>
<li> <a href="download.shtml#general">General notes</a>
<li> <a href="download.shtml#msdos">MS-DOS notes</a>
<li> <a href="download.shtml#nix">Unix/Linux notes</a>
<li> <a href="download.shtml#compilation">Compilation</a>
<li> <a href="download.shtml#input">Using the command line</a>
<li> <a href="download.shtml#output">Viewing output files</a>
<li> <a href="changelog.shtml">Version history</a>
</ul>
<a href="reference.shtml">4. Command reference table</a> </p>
<ul>
<li> <a href="reference.shtml#options">List of options</a>
<li> <a href="reference.shtml#output">List of output files</a>
<li> <a href="newfeat.shtml">Under development</a>
</ul>
<a href="data.shtml">5. Basic usage/data formats</a>
<ul>
<li> <a href="data.shtml#plink">Running PLINK</a>
<li> <a href="data.shtml#ped">PED files</a>
<li> <a href="data.shtml#map">MAP files</a>
<li> <a href="data.shtml#tr">Transposed filesets</a>
<li> <a href="data.shtml#long">Long-format filesets</a>
<li> <a href="data.shtml#bed">Binary PED files</a>
<li> <a href="data.shtml#pheno">Alternate phenotypes</a>
<li> <a href="data.shtml#covar">Covariate files</a>
<li> <a href="data.shtml#clst">Cluster files</a>
<li> <a href="data.shtml#sets">Set files</a>
</ul>
<a href="dataman.shtml">6. Data management</a> </p>
<ul>
<li> <a href="dataman.shtml#recode">Recode</a>
<li> <a href="dataman.shtml#recode">Reorder</a>
<li> <a href="dataman.shtml#snplist">Write SNP list</a>
<li> <a href="dataman.shtml#updatemap">Update SNP map</a>
<li> <a href="dataman.shtml#updateallele">Update allele information</a>
<li> <a href="dataman.shtml#refallele">Force reference allele</a>
<li> <a href="dataman.shtml#updatefam">Update individuals</a>
<li> <a href="dataman.shtml#wrtcov">Write covariate files</a>
<li> <a href="dataman.shtml#wrtclst">Write cluster files</a>
<li> <a href="dataman.shtml#flip">Flip strand</a>
<li> <a href="dataman.shtml#flipscan">Scan for strand problem</a>
<li> <a href="dataman.shtml#merge">Merge two files</a>
<li> <a href="dataman.shtml#mergelist">Merge multiple files</a>
<li> <a href="dataman.shtml#extract">Extract SNPs</a>
<li> <a href="dataman.shtml#exclude">Remove SNPs</a>
<li> <a href="dataman.shtml#zero">Zero out sets of genotypes</a>
<li> <a href="dataman.shtml#keep">Extract Individuals</a>
<li> <a href="dataman.shtml#remove">Remove Individuals</a>
<li> <a href="dataman.shtml#filter">Filter Individuals</a>
<li> <a href="dataman.shtml#attrib">Attribute filters</a>
<li> <a href="dataman.shtml#makeset">Create a set file</a>
<li> <a href="dataman.shtml#tabset">Tabulate SNPs by sets</a>
<li> <a href="dataman.shtml#snp-qual">SNP quality scores</a>
<li> <a href="dataman.shtml#geno-qual">Genotypic quality scores</a>
</ul>
<a href="summary.shtml">7. Summary stats</a>
<ul>
<li> <a href="summary.shtml#missing">Missingness</a>
<li> <a href="summary.shtml#oblig_missing">Obligatory missingness</a>
<li> <a href="summary.shtml#clustermissing">IBM clustering</a>
<li> <a href="summary.shtml#testmiss">Missingness by phenotype</a>
<li> <a href="summary.shtml#mishap">Missingness by genotype</a>
<li> <a href="summary.shtml#hardy">Hardy-Weinberg</a>
<li> <a href="summary.shtml#freq">Allele frequencies</a>
<li> <a href="summary.shtml#prune">LD-based SNP pruning</a>
<li> <a href="summary.shtml#mendel">Mendel errors</a>
<li> <a href="summary.shtml#sexcheck">Sex check</a>
<li> <a href="summary.shtml#pederr">Pedigree errors</a>
</ul>
<a href="thresh.shtml">8. Inclusion thresholds</a>
<ul>
<li> <a href="thresh.shtml#miss2">Missing/person</a>
<li> <a href="thresh.shtml#maf">Allele frequency</a>
<li> <a href="thresh.shtml#miss1">Missing/SNP</a>
<li> <a href="thresh.shtml#hwd">Hardy-Weinberg</a>
<li> <a href="thresh.shtml#mendel">Mendel errors</a>
</ul>
<a href="strat.shtml">9. Population stratification</a>
<ul>
<li> <a href="strat.shtml#cluster">IBS clustering</a>
<li> <a href="strat.shtml#permtest">Permutation test</a>
<li> <a href="strat.shtml#options">Clustering options</a>
<li> <a href="strat.shtml#matrix">IBS matrix</a>
<li> <a href="strat.shtml#mds">Multidimensional scaling</a>
<li> <a href="strat.shtml#outlier">Outlier detection</a>
</ul>
<a href="ibdibs.shtml">10. IBS/IBD estimation</a>
<ul>
<li> <a href="ibdibs.shtml#genome">Pairwise IBD</a>
<li> <a href="ibdibs.shtml#inbreeding">Inbreeding</a>
<li> <a href="ibdibs.shtml#homo">Runs of homozygosity</a>
<li> <a href="ibdibs.shtml#segments">Shared segments</a>
</ul>
<a href="anal.shtml">11. Association</a>
<ul>
<li> <a href="anal.shtml#cc">Case/control</a>
<li> <a href="anal.shtml#fisher">Fisher's exact</a>
<li> <a href="anal.shtml#model">Full model</a>
<li> <a href="anal.shtml#strat">Stratified analysis</a>
<li> <a href="anal.shtml#homog">Tests of heterogeneity</a>
<li> <a href="anal.shtml#hotel">Hotelling's T(2) test</a>
<li> <a href="anal.shtml#qt">Quantitative trait</a>
<li> <a href="anal.shtml#qtmeans">Quantitative trait means</a>
<li> <a href="anal.shtml#qtgxe">Quantitative trait GxE</a>
<li> <a href="anal.shtml#glm">Linear and logistic models</a>
<li> <a href="anal.shtml#set">Set-based tests</a>
<li> <a href="anal.shtml#adjust">Multiple-test correction</a>
</ul>
<a href="fanal.shtml">12. Family-based association</a>
<ul>
<li> <a href="fanal.shtml#tdt">TDT</a>
<li> <a href="fanal.shtml#ptdt">ParenTDT</a>
<li> <a href="fanal.shtml#poo">Parent-of-origin</a>
<li> <a href="fanal.shtml#dfam">DFAM test</a>
<li> <a href="fanal.shtml#qfam">QFAM test</a>
</ul>
<a href="perm.shtml">13. Permutation procedures</a>
<ul>
<li> <a href="perm.shtml#perm">Basic permutation</a>
<li> <a href="perm.shtml#aperm">Adaptive permutation</a>
<li> <a href="perm.shtml#mperm">max(T) permutation</a>
<li> <a href="perm.shtml#rank">Ranked permutation</a>
<li> <a href="perm.shtml#genedropmodel">Gene-dropping</a>
<li> <a href="perm.shtml#cluster">Within-cluster</a>
<li> <a href="perm.shtml#mkphe">Permuted phenotypes files</a>
</ul>
<a href="ld.shtml">14. LD calculations</a>
<ul>
<li> <a href="ld.shtml#ld1">2 SNP pairwise LD</a>
<li> <a href="ld.shtml#ld2">N SNP pairwise LD</a>
<li> <a href="ld.shtml#tags">Tagging options</a>
<li> <a href="ld.shtml#blox">Haplotype blocks</a>
</ul>
<a href="haplo.shtml">15. Multimarker tests</a>
<ul>
<li> <a href="haplo.shtml#hap1">Imputing haplotypes</a>
<li> <a href="haplo.shtml#precomputed">Precomputed lists</a>
<li> <a href="haplo.shtml#hap2">Haplotype frequencies</a>
<li> <a href="haplo.shtml#hap3">Haplotype-based association</a>
<li> <a href="haplo.shtml#hap3c">Haplotype-based GLM tests</a>
<li> <a href="haplo.shtml#hap3b">Haplotype-based TDT</a>
<li> <a href="haplo.shtml#hap4">Haplotype imputation</a>
<li> <a href="haplo.shtml#hap5">Individual phases</a>
</ul>
<a href="whap.shtml">16. Conditional haplotype tests</a>
<ul>
<li> <a href="whap.shtml#whap1">Basic usage</a>
<li> <a href="whap.shtml#whap2">Specifying type of test</a>
<li> <a href="whap.shtml#whap3">General haplogrouping</a>
<li> <a href="whap.shtml#whap4">Covariates and other SNPs</a>
</ul>
<a href="proxy.shtml">17. Proxy association</a>
<ul>
<li> <a href="proxy.shtml#proxy1">Basic usage</a>
<li> <a href="proxy.shtml#proxy2">Refining a signal</a>
<li> <a href="proxy.shtml#proxy2b">Multiple reference SNPs</a>
<li> <a href="proxy.shtml#proxy3">Haplotype-based SNP tests</a>
</ul>
<a href="pimputation.shtml">18. Imputation (beta)</a>
<ul>
<li> <a href="pimputation.shtml#impute1">Making reference set</a>
<li> <a href="pimputation.shtml#impute2">Basic association test</a>
<li> <a href="pimputation.shtml#impute3">Modifying parameters</a>
<li> <a href="pimputation.shtml#impute4">Imputing discrete calls</a>
<li> <a href="pimputation.shtml#impute5">Verbose output options</a>
</ul>
<a href="dosage.shtml">19. Dosage data</a>
<ul>
<li> <a href="dosage.shtml#format">Input file formats</a>
<li> <a href="dosage.shtml#assoc">Association analysis</a>
<li> <a href="dosage.shtml#output">Outputting dosage data</a>
</ul>
<a href="metaanal.shtml">20. Meta-analysis</a>
<ul>
<li> <a href="metaanal.shtml#basic">Basic usage</a>
<li> <a href="metaanal.shtml#opt">Misc. options</a>
</ul>
<a href="annot.shtml">21. Annotation</a>
<ul>
<li> <a href="annot.shtml#basic">Basic usage</a>
<li> <a href="annot.shtml#opt">Misc. options</a>
</ul>
<a href="clump.shtml">22. LD-based results clumping</a>
<ul>
<li> <a href="clump.shtml#clump1">Basic usage</a>
<li> <a href="clump.shtml#clump2">Verbose reporting</a>
<li> <a href="clump.shtml#clump3">Combining multiple studies</a>
<li> <a href="clump.shtml#clump4">Best single proxy</a>
</ul>
<a href="grep.shtml">23. Gene-based report</a>
<ul>
<li> <a href="grep.shtml#grep1">Basic usage</a>
<li> <a href="grep.shtml#grep2">Other options</a>
</ul>
<a href="epi.shtml">24. Epistasis</a>
<ul>
<li> <a href="epi.shtml#snp">SNP x SNP</a>
<li> <a href="epi.shtml#case">Case-only</a>
<li> <a href="epi.shtml#gene">Gene-based</a>
</ul>
<a href="cnv.shtml">25. Rare CNVs</a>
<ul>
<li> <a href="cnv.shtml#format">File format</a>
<li> <a href="cnv.shtml#maps">MAP file construction</a>
<li> <a href="cnv.shtml#loading">Loading CNVs</a>
<li> <a href="cnv.shtml#olap_check">Check for overlap</a>
<li> <a href="cnv.shtml#type_filter">Filter on type </a>
<li> <a href="cnv.shtml#gene_filter">Filter on genes </a>
<li> <a href="cnv.shtml#freq_filter">Filter on frequency </a>
<li> <a href="cnv.shtml#burden">Burden analysis</a>
<li> <a href="cnv.shtml#burden2">Geneset enrichment</a>
<li> <a href="cnv.shtml#assoc">Mapping loci</a>
<li> <a href="cnv.shtml#reg-assoc">Regional tests</a>
<li> <a href="cnv.shtml#qt-assoc">Quantitative traits</a>
<li> <a href="cnv.shtml#write_cnvlist">Write CNV lists</a>
<li> <a href="cnv.shtml#report">Write gene lists</a>
<li> <a href="cnv.shtml#groups">Grouping CNVs </a>
</ul>
<a href="gvar.shtml">26. Common CNPs</a>
<ul>
<li> <a href="gvar.shtml#cnv2"> CNPs/generic variants</a>
<li> <a href="gvar.shtml#cnv2b"> CNP/SNP association</a>
</ul>
<a href="rfunc.shtml">27. R-plugins</a>
<ul>
<li> <a href="rfunc.shtml#rfunc1">Basic usage</a>
<li> <a href="rfunc.shtml#rfunc2">Defining the R function</a>
<li> <a href="rfunc.shtml#rfunc2b">Example of debugging</a>
<li> <a href="rfunc.shtml#rfunc3">Installing Rserve</a>
</ul>
<a href="psnp.shtml">28. Annotation web-lookup</a>
<ul>
<li> <a href="psnp.shtml#psnp1">Basic SNP annotation</a>
<li> <a href="psnp.shtml#psnp2">Gene-based SNP lookup</a>
<li> <a href="psnp.shtml#psnp3">Annotation sources</a>
</ul>
<a href="simulate.shtml">29. Simulation tools</a>
<ul>
<li> <a href="simulate.shtml#sim1">Basic usage</a>
<li> <a href="simulate.shtml#sim2">Resampling a population</a>
<li> <a href="simulate.shtml#sim3">Quantitative traits</a>
</ul>
<a href="profile.shtml">30. Profile scoring</a>
<ul>
<li> <a href="profile.shtml#prof1">Basic usage</a>
<li> <a href="profile.shtml#prof2">SNP subsets</a>
<li> <a href="profile.shtml#dose">Dosage data</a>
<li> <a href="profile.shtml#prof3">Misc options</a>
</ul>
<a href="ids.shtml">31. ID helper</a>
<ul>
<li> <a href="ids.shtml#ex">Overview/example</a>
<li> <a href="ids.shtml#intro">Basic usage</a>
<li> <a href="ids.shtml#check">Consistency checks</a>
<li> <a href="ids.shtml#alias">Aliases</a>
<li> <a href="ids.shtml#joint">Joint IDs</a>
<li> <a href="ids.shtml#lookup">Lookups</a>
<li> <a href="ids.shtml#replace">Replace values</a>
<li> <a href="ids.shtml#match">Match files</a>
<li> <a href="ids.shtml#qmatch">Quick match files</a>
<li> <a href="ids.shtml#misc">Misc.</a>
</ul>
<a href="res.shtml">32. Resources</a>
<ul>
<li> <a href="res.shtml#hapmap">HapMap (PLINK format)</a>
<li> <a href="res.shtml#teach">Teaching materials</a>
<li> <a href="res.shtml#mmtests">Multimarker tests</a>
<li> <a href="res.shtml#sets">Gene-set lists</a>
<li> <a href="res.shtml#glist">Gene range lists</a>
<li> <a href="res.shtml#attrib">SNP attributes</a>
</ul>
<a href="flow.shtml">33. Flow-chart</a>
<ul>
<li> <a href="flow.shtml">Order of commands</a>
</ul>
<a href="misc.shtml">34. Miscellaneous</a>
<ul>
<li> <a href="misc.shtml#opt">Command options/modifiers</a>
<li> <a href="misc.shtml#output">Association output modifiers</a>
<li> <a href="misc.shtml#species">Different species</a>
<li> <a href="misc.shtml#bugs">Known issues</a>
</ul>
<a href="faq.shtml">35. FAQ & Hints</a>
</p>
<a href="gplink.shtml">36. gPLINK</a>
<ul>
<li> <a href="gplink.shtml">gPLINK mainpage</a>
<li> <a href="gplink_tutorial/index.html">Tour of gPLINK</a>
<li> <a href="gplink.shtml#overview">Overview: using gPLINK</a>
<li> <a href="gplink.shtml#locrem">Local versus remote modes</a>
<li> <a href="gplink.shtml#start">Starting a new project</a>
<li> <a href="gplink.shtml#config">Configuring gPLINK</a>
<li> <a href="gplink.shtml#plink">Initiating PLINK jobs</a>
<li> <a href="gplink.shtml#view">Viewing PLINK output</a>
<li> <a href="gplink.shtml#hv">Integration with Haploview</a>
<li> <a href="gplink.shtml#down">Downloading gPLINK</a></p>
</ul>
</font>
</td><td width=5%>
<td valign="top">
</p>
<h1>Association analysis</h1>
</p>
The basic association test is for a disease trait and is based on
comparing allele frequencies between cases and controls (asymptotic
and empirical p-values are available). Also implemented are the
Cochran-Armitage trend test, Fisher's exact test, different genetic models
(dominant, recessive and general), tests for stratified samples (e.g.
Cochran-Mantel-Haenszel, Breslow-Day tests), a test for a quantitative
trait; a test for differences in missing genotype rate between cases and
controls; multilocus tests, using either Hotelling's T(2) statistic or
a sum-statistic approach (evaluated by permutation) as well as <a
href="haplo.shtml">haplotype tests</a>. The basic
tests can be performed with permutation, described in the
<a href="perm.shtml">following section</a> to provide empirical
p-values, and allow for different designs (e.g. by use of structured,
within-cluster permutation). Family-based tests are described in
the <a href="fanal.shtml">next section</a>
</p>
<strong>HINT</strong> The basic association commands (<tt>--assoc</tt>,
<tt>--model</tt>, <tt>--fisher</tt>, <tt>--linear</tt> and <tt>--logistic</tt>) will
test only a single phenotype. If your alternate phenotype file contains more than one
phenotype, then adding the <tt>--all-pheno</tt> flag will make PLINK cycle over each
phenotype, e.g. instead of a single <tt>plink.assoc</tt> output file, if there are
100 phenotypes, PLINK will now show
<pre>
plink.P1.assoc
plink.P2.assoc
...
plink.P100.assoc
</pre>
Naturally, it will take 100 times longer... If you are testing a very large number
of phenotypes, it might be worth specifying <tt>--pfilter</tt> also, to reduce the
amount of amount (e.g. only outputing tests significant at p=1e-4 if <tt>--pfilter
1e-4</tt> is specified).
<a name="cc">
<h2>Basic case/control association test</h2>
</a>
</p>
To perform a standard case/control association analysis, use the option:
<h5>
plink --file mydata --assoc
</h5></p>
which generates a file
<pre>
plink.assoc
</pre>
which contains the fields:
<pre>
CHR Chromosome
SNP SNP ID
BP Physical position (base-pair)
A1 Minor allele name (based on whole sample)
F_A Frequency of this allele in cases
F_U Frequency of this allele in controls
A2 Major allele name
CHISQ Basic allelic test chi-square (1df)
P Asymptotic p-value for this test
OR Estimated odds ratio (for A1, i.e. A2 is reference)
</pre>
</P><strong>Hint</strong> In addition, if the optional command
<tt>--ci</tt> <em>X</em> (where <em>X</em> is the desired coverage for a
confidence interval, e.g. 0.95 or 0.99) is included, then two extra
fields are appended to this output:
<pre>
L95 Lower bound of 95% confidence interval for odds ratio
U95 Upper bound of 95% confidence interval for odds ratio
</pre>
(where 95 would change if a different value was used with the
<tt>--ci</tt> option, naturally).
</p>
Adding the option
<pre>
--counts
</pre>
with <tt>--assoc</tt> will make PLINK report allele counts, rather than frequencies, in cases and controls.
</p>
See the <a href="perm.shtml">next section on permutation</a> to learn how
to generate empirical p-values and use other aspects of permutation-based
testing.</p>
See the <a href="haplo.shtml">section on multimarker tests</a> to learn how to perform haplotype-based
tests of association.
</p>
This analysis should appropriately handle X/Y chromosome SNPs automatically.
</p>
<a name="fisher">
<h2>Fisher's Exact test (allelic association) </h2>
</a>
</p>
To perform a standard case/control association analysis using
Fisher's exact test to generate significance, use the option:
<h5>
plink --file mydata --fisher
</h5></p>
which generates a file
<pre>
plink.fisher
</pre>
which contains the fields:
<pre>
CHR Chromosome
SNP SNP ID
BP Physical position (base-pair)
A1 Minor allele name (based on whole sample)
F_A Frequency of this allele in cases
F_U Frequency of this allele in controls
A2 Major allele name
P Exact p-value for this test
OR Estimated odds ratio (for A1)
</pre>
As described below, if <tt>--fisher</tt> is specified with <tt>--model</tt> as well,
<tt>PLINK</tt> will perform genotypic tests using Fisher's exact test.
</P><strong>Note</strong> You can also use permutation to generate exact,
empirical significance values that would also be valid in small samples,
etc. </p>
<a name="model">
<h2>Alternate / full model association tests</h2>
</a></p>
It is possible to perform tests of association between a disease and a variant other than the
basic allelic test (which compares frequencies of alleles in cases versus controls), by using
the <tt>--model</tt> option. The tests offered here are (in addition to the basic allelic test):
<ul>
<li> Cochran-Armitage trend test
<li> Genotypic (2 df) test
<li> Dominant gene action (1df) test
<li> Recessive gene action (1df) test
</ul>
One advantage of the Cochran-Armitage test is that it does not assume Hardy-Weinberg equilibrium,
as the individual, not the allele, is the unit of analysis (although the permutation-based empirical
p-values from the basic allelic test also have this property). It is important to remember that SNPs
showing severe deviations from Hardy-Weinberg are often likely to be bad
SNPs, or reflect stratification in the sample, however, and so are probably
best excluded in many cases.
</p>
The genotypic test provides a general test of association in the 2-by-3 table of disease-by-genotype. The
dominant and recessive models are tests for the minor allele (which is the
minor allele can be found in the output of either the <tt>--assoc</tt>
or the <tt>--freq</tt> commands. That is, if <tt>D</tt> is the minor
allele (and <tt>d</tt> is the major allele):
<pre>
Allelic: D versus d
Dominant: (DD, Dd) versus dd
Recessive: DD versus (Dd, dd)
Genotypic: DD versus Dd versus dd
</pre>
As mentioned above, these tests are generated with option:
<h5>
plink --file mydata --model
</h5></p>
which generates a file
<pre>
plink.model
</pre>
which contains the following fields:
<pre>
CHR Chromosome number
SNP SNP identifier
TEST Type of test
AFF Genotypes/alleles in cases
UNAFF Genotypes/alleles in controls
CHISQ Chi-squated statistic
DF Degrees of freedom for test
P Asymptotic p-value
</pre>
Each SNP will feature on five rows of the output, correspondnig to the
five tests applied. The column <tt>TEST</tt> refers to
either <tt>ALLELIC</tt>, <tt>TREND</tt>, <tt>GENO</tt>,
<tt>DOM</tt> or <tt>REC</tt>, refering to the different types of test
mentioned above. The genotypic or allelic counts are given for cases
and controls separately. For recessive and dominant tests, the counts
represent the genotypes, with two of the classes pooled.
</p>
These tests only consider diploid genotypes: that is, for the X
chromosome males will be excluded even from the ALLELIC test. This way
the same data are used for the five tests presented here. Note that,
in contrast, the basic association commands (<tt>--assoc</tt>
and <tt>--linear</tt>, etc) include single male X chromosomes, and so the
results may differ.
</p>
The genotypic and dominant/recessive tests will only be conducted if
there is a minimum number of observations per cell in the 2-by-3
table: by default, if at least one of the cells has a frequency less
than 5, then we skip the alternate tests (<tt>NA</tt> is written in
the results file). The Cochran-Armitage and allelic tests are
performed in all cases. This threshold can be altered with the
<tt>--cell</tt> option:
<h5>
plink --file mydata --model --cell 20
</h5></p>
</p>
If permutation (with the <tt>--mperm</tt> or <tt>--perm</tt> options)
is specified, the <tt>-model</tt> option will by default perform a
permutation test based on the most significant result
of <tt>ALLELIC</tt>, <tt>DOM</tt> and <tt>REC</tt> models. That is,
for each SNP, the best original result will be compared against the
best of these three tests for that SNP for every replicate. In max(T)
permutation mode, this will also be compared against the best result
from all SNPs for the <tt>EMP2</tt> field. This procedure controls for
the fact that we have selected the best out of three correlated tests
for each SNP. The output will be generated in the file
<pre>
plink.model.best.perm
</pre>
or
<pre>
plink.model.best.mperm
</pre>
depending on whether adaptive or max(T) permutation was used.
</p>
The behavior of the <tt>--model</tt> command can be changed by adding
the <tt>--model-gen</tt>, <tt>--model-trend</tt>, <tt>--model-dom</tt>
or <tt>--model-rec</tt> flags to make the permutation use the
genotypic, the Cochram-Armitage trend test, the dominant test or the
recessive test as the basis for permutation instead. In this case, one
of the the following files will be generated:
<pre>
plink.model.gen.perm plink.model.gen.mperm
plink.model.trend.perm plink.model.trend.mperm
plink.model.dom.perm plink.model.dom.mperm
plink.model.rec.perm plink.model.rec.mperm
</pre>
It is also possible to add the <tt>--fisher</tt> flag to obtain exact p-values:
<h5>
./plink --bfile mydata --model --fisher
</h5></p> in which case the <tt>CHISQ</tt> field does not appear. Note
that the genotypic, allelic, dominant and recessive models use the
Fisher's exact; the trend-test does not and will give the same p-value as without
the <tt>--fisher</tt> flag. Also, by default, when <tt>--fisher</tt> is added, the
<tt>--cell</tt> field is set to 0, i.e. to include all SNPs.
<a name="strat">
<h2>Stratified analyses</h2>
</a>
</p>
When a cluster variable has been specified, by pointing to a file
that contains this information, with the <tt>--within</tt> command,
it is possible to perform a number of tests of case/control association
that take this clustering into account, or explicitly test for
homogeneity of effect between clusters.
</p>
</p><strong>Note</strong> In many cases, permutation procedures can also
be used to account for clusters in
the data. See the <a href="perm.shtml">next section</a> for more details.
The tests presented below
are only applicable for case/control data, so permutation might be useful
for quantitative trait outcomes,
etc.
</p>
There are two basic classes of test:
<ul>
<li> Testing for overall disease/gene association, controlling for clusters
<li> Testing for heterogeneity of the disease/gene assocation between different clusters
</ul>
The type of cluster structure will vary in terms of how many clusters
there are in the sample, and how many
people belong to each cluster. At one extreme, we might have two only 2
clusters in the sample, each with
a large number of cases and controls. At the other extreme, we might have
a very large number of clusters,
such that each cluster only has 2 individuals. These factors will
influence the choice of stratified
analysis. </p>
The tests offered are:
<ul>
<li> Cochran-Mantel-Haenszel test for 2x2xK stratified tables
<li> Cochran-Mantel-Haenszel test for IxJxK stratified tables
<li> Breslow-Day test of homogeneity of odds ratio
<li> Partitioning the total association chi-square to perform between and within cluster
association, and a test of homogeneity of effect
</ul>
The Cochran-Mantel-Haenszel (CMH) tests are valid with both a large
number of small clusters and a small
number of large clusters. These tests provide a test based on an
"average" odds ratio that controls for the potential confounding due to
the cluster variable.
</p>
The Breslow-Day test asks whether different clusters have different
disease/gene odds ratios: this test
assumes a moderate sample size within each cluster. The partitioning
total association test, which is
conceptually similar to the Breslow-Day test, also makes the same
assumption.
</p>
As mentioned above, the CMH test comes in two flavours: 2x2xK and
IxJxK. Currently, the 2x2xK test represents a <tt> { disease x SNP |
cluster }</tt> test. The generalized form, the IxJxK, represents a
test of <tt> { cluster x SNP | disease }</tt>, i.e. does the SNP vary
between clusters, controlling for any possible true SNP/disease
association. This latter test might be useful in interpreting
significant associations in stratified samples. Typically, the first
form of the test will be of more interest, however. These two tests
are run by using the options:
<h5>
plink --file mydata --mh --within mycluster.dat
</h5></p>
for the basic CMH test, or
<h5>
plink --file mydata --mh2 --within mycluster.dat
</h5></p>
for the IxJxK test.
</p>
The <tt>--mh</tt> option generates the file
<pre>
plink.cmh
</pre>
which contains the fields
<pre>
CHR Chromosome number
SNP SNP identifier
A1 Minor allele code
A2 Major allele code
BP Physical position (base-pair)
CHISQ Cochran-Mantel-Haenszel statistic (1df)
P Asymptotic p-value for CMH test
OR CMH odds ratio
L95 Lower bound on confidence interval for CMH odds ratio
U95 Upper bound on confidence interval for CMH odds ratio
</pre>
The range of the confidence interval with the <tt>--mh</tt> option can be
changed with the <tt>--ci</tt> option:
<h5>
plink --file mydata --mh --within mycluster.dat --ci 0.99
</h5></p>
The <tt>--mh2</tt> option generates the file
<pre>
plink.cmh2
</pre>
which contains the fields:
<pre>
CHR Chromosome
SNP SNP identifier
CHISQ_CMH2 Cochran-Mantel-Haenszel test for IxJxK tables
P_CMH2 Asymptotic p-value for this test
</pre>
It is not possible to obtain confidence intervals or
odds ratios for <tt>--mh2</tt> tests.
</p><strong>Hint</strong> A trick to analyse phenotypes with more two
categories (but only with nominal, not ordinal outcomes) is to use the
<tt>--mh2</tt> option with the
phenotype in the cluster file and the phenotype in the PED file set all
to a single value.
<a name="homog">
<h2>Testing for heterogeneous association</h2>
</a></p>
As mentioned in the previous section, two methods are provided to test
for between-cluster differences in association when using a case/control
design. The Breslow-Day test is specified with the option:
<h5>
plink --file mydata --bd --within myclst.txt
</h5></p>
which runs and generates the same files as the <tt>--mh</tt>
option, described above, but with two extra fields appended:
<pre>
CHISQ_BD Breslow-Day test
P_BD Asymptotic p-value
</pre>
where a significant value indicates between-cluster heterogeneoty
in the odds ratios for the disease/SNP association.
</p>
A similar test of the homogeneity of odds ratio tests based on
partitioning the chi-square statistic is given by:
<h5>
plink --file mydata --homog --within myclst.txt
</h5></p>
which generates the file
<pre>
plink.homog
</pre>
which contains the fields
<pre>
CHR Chromosome number
SNP SNP identifier
A1 Minor allele code
A2 Major allele code
F_A Case allele frequency
F_U Control allele frequency
N_A Case allele count
N_U Control allele count
TEST Type of test
CHISQ Chi-squared association statistic
DF Degrees of freedom
P Asymptotic p-value
OR Odds ratio
</pre>
The <tt>TEST</tt> type is either
<pre>
TOTAL Total SNP & strata association
ASSOC SNP association controlling for strata
HOMOG Between-strata heterogeneity test
X_1 Association in first stratum
X_2 Association in second stratum
...
</pre>
<a name="hotel">
<h2>Hotelling's T(2) multilocus association test</h2>
</a></p>
<strong>IMPORTANT</strong> This command has been temporarily disabled
</p>
For disease-traits, <tt>PLINK</tt> provides support for a
multilocus, genotype-based test using Hotelling's T2 (T-squared)
statistic. The <tt>--set</tt> option should be used to specify
which SNPs are to be grouped, as follows:
<h5>
plink --file data --set mydata.set --T2
</h5></p>
where <tt>mydata.set</tt> defines which SNPs are in which set (see
<a href="data.shtml#sets">this section</a> for more information on
defining sets).
</p>
This command will generate a file
<pre>
plink.T2
</pre>
which contains the fields
<pre>
SET Set name
SIZE Number of SNPs in this set
F F-statistic from Hotelling's test
DF1 Degrees of freedom 1
DF2 Degrees of freedom 2
P_HOTEL Asymptotic p-value
</pre>
</p>
<strong>HINT</strong> Use the <tt>--genedrop</tt> permutation to
perform a family-based application of the Hotelling's T2 test.
This command can be used with all permutation methods (label-swapping
or gene-dropping, adaptive or max(T)). In fact, the permutation test
is based on 1-p in order to make the between set comparisons for the
max(T) statistic more meaningful (as different sized sets would have
F-statistics with different degrees of freedom otherwise). Using
permutation will generate one of the following files:
<pre>
plink.T2.perm
</pre>
which contain the fields
<pre>
SET Set name
SIZE Number of SNPs in this set
EMP1 Empirical p-value
NR Number of permutation replicates
</pre>
or, if <tt>--mperm</tt> was used,
<pre>
plink.T2.mperm
</pre>
which contain the fields
<pre>
SET Set name
SIZE Number of SNPs in this set
EMP1 Empirical p-value
EMP2 max(T) empirical p-value
</pre>
Note that this test uses a simple approach to missing data: rather
than case-wise deletion (removing an individual if they have at least
one missing observation) we impute the mean allelic value. Although
this retains power under most scenarios, it can also cause some bias
when there are lots of missing data points. Using permutation is a
good way around this issue.
<a name="qt">
<h2>Quantitative trait association</h2>
</a>
</p>
Quantitative traits can be tested for association also, using either
asymptotic (likelihood ratio test and Wald test) or empirical
significance values. If the phenotype (column 6 of the PED file or the
phenotype as specified with the <tt>--pheno</tt> option) is
quantitative (i.e. contains values other than 1, 2, 0 or missing)
then <tt>PLINK</tt> will automatically treat the analysis as a
quantitative trait analysis. That is, the same command as for
disease-trait association:
<h5>
plink --file mydata --assoc
</h5></p>
will generate the file
<pre>
plink.qassoc
</pre>
with fields as follows:
<pre>
CHR Chromosome number
SNP SNP identifier
BP Physical position (base-pair)
NMISS Number of non-missing genotypes
BETA Regression coefficient
SE Standard error
R2 Regression r-squared
T Wald test (based on t-distribtion)
P Wald test asymptotic p-value
</pre>
If permutations were also requested, then an extra file, either
<pre>
plink.assoc.perm
</pre>
or
<pre>
plink.assoc.mperm
</pre>
will be generated, depending on whether adaptive or max(T) permutation
was used (see the <a href="perm.shtml">next section</a> for more
details). The empirical p-values are based on the Wald statistic.
</p>
<a name="qtmeans">
<h2>Genotype means for quantitative traits</h2>
</a>
</p>
Adding the flag <tt>--qt-means</tt> along with the <tt>--assoc</tt>
command, when run with a quantitative trait, will produce an