forked from opain/GenoPred
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Ancestry-based_prediction.html
724 lines (609 loc) · 25.4 KB
/
Ancestry-based_prediction.html
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
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8" />
<meta name="generator" content="pandoc" />
<meta http-equiv="X-UA-Compatible" content="IE=EDGE" />
<title>Ancestry-based prediction</title>
<script src="site_libs/jquery-1.11.3/jquery.min.js"></script>
<meta name="viewport" content="width=device-width, initial-scale=1" />
<link href="site_libs/bootstrap-3.3.5/css/united.min.css" rel="stylesheet" />
<script src="site_libs/bootstrap-3.3.5/js/bootstrap.min.js"></script>
<script src="site_libs/bootstrap-3.3.5/shim/html5shiv.min.js"></script>
<script src="site_libs/bootstrap-3.3.5/shim/respond.min.js"></script>
<style>h1 {font-size: 34px;}
h1.title {font-size: 38px;}
h2 {font-size: 30px;}
h3 {font-size: 24px;}
h4 {font-size: 18px;}
h5 {font-size: 16px;}
h6 {font-size: 12px;}
code {color: inherit; background-color: rgba(0, 0, 0, 0.04);}
pre:not([class]) { background-color: white }</style>
<script src="site_libs/jqueryui-1.11.4/jquery-ui.min.js"></script>
<link href="site_libs/tocify-1.9.1/jquery.tocify.css" rel="stylesheet" />
<script src="site_libs/tocify-1.9.1/jquery.tocify.js"></script>
<script src="site_libs/navigation-1.1/tabsets.js"></script>
<link href="site_libs/highlightjs-9.12.0/default.css" rel="stylesheet" />
<script src="site_libs/highlightjs-9.12.0/highlight.js"></script>
<link href="site_libs/font-awesome-5.1.0/css/all.css" rel="stylesheet" />
<link href="site_libs/font-awesome-5.1.0/css/v4-shims.css" rel="stylesheet" />
<style type="text/css">
code{white-space: pre-wrap;}
span.smallcaps{font-variant: small-caps;}
span.underline{text-decoration: underline;}
div.column{display: inline-block; vertical-align: top; width: 50%;}
div.hanging-indent{margin-left: 1.5em; text-indent: -1.5em;}
ul.task-list{list-style: none;}
</style>
<style type="text/css">code{white-space: pre;}</style>
<script type="text/javascript">
if (window.hljs) {
hljs.configure({languages: []});
hljs.initHighlightingOnLoad();
if (document.readyState && document.readyState === "complete") {
window.setTimeout(function() { hljs.initHighlighting(); }, 0);
}
}
</script>
<style type = "text/css">
.main-container {
max-width: 940px;
margin-left: auto;
margin-right: auto;
}
img {
max-width:100%;
}
.tabbed-pane {
padding-top: 12px;
}
.html-widget {
margin-bottom: 20px;
}
button.code-folding-btn:focus {
outline: none;
}
summary {
display: list-item;
}
pre code {
padding: 0;
}
</style>
<style type="text/css">
.dropdown-submenu {
position: relative;
}
.dropdown-submenu>.dropdown-menu {
top: 0;
left: 100%;
margin-top: -6px;
margin-left: -1px;
border-radius: 0 6px 6px 6px;
}
.dropdown-submenu:hover>.dropdown-menu {
display: block;
}
.dropdown-submenu>a:after {
display: block;
content: " ";
float: right;
width: 0;
height: 0;
border-color: transparent;
border-style: solid;
border-width: 5px 0 5px 5px;
border-left-color: #cccccc;
margin-top: 5px;
margin-right: -10px;
}
.dropdown-submenu:hover>a:after {
border-left-color: #adb5bd;
}
.dropdown-submenu.pull-left {
float: none;
}
.dropdown-submenu.pull-left>.dropdown-menu {
left: -100%;
margin-left: 10px;
border-radius: 6px 0 6px 6px;
}
</style>
<script type="text/javascript">
// manage active state of menu based on current page
$(document).ready(function () {
// active menu anchor
href = window.location.pathname
href = href.substr(href.lastIndexOf('/') + 1)
if (href === "")
href = "index.html";
var menuAnchor = $('a[href="' + href + '"]');
// mark it active
menuAnchor.tab('show');
// if it's got a parent navbar menu mark it active as well
menuAnchor.closest('li.dropdown').addClass('active');
// Navbar adjustments
var navHeight = $(".navbar").first().height() + 15;
var style = document.createElement('style');
var pt = "padding-top: " + navHeight + "px; ";
var mt = "margin-top: -" + navHeight + "px; ";
var css = "";
// offset scroll position for anchor links (for fixed navbar)
for (var i = 1; i <= 6; i++) {
css += ".section h" + i + "{ " + pt + mt + "}\n";
}
style.innerHTML = "body {" + pt + "padding-bottom: 40px; }\n" + css;
document.head.appendChild(style);
});
</script>
<!-- tabsets -->
<style type="text/css">
.tabset-dropdown > .nav-tabs {
display: inline-table;
max-height: 500px;
min-height: 44px;
overflow-y: auto;
border: 1px solid #ddd;
border-radius: 4px;
}
.tabset-dropdown > .nav-tabs > li.active:before {
content: "";
font-family: 'Glyphicons Halflings';
display: inline-block;
padding: 10px;
border-right: 1px solid #ddd;
}
.tabset-dropdown > .nav-tabs.nav-tabs-open > li.active:before {
content: "";
border: none;
}
.tabset-dropdown > .nav-tabs.nav-tabs-open:before {
content: "";
font-family: 'Glyphicons Halflings';
display: inline-block;
padding: 10px;
border-right: 1px solid #ddd;
}
.tabset-dropdown > .nav-tabs > li.active {
display: block;
}
.tabset-dropdown > .nav-tabs > li > a,
.tabset-dropdown > .nav-tabs > li > a:focus,
.tabset-dropdown > .nav-tabs > li > a:hover {
border: none;
display: inline-block;
border-radius: 4px;
background-color: transparent;
}
.tabset-dropdown > .nav-tabs.nav-tabs-open > li {
display: block;
float: none;
}
.tabset-dropdown > .nav-tabs > li {
display: none;
}
</style>
<!-- code folding -->
<style type="text/css">
#TOC {
margin: 25px 0px 20px 0px;
}
@media (max-width: 768px) {
#TOC {
position: relative;
width: 100%;
}
}
@media print {
.toc-content {
/* see https://github.com/w3c/csswg-drafts/issues/4434 */
float: right;
}
}
.toc-content {
padding-left: 30px;
padding-right: 40px;
}
div.main-container {
max-width: 1200px;
}
div.tocify {
width: 20%;
max-width: 260px;
max-height: 85%;
}
@media (min-width: 768px) and (max-width: 991px) {
div.tocify {
width: 25%;
}
}
@media (max-width: 767px) {
div.tocify {
width: 100%;
max-width: none;
}
}
.tocify ul, .tocify li {
line-height: 20px;
}
.tocify-subheader .tocify-item {
font-size: 0.90em;
}
.tocify .list-group-item {
border-radius: 0px;
}
</style>
</head>
<body>
<div class="container-fluid main-container">
<!-- setup 3col/9col grid for toc_float and main content -->
<div class="row">
<div class="col-sm-12 col-md-4 col-lg-3">
<div id="TOC" class="tocify">
</div>
</div>
<div class="toc-content col-sm-12 col-md-8 col-lg-9">
<div class="navbar navbar-default navbar-fixed-top" role="navigation">
<div class="container">
<div class="navbar-header">
<button type="button" class="navbar-toggle collapsed" data-toggle="collapse" data-target="#navbar">
<span class="icon-bar"></span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
</button>
<a class="navbar-brand" href="index.html">GenoPred</a>
</div>
<div id="navbar" class="navbar-collapse collapse">
<ul class="nav navbar-nav">
</ul>
<ul class="nav navbar-nav navbar-right">
<li>
<a href="https://github.com/opain/GenoPred">
<span class="fas fa-github fa-lg"></span>
</a>
</li>
</ul>
</div><!--/.nav-collapse -->
</div><!--/.container -->
</div><!--/.navbar -->
<div id="header">
<h1 class="title toc-ignore">Ancestry-based prediction</h1>
</div>
<style>
p.caption {
font-size: 1.5em;
}
</style>
<style type="text/css">
pre code, pre, code {
white-space: pre !important;
overflow-x: scroll !important;
word-break: keep-all !important;
word-wrap: initial !important;
}
</style>
<hr />
<div id="introduction" class="section level1">
<h1><span class="header-section-number">1</span> Introduction</h1>
<p>Genotypes vary in frequency between different ancestral groups. As such, it is important to consider an individuals’ ancestry in genetic association studies to avoid population stratification and to avoid false positives. However, there are many aspects of health that do vary between ancestral groups, however we are unable to determine whether a variant is causally associated with an outcome, or if it is associated by chance.</p>
<p>In the context of prediction, where the aim is to explain the maximum amount of variance in a phenotype with no consideration for what is causal, difference in peoples ancestry may be a useful predictor. Therefore, here we explore whether principal components of population structure (mainly ancestry), typically used as covariates in association studies, can be used a predictor for a range of health outcomes.</p>
<p>Principal components are typically estimated within the sample of interest to capture sources of population structure which are unique to the sample. However, in this context we want the principal components (PCs), to capture variance that can be generalised across samples, to improve the external validity of our prediction models. One way to achieve this is to derive the PCs in a reference sample, and then project the derived PCs into subsequent target samples, ensuring that PCs correspond to the same variance across samples.</p>
<p>The allele frequency differences between ancestries are not only captured by PCs, but also by other genotype-based scores. Hence why PCs are used as covariates for inference when using polygenic risk scores. Therefore, it is also of interest to determine whether PCs can improve prediction over and above polygenic scores.</p>
<p><br/></p>
<hr />
</div>
<div id="aims" class="section level1">
<h1><span class="header-section-number">2</span> Aims</h1>
<ol style="list-style-type: decimal">
<li>Test whether reference-projected PCs can significantly predict variance in a range of phenotypes.</li>
<li>Test whether reference-projected PCs can significantly improve prediction over polygenic scores alone.</li>
<li>Compare the predictive utility of ancestry-specific PCs and across-ancestry PCs.</li>
<li>Determine the optimal number of PCs for prediction.</li>
</ol>
<p><br/></p>
<hr />
</div>
<div id="methods" class="section level1">
<h1><span class="header-section-number">3</span> Methods</h1>
<div id="samples" class="section level2">
<h2><span class="header-section-number">3.1</span> Samples</h2>
<ul>
<li>UK Biobank</li>
<li>TEDS</li>
</ul>
</div>
<div id="outcomes" class="section level2">
<h2><span class="header-section-number">3.2</span> Outcomes</h2>
<ul>
<li>UK Biobank
<ul>
<li>Depression (binary)</li>
<li>Intelligence (continuous)</li>
<li>Body mass index (BMI - continuous)</li>
<li>Height (continuous)</li>
<li>Coronary Artery Disease (CAD - Binary)</li>
<li>Type II Diabetes (T2D - Binary)</li>
</ul></li>
<li>TEDS
<ul>
<li>ADHD traits (continuous)</li>
<li>Height (continuous)</li>
<li>Body mass index (BMI - continuous)</li>
<li>GCSE scores (continuous)</li>
</ul></li>
</ul>
</div>
<div id="genotypic-data" class="section level2">
<h2><span class="header-section-number">3.3</span> Genotypic data</h2>
<p>HapMap3 variants were extracted from the HRC imputed genetic data, and converted to hard-call PLINK format with no hard-call threshold to maximise overlap with the HapMap3 SNP list. Individuals were excluded if they had extensive missing data, had non-European ancestry, or were closely-related to other individuals in the sample. This was done prior to this project.</p>
</div>
<div id="ancestry-scoring" class="section level2">
<h2><span class="header-section-number">3.4</span> Ancestry scoring</h2>
<p>Calculation of the first 100 European-specific and 100 all-ancestry PCs was perfomed using a reference standardised approach (see <a href="https://opain.github.io/GenoPred/Pipeline_prep.html#3_ancestry_scoring">here</a>). In brief, PCs were derived in the 1000 Genomes reference, then projected onto the UK Biobank sample, and subsequent factor scores were then centered and scaled according to the mean and standard deviation of the PCs in an ancestry matched sample. The standard deviation of PCs in UKBB and TEDS was estimated for comparison as this may effect the variance that PCs can explain in each target sample.</p>
</div>
<div id="polygenic-scoring" class="section level2">
<h2><span class="header-section-number">3.5</span> Polygenic scoring</h2>
<p>Polygenic scores were derived using PRScs, a Bayesian shrinkage method. Polygenic scores based on a range of global shrinkage parameters were used as this approach was shown to explain the most variance (see <a href="https://opain.github.io/GenoPred/Determine_optimal_polygenic_scoring_approach.html">here</a>). Furthermore, PRScs-based polygenic scores for a range of outcomes were included, as this approach has been shown to increase the variance explained by polygenic scores (see <a href="https://opain.github.io/GenoPred/Determine_optimal_polygenic_scoring_approach.html">here</a>). Maximising the variance explained by polygenic scores will provide more robust evidence that PCs can explain additional variance to polygenic scores, and therefore are a useful predictor.</p>
<p>Polygenic scores were calculated in UK Biobankand TEDS using a reference-standardised approach (see <a href="https://opain.github.io/GenoPred/Pipeline_prep.html#4_polygenic_scoring">here</a>). In brief, all scores were derived using HapMap3 SNPs only, modelling LD based on European individuals within the 1000 Genomes reference. Any HapMap3 missing in the target sample are imputed using the allele frequency measured in the European subset of the 1000 Genomes reference. Score were centreed and scaled based on the mean and standard deviation of scores in an ancestry matched samples.</p>
</div>
<div id="estimating-predictive-ability" class="section level2">
<h2><span class="header-section-number">3.6</span> Estimating predictive ability</h2>
<p>All models were derived using elastic-net regularisation to reduce the likelihood of overfitting and account for multicollinearity when modelling highly correlated predictors. 10-fold cross validation was performed using 80% of individuals to identify optimal parameters, with subsequent test-set validation in the remaining 20% of individuals to estimate the predictive utility among individuals not included in the parameter selection process. Permutation testing was used to determine whether the difference in variance explained was significantly greater than 0.</p>
<p>Model building and evaluation was performed using an Rscript called Model_builder.R (more information <a href="https://github.com/opain/GenoPred/tree/master/Scripts/Model_builder">here</a>).</p>
</div>
<div id="code" class="section level2">
<h2><span class="header-section-number">3.7</span> Code</h2>
<div id="calculate-descriptives-of-pcs-in-teds-and-ukbb" class="section level3">
<h3><span class="header-section-number">3.7.1</span> Calculate descriptives of PCs in TEDS and UKBB</h3>
<details>
<p><summary>Evaluate PCs alone</summary></p>
<pre class="bash"><code>module add general/R/3.5.0
R
library(data.table)
##
# UKBB
##
# Read in the UKBB PCs
UKBB_EUR_PCs<-fread('/users/k1806347/brc_scratch/Data/UKBB/Projected_PCs/EUR/UKBB.w_hm3.EUR.eigenvec')
UKBB_All_PCs<-fread('/users/k1806347/brc_scratch/Data/UKBB/Projected_PCs/EUR/AllAncestry/UKBB.w_hm3.AllAncestry.EUR.eigenvec')
# Read in the TEDS PCs
TEDS_EUR_PCs<-fread('/users/k1806347/brc_scratch/Data/TEDS/Projected_PCs/EUR/EUR_specific/TEDS.w_hm3.EUR.EUR.eigenvec')
TEDS_All_PCs<-fread('/users/k1806347/brc_scratch/Data/TEDS/Projected_PCs/EUR/AllAncestry/TEDS.w_hm3.AllAncestry.EUR.eigenvec')
# Extract individuals with European ancestry
UKBB_EUR_keep<-fread('/users/k1806347/brc_scratch/Data/UKBB/Projected_PCs/Ancestry_idenitfier/UKBB.w_hm3.AllAncestry.EUR.keep')
UKBB_EUR_PCs<-UKBB_EUR_PCs[(UKBB_EUR_PCs$IID %in% UKBB_EUR_keep$V2),]
UKBB_All_PCs<-UKBB_All_PCs[(UKBB_All_PCs$IID %in% UKBB_EUR_keep$V2),]
TEDS_EUR_keep<-fread('/users/k1806347/brc_scratch/Data/TEDS/Projected_PCs/Ancestry_idenitfier/TEDS.w_hm3.AllAncestry.EUR.keep')
TEDS_EUR_PCs<-TEDS_EUR_PCs[(TEDS_EUR_PCs$IID %in% TEDS_EUR_keep$V2),]
TEDS_All_PCs<-TEDS_All_PCs[(TEDS_All_PCs$IID %in% TEDS_EUR_keep$V2),]
# Calculate mean and sd
UKBB_EUR_PCs_desc<-NULL
for(i in names(UKBB_EUR_PCs)[-1:-2]){
tmp<-data.frame(Mean_UKBB_EUR=mean(UKBB_EUR_PCs[[i]]),
SD_UKBB_EUR=sd(UKBB_EUR_PCs[[i]]))
UKBB_EUR_PCs_desc<-rbind(UKBB_EUR_PCs_desc,tmp)
}
UKBB_All_PCs_desc<-NULL
for(i in names(UKBB_All_PCs)[-1:-2]){
tmp<-data.frame(Mean_UKBB_All=mean(UKBB_All_PCs[[i]]),
SD_UKBB_All=sd(UKBB_All_PCs[[i]]))
UKBB_All_PCs_desc<-rbind(UKBB_All_PCs_desc,tmp)
}
TEDS_EUR_PCs_desc<-NULL
for(i in names(TEDS_EUR_PCs)[-1:-2]){
tmp<-data.frame(Mean_TEDS_EUR=mean(TEDS_EUR_PCs[[i]]),
SD_TEDS_EUR=sd(TEDS_EUR_PCs[[i]]))
TEDS_EUR_PCs_desc<-rbind(TEDS_EUR_PCs_desc,tmp)
}
TEDS_All_PCs_desc<-NULL
for(i in names(TEDS_All_PCs)[-1:-2]){
tmp<-data.frame(Mean_TEDS_All=mean(TEDS_All_PCs[[i]]),
SD_TEDS_All=sd(TEDS_All_PCs[[i]]))
TEDS_All_PCs_desc<-rbind(TEDS_All_PCs_desc,tmp)
}
PCs_desc<-do.call(cbind, list(UKBB_EUR_PCs_desc,UKBB_All_PCs_desc,TEDS_EUR_PCs_desc,TEDS_All_PCs_desc))
PCs_desc<-data.frame(PC=1:dim(PCs_desc)[1], PCs_desc)
write.csv(PCs_desc[c('PC','Mean_UKBB_EUR','SD_UKBB_EUR','Mean_UKBB_All','SD_UKBB_All')], '/users/k1806347/brc_scratch/Analyses/UKBB_outcomes_for_prediction/UKBB_PC_descriptives.csv', col.names=T, row.names=F, quote=F)
write.csv(PCs_desc[c('PC','Mean_TEDS_EUR','SD_TEDS_EUR','Mean_TEDS_All','SD_TEDS_All')], '/users/k1806347/brc_scratch/Analyses/TEDS_outcomes_for_prediction/TEDS_PC_descriptives.csv', col.names=T, row.names=F, quote=F)
q()
n
</code></pre>
</details>
</div>
</div>
<div id="pcs-alone" class="section level2">
<h2><span class="header-section-number">3.8</span> PCs alone</h2>
<div id="uk-biobank" class="section level3">
<h3><span class="header-section-number">3.8.1</span> UK Biobank</h3>
<details>
<p><summary>Evaluate PCs alone</summary></p>
<pre class="bash"><code># Make required directories
for pheno_i in $(echo Depression Intelligence BMI Height T2D CAD);do
mkdir -p /users/k1806347/brc_scratch/Analyses/UKBB_outcomes_for_prediction/${pheno_i}/Association_withPCs
done
######
# Create a file to group the predictors
######
module add general/R/3.5.0
R
library(data.table)
groups<-data.frame( predictor=c('/users/k1806347/brc_scratch/Data/UKBB/Projected_PCs/EUR/UKBB.w_hm3.EUR.eigenvec','/users/k1806347/brc_scratch/Data/UKBB/Projected_PCs/EUR/AllAncestry/UKBB.w_hm3.AllAncestry.EUR.eigenvec'),
group=c('EUR_PCs','AllAncestry_PCs'))
write.table(groups, '/users/k1806347/brc_scratch/Analyses/UKBB_outcomes_for_prediction/UKBB-projected-PCs.predictor_groups', col.names=T, row.names=F, quote=F)
q()
n
######
# Derive and evaluate models
######
pheno=$(echo Depression Intelligence BMI Height T2D CAD)
pheno_file=$(echo ever_depressed_pheno_final.UpdateIDs.txt UKBB_Fluid.intelligence.score.UpdateIDs.pheno UKBB_BMI.score.UpdateIDs.pheno UKBB_Height.score.UpdateIDs.pheno t2d_only_111119.UpdateIDs.txt cad_only_111119.UpdateIDs.txt)
gwas=$(echo DEPR06 COLL01 BODY03 HEIG03 DIAB05 COAD01)
prev=$(echo 0.15 NA NA NA 0.05 0.03)
for i in $(seq 1 6);do
pheno_i=$(echo ${pheno} | cut -f ${i} -d ' ')
pheno_file_i=$(echo ${pheno_file} | cut -f ${i} -d ' ')
gwas_i=$(echo ${gwas} | cut -f ${i} -d ' ')
prev_i=$(echo ${prev} | cut -f ${i} -d ' ')
qsub -l h_vmem=10G -pe smp 5 /users/k1806347/brc_scratch/Software/Rscript.sh /users/k1806347/brc_scratch/Software/Scripts/Model_builder.R \
--pheno /users/k1806347/brc_scratch/Data/UKBB/Phenotype/${pheno_i}/${pheno_file_i} \
--keep /users/k1806347/brc_scratch/Data/UKBB/Projected_PCs/Ancestry_idenitfier/UKBB.w_hm3.AllAncestry.EUR.keep \
--out /users/k1806347/brc_scratch/Analyses/UKBB_outcomes_for_prediction/${pheno_i}/Association_withPCs/UKBB-projected-PCs \
--n_core 5 \
--assoc T \
--outcome_pop_prev ${prev_i} \
--predictors /users/k1806347/brc_scratch/Analyses/UKBB_outcomes_for_prediction/UKBB-projected-PCs.predictor_groups
done
</code></pre>
</details>
</div>
<div id="teds" class="section level3">
<h3><span class="header-section-number">3.8.2</span> TEDS</h3>
<details>
<p><summary>Evaluate PCs alone</summary></p>
<pre class="bash"><code># Make required directories
for pheno_i in $(echo Height21 BMI21 GCSE ADHD);do
mkdir -p /users/k1806347/brc_scratch/Analyses/TEDS_outcomes_for_prediction/${pheno_i}/Association_withPCs
done
######
# Create a file to group the predictors
######
module add general/R/3.5.0
R
library(data.table)
groups<-data.frame( predictor=c('/users/k1806347/brc_scratch/Data/TEDS/Projected_PCs/EUR/EUR_specific/TEDS.w_hm3.EUR.EUR.eigenvec','/users/k1806347/brc_scratch/Data/TEDS/Projected_PCs/EUR/AllAncestry/TEDS.w_hm3.AllAncestry.EUR.eigenvec'),
group=c('EUR_PCs','AllAncestry_PCs'))
write.table(groups, '/users/k1806347/brc_scratch/Analyses/TEDS_outcomes_for_prediction/TEDS-projected-PCs.predictor_groups', col.names=T, row.names=F, quote=F)
q()
n
######
# Derive and evaluate models
######
pheno=$(echo Height21 BMI21 GCSE ADHD)
gwas=$(echo HEIG03 BODY11 EDUC03 ADHD04)
prev=$(echo NA NA NA NA)
for i in $(seq 1 4);do
pheno_i=$(echo ${pheno} | cut -f ${i} -d ' ')
gwas_i=$(echo ${gwas} | cut -f ${i} -d ' ')
prev_i=$(echo ${prev} | cut -f ${i} -d ' ')
qsub -l h_vmem=6G -pe smp 3 /users/k1806347/brc_scratch/Software/Rscript.sh /users/k1806347/brc_scratch/Software/Scripts/Model_builder.R \
--pheno /users/k1806347/brc_scratch/Data/TEDS/Phenotypic/Derived_outcomes/TEDS_${pheno_i}.txt \
--keep /users/k1806347/brc_scratch/Data/TEDS/Projected_PCs/Ancestry_idenitfier/TEDS.w_hm3.AllAncestry.EUR.keep \
--out /users/k1806347/brc_scratch/Analyses/TEDS_outcomes_for_prediction/${pheno_i}/Association_withPCs/TEDS-projected-PCs \
--n_core 3 \
--assoc T \
--outcome_pop_prev ${prev_i} \
--predictors /users/k1806347/brc_scratch/Analyses/TEDS_outcomes_for_prediction/TEDS-projected-PCs.predictor_groups
done
</code></pre>
</details>
<p><br/></p>
</div>
</div>
<div id="pcs-prss" class="section level2">
<h2><span class="header-section-number">3.9</span> PCs + PRSs</h2>
<p>Here we will test PCs can significantly improve prediction over multiple-trait PRSs alone.</p>
</div>
<div id="prss-controlled-for-pcs" class="section level2">
<h2><span class="header-section-number">3.10</span> PRSs controlled for PCs</h2>
<p>Here will test whether PRSs controlled for PCs improve prediction. We will adjust PRSs for PCs using the EUR 1KG reference sample.</p>
<hr />
</div>
</div>
<div id="results" class="section level1">
<h1><span class="header-section-number">4</span> Results</h1>
<div id="descriptives" class="section level2">
<h2><span class="header-section-number">4.1</span> Descriptives</h2>
<details>
<p><summary>Show PC descriptives</summary></p>
</details>
<p><br/></p>
</div>
<div id="pcs-alone-1" class="section level2">
<h2><span class="header-section-number">4.2</span> PCs alone</h2>
<div id="uk-biobank-1" class="section level3">
<h3><span class="header-section-number">4.2.1</span> UK biobank</h3>
<details>
<p><summary>Show UK Biobank results table</summary></p>
</details>
<p><br/></p>
</div>
<div id="teds-1" class="section level3">
<h3><span class="header-section-number">4.2.2</span> TEDS</h3>
<details>
<p><summary>Show TEDS results table</summary></p>
</details>
<p><br/></p>
<p>PCs show significant prediction in the test set of UKBB, but not in TEDS. The estimates of correlation between predicted and observed values are more accurate in UKBB than TEDS due to a larger sample size. TEDS may not have sufficient sample size to detect significant prediction using PCs.</p>
<p><br/></p>
<hr />
</div>
</div>
</div>
<div id="discussion" class="section level1">
<h1><span class="header-section-number">5</span> Discussion</h1>
<p><br/></p>
<hr />
</div>
<div id="conclusion" class="section level1">
<h1><span class="header-section-number">6</span> Conclusion</h1>
<hr />
</div>
</div>
</div>
</div>
<script>
// add bootstrap table styles to pandoc tables
function bootstrapStylePandocTables() {
$('tr.odd').parent('tbody').parent('table').addClass('table table-condensed');
}
$(document).ready(function () {
bootstrapStylePandocTables();
});
</script>
<!-- tabsets -->
<script>
$(document).ready(function () {
window.buildTabsets("TOC");
});
$(document).ready(function () {
$('.tabset-dropdown > .nav-tabs > li').click(function () {
$(this).parent().toggleClass('nav-tabs-open');
});
});
</script>
<!-- code folding -->
<script>
$(document).ready(function () {
// move toc-ignore selectors from section div to header
$('div.section.toc-ignore')
.removeClass('toc-ignore')
.children('h1,h2,h3,h4,h5').addClass('toc-ignore');
// establish options
var options = {
selectors: "h1,h2,h3",
theme: "bootstrap3",
context: '.toc-content',
hashGenerator: function (text) {
return text.replace(/[.\\/?&!#<>]/g, '').replace(/\s/g, '_');
},
ignoreSelector: ".toc-ignore",
scrollTo: 0
};
options.showAndHide = true;
options.smoothScroll = true;
// tocify
var toc = $("#TOC").tocify(options).data("toc-tocify");
});
</script>
<!-- dynamically load mathjax for compatibility with self-contained -->
<script>
(function () {
var script = document.createElement("script");
script.type = "text/javascript";
script.src = "https://mathjax.rstudio.com/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML";
document.getElementsByTagName("head")[0].appendChild(script);
})();
</script>
</body>
</html>