43 |
44 | ## Installation
45 |
46 | The current github version of *PhenotypeSimulator* is 0.3.4 and can be
47 | installed via:
48 |
49 | ``` r
50 | library(devtools)
51 | install_github("HannahVMeyer/PhenotypeSimulator")
52 | ```
53 |
54 | The current CRAN version of *PhenotypeSimulator* is 0.3.4 and can be
55 | installed via:
56 |
57 | ``` r
58 | install.packages("PhenotypeSimulator")
59 | ```
60 |
61 | A log of version changes can be found
62 | [here](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/NEWS.md).
63 |
64 | ## Citation
65 |
66 | Meyer, HV & Birney E (2018) [PhenotypeSimulator: A comprehensive
67 | framework for simulating multi-trait, multi-locus genotype to phenotype
68 | relationships](https://doi.org/10.1093/bioinformatics/bty197),
69 | *Bioinformatics*, 34(17):2951–2956
70 |
--------------------------------------------------------------------------------
/README.Rmd:
--------------------------------------------------------------------------------
1 | ---
2 | output: github_document
3 | ---
4 |
5 |
6 |
7 | ```{r, echo = FALSE}
8 | knitr::opts_chunk$set(
9 | collapse = TRUE,
10 | comment = "#>",
11 | fig.path = "README-"
12 | )
13 | ```
14 |
15 | [](https://cran.r-project.org/package=PhenotypeSimulator)
16 | [](https://travis-ci.org/HannahVMeyer/PhenotypeSimulator)
17 | [](https://opensource.org/licenses/MIT)
18 | [](https://CRAN.R-project.org/package=PhenotypeSimulator)
19 |
20 |
21 | ## PhenotypeSimulator
22 |
23 | **PhenotypeSimulator** allows for the flexible simulation of phenotypes from
24 | different genetic and non-genetic (noise) components.
25 |
26 | In quantitative genetics, genotype to phenotype mapping is commonly realised by
27 | fitting a linear model to the genotype as the explanatory variable and the
28 | phenotype as the response variable. Other explanatory variable such as
29 | additional sample measures (e.g. age, height, weight) or batch effects can also
30 | be included. For linear mixed models, in addition to the fixed effects of the
31 | genotype and the covariates, different random effect components can be included,
32 | accounting for population structure in the study cohort or environmental
33 | effects. The application of linear and linear mixed models in quantitative
34 | genetics ranges from genetic studies in model organism such as yeast and
35 | *Arabidopsis thaliana* to human molecular, morphological or imaging derived
36 | traits. Developing new methods for increasing numbers of sample cohorts,
37 | phenotypic measurements or complexity of phenotypes to analyse, often requires
38 | the simulation of datasets with a specific underlying phenotype structure.
39 |
40 | **PhenotypeSimulator** allows for the simulation of complex phenotypes under
41 | different models, including genetic variant effects and infinitesimal genetic
42 | effects (reflecting population structure) as well as correlated, non-genetic
43 | covariates and observational noise effects. Different phenotypic effects can be
44 | combined into a final phenotype while controlling for the proportion of variance
45 | explained by each of the components. For each component, the number of
46 | variables, their distribution and the design of their effect across traits can
47 | be customised.
48 |
49 | ## Installation
50 |
51 | Full documentation of **PhenotypeSimulator** is available at
52 | https://HannahVMeyer.github.io/PhenotypeSimulator/.
53 |
54 | The current github version of *PhenotypeSimulator* is 0.3.4 and can be
55 | installed via
56 |
57 | ```{r gh-installation, eval = FALSE}
58 | # install.packages("devtools")
59 | devtools::install_github("HannahVMeyer/PhenotypeSimulator")
60 | ```
61 |
62 | The current CRAN version of *PhenotypeSimulator* is 0.3.4 and can be installed
63 | via:
64 | ```{r, eval=FALSE}
65 | install.packages("PhenotypeSimulator")
66 | ```
67 |
68 | A log of version changes can be found [here](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/NEWS.md).
69 |
70 | ## Citation
71 | Meyer, HV & Birney E (2018) [PhenotypeSimulator: A comprehensive framework for simulating multi-trait, multi-locus genotype to phenotype relationships](https://doi.org/10.1093/bioinformatics/bty197), *Bioinformatics*, 34(17):2951–2956
72 |
--------------------------------------------------------------------------------
/INDEX.Rmd:
--------------------------------------------------------------------------------
1 | ---
2 | output: github_document
3 | ---
4 |
5 |
6 |
7 | ```{r, echo = FALSE}
8 | knitr::opts_chunk$set(
9 | collapse = TRUE,
10 | comment = "#>",
11 | fig.path = "README-"
12 | )
13 | ```
14 | [](https://cran.r-project.org/package=PhenotypeSimulator)
15 | [](https://travis-ci.org/HannahVMeyer/PhenotypeSimulator)
16 | [](https://opensource.org/licenses/MIT)
17 | [](http://cran.rstudio.com/web/packages/PhenotypeSimulator/index.html)
18 |
19 |
20 | ## PhenotypeSimulator
21 |
22 | **PhenotypeSimulator** allows for the flexible simulation of phenotypes from
23 | different genetic and non-genetic (noise) components.
24 |
25 | In quantitative genetics, genotype to phenotype mapping is commonly realised by
26 | fitting a linear model to the genotype as the explanatory variable and the
27 | phenotype as the response variable. Other explanatory variable such as
28 | additional sample measures (e.g. age, height, weight) or batch effects can also
29 | be included. For linear mixed models, in addition to the fixed effects of the
30 | genotype and the covariates, different random effect components can be included,
31 | accounting for population structure in the study cohort or environmental
32 | effects. The application of linear and linear mixed models in quantitative
33 | genetics ranges from genetic studies in model organism such as yeast and
34 | *Arabidopsis thaliana* to human molecular, morphological or imaging derived
35 | traits. Developing new methods for increasing numbers of sample cohorts,
36 | phenotypic measurements or complexity of phenotypes to analyse, often requires
37 | the simulation of datasets with a specific underlying phenotype structure.
38 |
39 | **PhenotypeSimulator** allows for the simulation of complex phenotypes under
40 | different models, including genetic variant effects and infinitesimal genetic
41 | effects (reflecting population structure) as well as correlated, non-genetic
42 | covariates and observational noise effects. Different phenotypic effects can be
43 | combined into a final phenotype while controlling for the proportion of variance
44 | explained by each of the components. For each component, the number of
45 | variables, their distribution and the design of their effect across traits can
46 | be customised.
47 |
48 |
49 | ```{r, echo = FALSE, out.width='100%'}
50 | knitr::include_graphics("docs/simulatedPhenotypes.png")
51 | ```
52 |
53 |
54 | ## Installation
55 |
56 | The current github version of *PhenotypeSimulator* is 0.3.4 and can be
57 | installed via:
58 | ```{r, eval=FALSE}
59 | library(devtools)
60 | install_github("HannahVMeyer/PhenotypeSimulator")
61 | ```
62 | The current CRAN version of *PhenotypeSimulator* is 0.3.4 and can be installed
63 | via:
64 | ```{r, eval=FALSE}
65 | install.packages("PhenotypeSimulator")
66 | ```
67 |
68 | A log of version changes can be found [here](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/NEWS.md).
69 |
70 | ## Citation
71 |
72 | Meyer, HV & Birney E (2018) [PhenotypeSimulator: A comprehensive framework for simulating multi-trait, multi-locus genotype to phenotype relationships](https://doi.org/10.1093/bioinformatics/bty197), *Bioinformatics*, 34(17):2951–2956
73 |
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
--------------------------------------------------------------------------------
/docs/pkgdown.js:
--------------------------------------------------------------------------------
1 | /* http://gregfranko.com/blog/jquery-best-practices/ */
2 | (function($) {
3 | $(function() {
4 |
5 | $('.navbar-fixed-top').headroom();
6 |
7 | $('body').css('padding-top', $('.navbar').height() + 10);
8 | $(window).resize(function(){
9 | $('body').css('padding-top', $('.navbar').height() + 10);
10 | });
11 |
12 | $('[data-toggle="tooltip"]').tooltip();
13 |
14 | var cur_path = paths(location.pathname);
15 | var links = $("#navbar ul li a");
16 | var max_length = -1;
17 | var pos = -1;
18 | for (var i = 0; i < links.length; i++) {
19 | if (links[i].getAttribute("href") === "#")
20 | continue;
21 | // Ignore external links
22 | if (links[i].host !== location.host)
23 | continue;
24 |
25 | var nav_path = paths(links[i].pathname);
26 |
27 | var length = prefix_length(nav_path, cur_path);
28 | if (length > max_length) {
29 | max_length = length;
30 | pos = i;
31 | }
32 | }
33 |
34 | // Add class to parent `, then ``, etc.) that has more than one element. Defaults to 1 (for ``).
76 | getTopLevel: function($scope) {
77 | for (var i = 1; i <= 6; i++) {
78 | var $headings = this.findOrFilter($scope, 'h' + i);
79 | if ($headings.length > 1) {
80 | return i;
81 | }
82 | }
83 |
84 | return 1;
85 | },
86 |
87 | // returns the elements for the top level, and the next below it
88 | getHeadings: function($scope, topLevel) {
89 | var topSelector = 'h' + topLevel;
90 |
91 | var secondaryLevel = topLevel + 1;
92 | var secondarySelector = 'h' + secondaryLevel;
93 |
94 | return this.findOrFilter($scope, topSelector + ',' + secondarySelector);
95 | },
96 |
97 | getNavLevel: function(el) {
98 | return parseInt(el.tagName.charAt(1), 10);
99 | },
100 |
101 | populateNav: function($topContext, topLevel, $headings) {
102 | var $context = $topContext;
103 | var $prevNav;
104 |
105 | var helpers = this;
106 | $headings.each(function(i, el) {
107 | var $newNav = helpers.generateNavItem(el);
108 | var navLevel = helpers.getNavLevel(el);
109 |
110 | // determine the proper $context
111 | if (navLevel === topLevel) {
112 | // use top level
113 | $context = $topContext;
114 | } else if ($prevNav && $context === $topContext) {
115 | // create a new level of the tree and switch to it
116 | $context = helpers.createChildNavList($prevNav);
117 | } // else use the current $context
118 |
119 | $context.append($newNav);
120 |
121 | $prevNav = $newNav;
122 | });
123 | },
124 |
125 | parseOps: function(arg) {
126 | var opts;
127 | if (arg.jquery) {
128 | opts = {
129 | $nav: arg
130 | };
131 | } else {
132 | opts = arg;
133 | }
134 | opts.$scope = opts.$scope || $(document.body);
135 | return opts;
136 | }
137 | },
138 |
139 | // accepts a jQuery object, or an options object
140 | init: function(opts) {
141 | opts = this.helpers.parseOps(opts);
142 |
143 | // ensure that the data attribute is in place for styling
144 | opts.$nav.attr('data-toggle', 'toc');
145 |
146 | var $topContext = this.helpers.createChildNavList(opts.$nav);
147 | var topLevel = this.helpers.getTopLevel(opts.$scope);
148 | var $headings = this.helpers.getHeadings(opts.$scope, topLevel);
149 | this.helpers.populateNav($topContext, topLevel, $headings);
150 | }
151 | };
152 |
153 | $(function() {
154 | $('nav[data-toggle="toc"]').each(function(i, el) {
155 | var $nav = $(el);
156 | Toc.init($nav);
157 | });
158 | });
159 | })();
160 |
--------------------------------------------------------------------------------
/vignettes/SimulationAndLinearModel.bib:
--------------------------------------------------------------------------------
1 | @article{Zhou2014,
2 | abstract = {Multivariate linear mixed models (mvLMMs) are powerful tools for testing associations between single-nucleotide polymorphisms and multiple correlated phenotypes while controlling for population stratification in genome-wide association studies. We present efficient algorithms in the genome-wide efficient mixed model association (GEMMA) software for fitting mvLMMs and computing likelihood ratio tests. These algorithms offer improved computation speed, power and P-value calibration over existing methods, and can deal with more than two phenotypes.},
3 | author = {Zhou, Xiang and Stephens, Matthew},
4 | doi = {10.1038/nmeth.2848},
5 | file = {:Users/hannah/Documents/Mendeley Desktop/2014{\_}Nature Methods{\_}Zhou, Stephens(3).pdf:pdf;:Users/hannah/Documents/Mendeley Desktop/2014{\_}Nature Methods{\_}Zhou, Stephens(4).pdf:pdf},
6 | isbn = {1548-7105 (Electronic)$\backslash$r1548-7091 (Linking)},
7 | issn = {1548-7105},
8 | journal = {Nature Methods},
9 | keywords = {Algorithms,Genome,Likelihood Functions,Linear Models,Multivariate Analysis,Polymorphism,Single Nucleotide,Single Nucleotide: genetics,Software},
10 | number = {4},
11 | pages = {407--9},
12 | pmid = {24531419},
13 | title = {{Efficient multivariate linear mixed model algorithms for genome-wide association studies.}},
14 | url = {http://www.ncbi.nlm.nih.gov/pubmed/24531419},
15 | volume = {11},
16 | year = {2014}
17 | }
18 | @article{Chang2015,
19 | abstract = {PLINK 1 is a widely used open-source C/C++ toolset for genome-wide association studies (GWAS) and research in population genetics. However, the steady accumulation of data from imputation and whole-genome sequencing studies has exposed a strong need for even faster and more scalable implementations of key functions. In addition, GWAS and population-genetic data now frequently contain probabilistic calls, phase information, and/or multiallelic variants, none of which can be represented by PLINK 1's primary data format. To address these issues, we are developing a second-generation codebase for PLINK. The first major release from this codebase, PLINK 1.9, introduces extensive use of bit-level parallelism, O(sqrt(n))-time/constant-space Hardy-Weinberg equilibrium and Fisher's exact tests, and many other algorithmic improvements. In combination, these changes accelerate most operations by 1-4 orders of magnitude, and allow the program to handle datasets too large to fit in RAM. This will be followed by PLINK 2.0, which will introduce (a) a new data format capable of efficiently representing probabilities, phase, and multiallelic variants, and (b) extensions of many functions to account for the new types of information. The second-generation versions of PLINK will offer dramatic improvements in performance and compatibility. For the first time, users without access to high-end computing resources can perform several essential analyses of the feature-rich and very large genetic datasets coming into use.},
20 | author = {Chang, Christopher C and Chow, Carson C and Tellier, Laurent CAM and Vattikuti, Shashaank and Purcell, Shaun M and Lee, James J},
21 | doi = {10.1186/s13742-015-0047-8},
22 | issn = {2047-217X},
23 | journal = {GigaScience},
24 | month = {feb},
25 | number = {1},
26 | pages = {7},
27 | title = {{Second-generation PLINK: rising to the challenge of larger and richer datasets}},
28 | url = {http://arxiv.org/abs/1410.4803},
29 | volume = {4},
30 | year = {2015}
31 | }
32 | @article{Halsey2015,
33 | author = {Halsey, Lewis G and Curran-Everett, Douglas and Vowler, Sarah L and Drummond, Gordon B},
34 | doi = {10.1038/nmeth.3288},
35 | issn = {1548-7091},
36 | journal = {Nature Methods},
37 | month = {feb},
38 | number = {3},
39 | pages = {179--185},
40 | pmid = {25719825},
41 | title = {{The fickle P value generates irreproducible results}},
42 | url = {http://www.ncbi.nlm.nih.gov/pubmed/25719825 http://www.nature.com/doifinder/10.1038/nmeth.3288},
43 | volume = {12},
44 | year = {2015}
45 | }
46 | @article{Cohen1992,
47 | abstract = {One possible reason for the continued neglect of statistical power analysis in research in the behavioral sciences is the inaccessibility of or difficulty with the standard material. A convenient, although not comprehensive, presentation of required sample sizes is provided here. Effect-size indexes and conventional values for these are given for operationally defined small, medium, and large effects. The sample sizes necessary for .80 power to detect effects at these levels are tabled for eight standard statistical tests: (a) the difference between independent means, (b) the significance of a product-moment correlation, (c) the difference between independent rs, (d) the sign test, (e) the difference between independent proportions, (f) chi-square tests for goodness of fit and contingency tables, (g) one-way analysis of variance, and (h) the significance of a multiple or multiple partial correlation.},
48 | author = {Cohen, J},
49 | issn = {0033-2909},
50 | journal = {Psychological Bulletin},
51 | month = {jul},
52 | number = {1},
53 | pages = {155--9},
54 | pmid = {19565683},
55 | title = {{A power primer.}},
56 | url = {http://www.ncbi.nlm.nih.gov/pubmed/19565683},
57 | volume = {112},
58 | year = {1992}
59 | }
60 | @article{Su2011,
61 | author = {Su, Zhan and Marchini, Jonathan and Donnelly, Peter},
62 | doi = {10.1093/bioinformatics/btr341},
63 | file = {:Users/hannah/Documents/Mendeley Desktop/2011{\_}Bioinformatics{\_}Su, Marchini, Donnelly.pdf:pdf},
64 | issn = {1460-2059},
65 | journal = {Bioinformatics},
66 | month = {aug},
67 | number = {16},
68 | pages = {2304--2305},
69 | publisher = {Oxford University Press},
70 | title = {{HAPGEN2: simulation of multiple disease SNPs}},
71 | url = {https://academic.oup.com/bioinformatics/article-lookup/doi/10.1093/bioinformatics/btr341},
72 | volume = {27},
73 | year = {2011}
74 | }
75 |
--------------------------------------------------------------------------------
/tests/testthat/test-utility.r:
--------------------------------------------------------------------------------
1 | context('Test utility functions')
2 |
3 | test_that("vmessage fails when non-character specified as separator", {
4 | expect_error(vmessage(c("hello", "world"), sep=9))
5 | })
6 |
7 | test_that("vmessage returns correct messsage", {
8 | expect_that(vmessage(c("Hello", "World"), sep="Little"),
9 | shows_message("HelloLittleWorld"))
10 | })
11 |
12 | test_that('vmessage returns NULL when verbose set to FALSE', {
13 | expect_equal(vmessage("Hello world", verbose=FALSE), NULL)
14 | })
15 |
16 | test_that('commaList2vector returns vector from comma-separated inputlist', {
17 | expect_equal(commaList2vector("1,2,3"), c(1,2,3))
18 | })
19 |
20 | test_that('commaList2vector returns NULL', {
21 | expect_equal(commaList2vector(), NULL)
22 | })
23 |
24 | test_that('commaList2vector fails with unknown type', {
25 | expect_error(commaList2vector(type='matrix', commastring="1,2,2,3,4"),
26 | "Unknown type of comma-separated list elements")
27 | })
28 | test_that('addNonNulls fails when dimensions of list elements are not the same',
29 | {
30 | expect_error(addNonNulls(list(matrix(1:10, ncol=2), matrix(11:20, ncol=2),
31 | matrix(21:30, ncol=5), NULL, NULL)),
32 | "Column dimensions of list elements are different")
33 | expect_error(addNonNulls(list(matrix(1:10, ncol=2), matrix(11:20, ncol=2),
34 | matrix(21:40, ncol=2), NULL, NULL)),
35 | "Row dimensions of list elements are different")
36 | })
37 |
38 | test_that('addNonNulls fails when non-list supplied as input', {
39 | expect_error(addNonNulls(matrix(1:10, ncol=2)),
40 | "addNonNulls expects input of type list")
41 | })
42 |
43 | test_that('simulateDist fails when distr provided is not one of "unif", "norm",
44 | "bin", "cat_norm" or "cat_unif" ', {
45 | expect_error(simulateDist(x=10, dist="gamma"),"Unknown distribution")
46 | })
47 |
48 | test_that('simulateDist fails when number of observations is less than 0', {
49 | expect_error(simulateDist(x=-10, dist="bin"),
50 | "The number of observations to simulate")
51 | })
52 |
53 | test_that('simulateDist fails when standard deviation for normal distribution
54 | less than 0', {
55 | expect_error(simulateDist(x=10, dist="norm", std=-0.3),
56 | "Simulating normal distribution")
57 | })
58 |
59 | test_that('simulateDist fails when probability for binomial dist is not
60 | provided', {
61 | expect_error(simulateDist(x=10, dist="bin"),
62 | "Simulating binomial distribution")
63 | })
64 |
65 | test_that('simulateDist fails when probability for binomial dist is less
66 | than zero', {
67 | expect_error(simulateDist(x=10, dist="bin", prob=-0.02),
68 | "Simulating binomial distribution")
69 | })
70 |
71 | test_that('simulateDist fails with unknown distribution', {
72 | expect_error(simulateDist(x=10, dist="beta"), "Unknown distribution")
73 | })
74 |
75 | test_that('simulateDist fails with length of distribution argument', {
76 | dist=c("unif", "norm", "bin", "cat_norm", "cat_unif")
77 | expect_error(simulateDist(x=10),
78 | paste("Please specify exactly one distribution to sample from,",
79 | "currently ", length(dist), " provided.", sep=""))
80 | })
81 |
82 | test_that('simulateDist fails when number of categories for categorical dist
83 | is not provided', {
84 | expect_error(simulateDist(x=10, dist="cat_unif"),
85 | "Simulating categorical distribution")
86 | })
87 |
88 | test_that('simulateDist fails when number of categories for categorical dist
89 | is less than zero', {
90 | expect_error(simulateDist(x=10, dist="cat_unif", categories=-2),
91 | "Simulating categorical distribution")
92 | })
93 |
94 | test_that("probGen2expGen fails when input is not a numeric vector", {
95 | expect_error( probGen2expGen(list(c(0.1, 0.9, 0))),
96 | "probGen2expGen takes")
97 | })
98 |
99 | test_that("probGen2expGen fails when input vector is not a multiple of three", {
100 | expect_error( probGen2expGen(c(0.1, 0.9)),
101 | "Length of genotype probabilty vector")
102 | })
103 |
104 | test_that("probGen2expGen fails with NA", {
105 | expect_error( probGen2expGen(c(0.1, 0.8, NA)),
106 | "Samples need to be fully genotyped")
107 | })
108 |
109 | test_that("probGen2expGen fails with incorrect probabilities", {
110 | expect_error( probGen2expGen(c(0.1, 0.8, 0)),
111 | "Genotype probabilities do not sum to 1")
112 | })
113 |
114 | test_that('expGen2probGen fails when wrong encoding is given', {
115 | expect_error(expGen2probGen(c(0,2,3)),
116 | "Genotypes can only be encoded as 0,1,2")
117 | })
118 |
119 | test_that('expGen2probGen fails when wrong input format is given', {
120 | expect_error(expGen2probGen(list(c(0,2,1))),
121 | "expGen2probGen takes a vector of ")
122 | })
123 |
124 | test_that('expGen2probGen returns triple NA', {
125 | expect_equal(expGen2probGen(c(NA)), c(NA,NA,NA))
126 | })
127 |
128 | test_that('expGen2probGen returns expected output', {
129 | expect_equal(expGen2probGen(c(2)), c(0,0,1))
130 | })
131 | test_that('probGen2expGen and expGen2probGen conversions work',{
132 | nrSamples <- 10
133 | genotype_prob <- rep(0, 3*nrSamples)
134 | genotype_prob[seq(1, nrSamples*3, 3) + sample(0:2, 10, replace=TRUE)] <- 1
135 | expect_equal(genotype_prob, expGen2probGen(probGen2expGen(genotype_prob)))
136 | })
137 |
138 |
--------------------------------------------------------------------------------
/man/getCausalSNPs.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/genotypeFunctions.R
3 | \name{getCausalSNPs}
4 | \alias{getCausalSNPs}
5 | \title{Draw random SNPs from genotypes.}
6 | \usage{
7 | getCausalSNPs(
8 | N,
9 | NrCausalSNPs = 20,
10 | genotypes = NULL,
11 | chr = NULL,
12 | NrSNPsOnChromosome = NULL,
13 | NrChrCausal = NULL,
14 | genoFilePrefix = NULL,
15 | genoFileSuffix = NULL,
16 | format = "delim",
17 | delimiter = ",",
18 | header = FALSE,
19 | skipFields = NULL,
20 | probabilities = FALSE,
21 | sampleID = "ID_",
22 | verbose = TRUE
23 | )
24 | }
25 | \arguments{
26 | \item{N}{Number [integer] of samples to simulate.}
27 |
28 | \item{NrCausalSNPs}{Number [integer] of SNPs to chose at random.}
29 |
30 | \item{genotypes}{[NrSamples x totalNrSNPs] Matrix of genotypes [integer]/
31 | [double].}
32 |
33 | \item{chr}{Vector of chromosome(s) [integer] to chose NrCausalSNPs from; only
34 | used when external genotype data is provided i.e. !is.null(genoFilePrefix).}
35 |
36 | \item{NrSNPsOnChromosome}{Vector of number(s) of SNPs [integer] per entry in
37 | chr (see above); has to be the same length as chr. If not provided, number of
38 | SNPS in file will be determined from line count (which can be slow for large
39 | files); (optional) header lines will be ignored, so accurate number of SNPs
40 | not lines in file should be specified.}
41 |
42 | \item{NrChrCausal}{Number [integer] of causal chromosomes to sample
43 | NrCausalSNPs from (as opposed to the actual chromosomes to chose from via chr
44 | ); only used when external genotype data is provided i.e.
45 | !is.null(genoFilePrefix).}
46 |
47 | \item{genoFilePrefix}{full path/to/chromosome-wise-genotype-file-ending-
48 | before-"chrChromosomeNumber" (no '~' expansion!) [string].}
49 |
50 | \item{genoFileSuffix}{[string] Following chromosome number including
51 | .fileformat (e.g. ".csv"); File described by genoFilePrefix-genoFileSuffix
52 | has to be a text format i.e. comma/tab/space separated.}
53 |
54 | \item{format}{Name [string] of genotype file format. Options are:
55 | "oxgen", "bimbam" or "delim". See \link{readStandardGenotypes} for details.}
56 |
57 | \item{delimiter}{Field separator [string] of genotypefile or
58 | genoFilePrefix-genoFileSuffix file if format == 'delim'.}
59 |
60 | \item{header}{[logical] Can be set to indicate if
61 | genoFilePrefix-genoFileSuffix file has a header for format == 'delim'.
62 | See details.}
63 |
64 | \item{skipFields}{Number [integer] of fields (columns) to skip in
65 | genoFilePrefix-genoFileSuffix file if format == 'delim'. See details.}
66 |
67 | \item{probabilities}{[boolean]. If set to TRUE, the genotypes in the files
68 | described by genoFilePrefix-genoFileSuffix are provided as triplets of
69 | probabilities (p(AA), p(Aa), p(aa)) and are converted into their expected
70 | genotype frequencies by 0*p(AA) + p(Aa) + 2p(aa) via \link{probGen2expGen}.}
71 |
72 | \item{sampleID}{Prefix [string] for naming samples (will be followed by
73 | sample number from 1 to N when constructing id_samples)}
74 |
75 | \item{verbose}{[boolean] If TRUE, progress info is printed to standard out}
76 | }
77 | \value{
78 | [N x NrCausalSNPs] Matrix of randomly drawn genotypes [integer]/
79 | [double]
80 | }
81 | \description{
82 | Draw random SNPs from genotypes provided or external genotype files.
83 | When drawing from external genotype files, only lines of randomly
84 | chosen SNPs are read, which is recommended for large genotype files. See
85 | details for more information. The latter option currently supports file in
86 | simple delim-formats (with specified delimiter and optional number of fields
87 | to skip) and the bimbam and the oxgen format.
88 | }
89 | \details{
90 | In order to chose SNPs from external genotype files without reading
91 | them into memory, genotypes for each chromosome need to be accesible as
92 | [SNPs x samples] in a separate file, containing "chrChromosomenumber" (e.g
93 | chr22) in the file name (e.g. /path/to/dir/related_nopopstructure_chr22.csv).
94 | All genotype files need to be saved in the same directory. genoFilePrefix
95 | (/path/to/dir/related_nopopstructure_) and genoFileSuffix (.csv) specify the
96 | strings leading and following the "chrChromosomenumber". If format== delim,
97 | the first column in each file needs to be the SNP_ID, the first row can
98 | either contain sample IDs or the first row of genotypes (specified with
99 | header). Subsequent columns containing additional SNP information can be
100 | skipped by setting skipFields. If format==oxgen or bimbam, files need to be
101 | in the oxgen or bimbam format (see \link{readStandardGenotypes} for details)
102 | and no additional information about delim, header or skipFields will be
103 | considered.
104 | getCausalSNPs generates a vector of chromosomes from which to sample the
105 | SNPs. For each of the chromosomes, it counts the number of SNPs in the
106 | chromosome file and creates vectors of random numbers ranging
107 | from 1:NrSNPSinFile. Only the lines corresponding to these numbers are then
108 | read into R. The example data provided for chromosome 22 contains genotypes
109 | (50 samples) of the first 500 SNPs on chromosome 22 with a minor allele
110 | frequency of greater than 2% from the European populations of the the 1000
111 | Genomes project.
112 | }
113 | \examples{
114 | # get causal SNPs from genotypes simulated within PhenotypeSimulator
115 | geno <- simulateGenotypes(N=10, NrSNP=10)
116 | causalSNPsFromSimulatedGenoStandardised <- getCausalSNPs(N=10,
117 | NrCausalSNPs=10, genotypes=geno$genotypes)
118 |
119 | # Get causal SNPs by sampling lines from large SNP files
120 | genotypeFile <- system.file("extdata/genotypes/",
121 | "genotypes_chr22.csv",
122 | package = "PhenotypeSimulator")
123 | genoFilePrefix <- gsub("chr.*", "", genotypeFile)
124 | genoFileSuffix <- ".csv"
125 | causalSNPsFromLines <- getCausalSNPs(N=50, NrCausalSNPs=10, chr=22,
126 | genoFilePrefix=genoFilePrefix,
127 | genoFileSuffix=genoFileSuffix)
128 | }
129 | \seealso{
130 | \code{\link{standardiseGenotypes}}
131 | }
132 |
--------------------------------------------------------------------------------
/man/noiseFixedEffects.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/variancecomponentFunctions.R
3 | \name{noiseFixedEffects}
4 | \alias{noiseFixedEffects}
5 | \title{Simulate noise fixed effects.}
6 | \usage{
7 | noiseFixedEffects(
8 | N,
9 | P,
10 | NrConfounders = 10,
11 | sampleID = "ID_",
12 | phenoID = "Trait_",
13 | id_samples = NULL,
14 | id_phenos = NULL,
15 | pTraitsAffected = 1,
16 | NrFixedEffects = 1,
17 | pIndependentConfounders = 0.4,
18 | pTraitIndependentConfounders = 0.2,
19 | keepSameIndependent = FALSE,
20 | distConfounders = "norm",
21 | mConfounders = 0,
22 | sdConfounders = 1,
23 | catConfounders = NULL,
24 | probConfounders = NULL,
25 | distBeta = "norm",
26 | mBeta = 0,
27 | sdBeta = 1,
28 | verbose = FALSE
29 | )
30 | }
31 | \arguments{
32 | \item{N}{Number [integer] of samples to simulate.}
33 |
34 | \item{P}{Number [integer] of phenotypes to simulate.}
35 |
36 | \item{NrConfounders}{Vector of number(s) [integer] of confounders from a
37 | specified distribution to simulate.}
38 |
39 | \item{sampleID}{Prefix [string] for naming samples.}
40 |
41 | \item{phenoID}{Prefix [string] for naming traits.}
42 |
43 | \item{id_samples}{Vector of [NrSamples] sample IDs [string]; if not provided
44 | constructed by paste(sampleID, 1:N, sep="").}
45 |
46 | \item{id_phenos}{Vector of [NrTraits] phenotype IDs [string]; if not provided
47 | constructed by paste(phenoID, 1:P, sep="").}
48 |
49 | \item{pTraitsAffected}{Vector of proportion(s) [double] of traits affected by
50 | the confounders. For non-integer results of pTraitsAffected*P, the ceiling of
51 | the result is used.}
52 |
53 | \item{NrFixedEffects}{Number [integer] of different confounder effects to
54 | simulate; allows to simulate fixed effects from different distributions or
55 | with different parameters; if only one type of confounder distribution is
56 | wanted, set NrFixedEffects=1 and choose the number of confounders with eg
57 | NrConfounders=10.}
58 |
59 | \item{pIndependentConfounders}{Vector of proportion(s) [double] of
60 | confounders to have a trait-independent effect.}
61 |
62 | \item{pTraitIndependentConfounders}{Vector of proportion(s) [double] of
63 | traits influenced by independent confounder effects.}
64 |
65 | \item{keepSameIndependent}{[boolean] If set to TRUE, the independent genetic
66 | effects always influence the same subset of traits.}
67 |
68 | \item{distConfounders}{Vector of name(s) [string] of distribution to use to
69 | simulate confounders; one of "unif", "norm", "bin", "cat_norm", "cat_unif".}
70 |
71 | \item{mConfounders}{Vector of mean/midpoint(s) [double] of normal/uniform
72 | distribution for confounders.}
73 |
74 | \item{sdConfounders}{Vector of standard deviation(s)/distance from
75 | midpoint(s) [double] of normal/uniform distribution for confounders.}
76 |
77 | \item{catConfounders}{Vector of number(s) of confounder categories [integer];
78 | required if distConfounders "cat_norm" or "cat_unif".}
79 |
80 | \item{probConfounders}{Vector of probability(s) [double] of binomial
81 | confounders (0/1); required if distConfounders "bin".}
82 |
83 | \item{distBeta}{Vector of name(s) [string] of distribution to use to simulate
84 | effect sizes of confounders; one of "unif" or "norm".}
85 |
86 | \item{mBeta}{Vector of mean/midpoint [double] of normal/uniform distribution
87 | for effect sizes of confounders.}
88 |
89 | \item{sdBeta}{Vector of standard deviation/distance from midpoint [double]
90 | of normal/uniform distribution for effect sizes of confounders.}
91 |
92 | \item{verbose}{[boolean] If TRUE, progress info is printed to standard out}
93 | }
94 | \value{
95 | Named list of shared confounder effects (shared: [N x P] matrix),
96 | independent confoudner effects (independent: [N x P] matrix),
97 | the confounders labeled as shared or independent effect
98 | (cov: [NrConfounders x N] matrix) and the simulated effect sizes of the
99 | confounders (cov_effect: [P x NrConfounders] dataframe).
100 | }
101 | \description{
102 | noiseFixedEffects simulates a number of non-genetic covariate effects
103 | (confounders). Confounders can have effects across all traits (shared)
104 | or to a number of traits only (independent); in addition, only a certain
105 | proportion of traits can be affected by the confounders.
106 | Confounders can be simulated as categorical variables or following a binomial
107 | , uniform or normal distribution. Effect sizes for the noise effects can be
108 | simulated from a uniform or normal distribution. Multiple confounder sets
109 | drawn from different distributions/different parameters of the same
110 | distribution can be simulated by specifying NrFixedEffects and supplying the
111 | respective distribution parameters.
112 | }
113 | \examples{
114 | # fixed noise effect with default setting
115 | noiseFE <- noiseFixedEffects(P=5, N=20)
116 |
117 | # 1 categorical fixed noise effect with uniform distribution of the
118 | # categories
119 | noiseFE_catUnif <- noiseFixedEffects(P=10, N=20, NrConfounders=1,
120 | distConfounders="cat_unif", catConfounders=3)
121 |
122 | # 10 fixed noise effect with uniform distribution between 1 and 5 (3 +/- 2)
123 | # categories
124 | noiseFE_uniformConfounders_normBetas <- noiseFixedEffects(P=10, N=20,
125 | NrConfounders=10, distConfounders="unif", mConfounders=3, sdConfounders=2,
126 | distBeta="norm", sdBeta=2)
127 |
128 | # 4 fixed noise effect with binomial distribution with p=0.2
129 | noiseFE_binomialConfounders_uniformBetas <- noiseFixedEffects(P=10, N=20,
130 | NrConfounders=4, distConfounders="bin", probConfounders=0.2, distBeta="norm",
131 | sdBeta=2)
132 |
133 | # 2 fixed noise effect with 1 binomial confounders and 1 normally
134 | # distributed confounder; the latter only affects 2 traits
135 | noiseFE_binomialandNormalConfounders <- noiseFixedEffects(P=10, N=20,
136 | NrFixedEffects=2, pTraitsAffected =c (1,0.2), NrConfounders=c(2,2),
137 | distConfounders=c("bin", "norm"), probConfounders=0.2)
138 | }
139 | \seealso{
140 | \code{\link{simulateDist}}
141 | }
142 |
--------------------------------------------------------------------------------
/NEWS.md:
--------------------------------------------------------------------------------
1 | # PhenotypeSimulator 0.3.4
2 | ## Minor changes
3 | 1. Fixed missing --genotypefile flag [issue 27](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/27)
4 | 2. Update vignettes with new location for impute files and commands to get the
5 | CEU samples [issue 24](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/24),
6 | thanks to @zfuller5280 for the suggestion!)
7 | 3. Standardise genotypes on row with major alleles [issue 21](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/21).
8 | Thank you for the detailed bug report by @alanw1!
9 | 4. Add option to imput missing genotypes to standardise genotype function;
10 | otherwise, if genotypes are missing, function will fail
11 | [issue 17](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/17)
12 | 5. Fix function description and passing of SNP IDs in readStandardGenotypes with
13 | delim option [issue 25](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/25),
14 | thanks @BSchmidt1.
15 |
16 | # PhenotypeSimulator 0.3.3
17 | ## Minor changes
18 | 1. Fixed bug that failed to return causal SNP name when only one SNP was chosen
19 | to be causal [issue
20 | 13](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/13).
21 |
22 | # PhenotypeSimulator 0.3.2
23 | ## Minor changes
24 | 1. Option for external, delimited genotype file to contain a header;
25 | additional checks to make sure the right data is received when
26 | sampling from the genotypes file [issue 10](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/10)
27 | 1. Fixed bug for reading external genotypes file [issue
28 | 9](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/9)
29 |
30 | # PhenotypeSimulator 0.3.1
31 | ## Minor changes
32 | 1. Adapted output file names for genotypes consistent with other
33 | filenames from genotypes.txt to Genotypes.txt
34 |
35 | # PhenotypeSimulator 0.3.0
36 | ## Major changes
37 | 1. Add option for non-linear transformation of simulated phenotypes: function (transformNonlinear)[https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/createphenotypeFunctions.R], accessible from [runSimulation](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/createphenotypeFunctions.R). Both transformed and original phenotypes are automatically returned with [savePheno](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R)
38 | 1. Replace parameter 'oxgen' in [readStandardGenotypes](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/genotypeFunctions.R) and [getCausalSNPs](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/genotypeFunctions.R) with 'format' - ensures proper
39 | specification of genotype format for all cases.
40 | ## Minor changes
41 | 1. In addition to full kinship, [savePheno](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R) and [writeStandardOutput](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R) write eigenvalues and eigenvalues of kinship matrix.
42 | 1. Output file names have been made more consistent in [savePheno](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R) and [writeStandardOutput](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R).
43 | 1. Causal SNPs are now also saved in specified standard output format.
44 | 1. LiMMBo has been added as output format in [savePheno](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R) and [writeStandardOutput](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R)([LiMMBo format](https://limmbo.readthedocs.io/en/latest/))
45 |
46 | # PhenotypeSimulator 0.2.2
47 | ## Minor changes
48 | 1. Update readStandardGenotypes to be compatible with latest
49 | release of data.table (v1.11.2), see
50 | [here](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/7)
51 |
52 | # PhenotypeSimulator 0.2.1
53 | ## Minor changes
54 | 1. Additional tests for compatibility of input parameters with variance
55 | components functions, genotype functions and output functions.
56 | 1. Bug fix in output function: savePheno now properly saves kinship matrix as
57 | .rds.
58 |
59 | # PhenotypeSimulator 0.2.0
60 | ## Major changes
61 | **Input**
62 | 1. *PhenotypeSimulator* now includes readStandardGenotypes which can read externally simulated or user-provided genotypes in plink, genome, oxgen (hapgen/impute2), bimbam or simple delimited format.
63 | 1. A user-specified correlation matrix can be provided for the simulation of the correlatedBdEffects.
64 | 1. Short option flags for command-line use of *PhenotypeSimulator* were removed.
65 |
66 | **Output**
67 | 1. *PhenotypeSimulator* provides the option to save the simulated phenotypes and genotypes in formats compatible with a number of commonly used genetic association software (gemma, bimbam, plink, snptest) via writeStandardOutput.
68 | 1. Intermediate phenotype components are now saved per default.
69 | 1. Saving additional subsets of the simulated data has been removed.
70 |
71 | **Variance components**
72 | 1. Genotype simulation and kinship estimation: functions for genotype simulation and kinship estimation have been rewritten for
73 | significant speed-ups of the computation time [benchmarking](https://github.com/HannahVMeyer/PhenotypeSimulator-profiling).
74 |
75 | 1. geneticFixedEffects and noiseFixedEffects:
76 | 1. The effect size distributions of the shared effects are now modelled as the product of two exponential distributions
77 | (to yield an approximately uniform distributions) or the product of a normal distribution with user-specified
78 | parameters and a standard normal distribution.
79 |
80 | 1. The independent effects can now be specified to affect the same subset or different subsets of traits (via
81 | keepSameIndependent).
82 |
83 | 1. The overall number of traits affected by the effects can now be specified via pTraitsAffected.
84 |
85 |
86 | 1. correlatedBgEffects: the additional correlation between the traits can be specified by the user by providing an external
87 | correlation matrix.
88 |
--------------------------------------------------------------------------------
/vignettes/sample-scripts-external-genotype-simulation.Rmd:
--------------------------------------------------------------------------------
1 | ---
2 | title: "Genotype simulation supported by *PhenotypeSimulator*"
3 | author: "Hannah Meyer"
4 | date: "`r Sys.Date()`"
5 | output: pdf_document
6 | bibliography: genotype-simulation.bib
7 | csl: plos-genetics.csl
8 | vignette: >
9 | %\VignetteIndexEntry{Supported Genotype Simulation}
10 | %\VignetteEngine{knitr::rmarkdown}
11 | %\VignetteEncoding{UTF-8}
12 | ---
13 | ```{r, echo = FALSE, message=FALSE}
14 | library(knitr)
15 | knitr::opts_chunk$set(collapse = TRUE, comment = "#>")
16 | ```
17 |
18 | There are a number of different strategies to generate genotype data for genetic
19 | association studies.
20 |
21 | 1. **Simple bi-allelic SNPs without structure**:
22 | In the most simple case and assuming bi-allelic SNPs, each SNP is simulated
23 | from a binomial distribution with two trials and probability equal to the
24 | given allele frequencies. This simple approach does not simulate any
25 | dependency between the genotypes as is observed with LD structure in the
26 | genome. Simple bi-allelic SNPs can be simulated with
27 | PhenotypeSimulator::simulateGenotypes or via PLINK @Chang2015. For large
28 | datasets (~ 1 million SNPs), simulation via PLINK is recommended.
29 |
30 | 1. **Coalescent approaches**:
31 | Coalescent methods simulate genealogical events backward in time. These
32 | events typically include the coalescence of two sequences into a single
33 | ancestral lineage, recombination within a sequence or migration between
34 | populations. A coalescent-based approach to simulate whole genome data is
35 | implemented in GENOME @Liang2007 and its output format is supported as an
36 | input format for *PhenotypeSimulator*.
37 |
38 | 1. **Forward-time approaches**:
39 | Forward-time simulation methods evolve a population forward in time, subject
40 | to arbitrary genetic and demographic factors @Peng2010. Several algorithms
41 | are available, many of which allow customisation by building the simulation
42 | scheme in R or python, hence the output format of the genotypes can be
43 | specified by the user. *PhenotypeSimulator* offers reading genotypes from
44 | delimited-files and as such, any genotypes generated by these programmes can
45 | be read (e.g. MetaSim @Strand2002 or simuPop @Peng2005).
46 |
47 | 1. **Resampling approaches**:
48 | Resampling-based approaches combine existing genotype data into the
49 | genotypes of the simulated samples, thereby retaining allele frequency and
50 | LD patterns @Wright2007. A standard resampling-based approach that uses
51 | common genotype formats (common for the
52 | [oxford genetics format](http://www.stats.ox.ac.uk/~marchini/software/gwas/file_format.html))
53 | is Hapgen2 and an example usage is given below.
54 |
55 | Examples for genotype simulation via PLINK, GENOME and Hapgen2 are given below.
56 | The corresponding data can be found in the extdata folder.
57 |
58 | ## Simple bi-allelic SNPs without structure via PLINK
59 | Download [PLINK](https://www.cog-genomics.org/plink/1.9/) and create a folder
60 | for the PLINK output files. The example below uses PLINK to simulate 1000 SNPs,
61 | with allele frequencies between 0 and 1 for 100 controls and 0 cases.
62 | PhenotypeSimulator::readStandardGenotypes reads the resulting .bim, .fam, .bed
63 | files.
64 |
65 | ```{bash, eval=FALSE}
66 | # Serves as output directory for simulation and parameter file
67 | mkdir -p ~/PhenotypeSimulator/inst/extdata/genotypes/plink
68 | cd ~/PhenotypeSimulator/inst/extdata/genotypes/plink
69 |
70 | # Write paramter file to simulate 1000 SNPs, with allele frequencies between 0
71 | # and 1 for 100 controls and 0 cases
72 | echo -e "1000\tSNP\t0.00\t1.00\t1.00\t1.00" > plink_sim_par.txt
73 |
74 | plink --simulate plink_sim_par.txt \
75 | --simulate-ncases 0 \
76 | --simulate-ncontrols 100 \
77 | --out genotypes_plink
78 | ```
79 |
80 |
81 |
82 | ## Coalescent simulation via Genome
83 | Download [GENOME](http://csg.sph.umich.edu/liang/genome/download.html) and
84 | create a folder for the GENOME output files. The example below uses GENOME to
85 | simulate genetic data for a population comprised of three sub-populations with
86 | 30, 30 and 40 samples. The simulated POPULATION PROFILE, SNP positions and the
87 | genotypes are all saved in genotypes_genome.txt.
88 | PhenotypeSimulator::readStandardGenotypes reads genotype information by parsing
89 | this output file and extracting the samples information following the line
90 | 'Samples:'
91 |
92 | ```{bash, eval=FALSE}
93 | mkdir -p ~/PhenotypeSimulator/inst/extdata/genotypes/genome
94 | cd ~/PhenotypeSimulator/inst/extdata/genotypes/genome
95 |
96 | # subpopulation with 30, 30 and 40 individuals each
97 | genome -pop 3 30 30 40 > genotypes_genome.txt
98 | ```
99 |
100 | ## Resampling-based simulation via Hapgen2
101 | Download [hapgen2](http://mathgen.stats.ox.ac.uk/genetics_software/hapgen/hapgen2.html),
102 | [hapgen example files](http://mathgen.stats.ox.ac.uk/genetics_software/hapgen/download/example/hapgen2.example.tgz)
103 | and the 1000Genomes data from the
104 | [impute2 webpage](https://mathgen.stats.ox.ac.uk/impute/impute_v1.html#Using_IMPUTE_with_the_HapMap_Data)
105 | as starting point for the resampling-based genotype simulation with hapgen2.
106 | For formating of the 1000Genomes data, have a look at this [vignette](https://hannahvmeyer.github.io/PhenotypeSimulator/articles/Simulation-and-LinearModel.html#genotype-simulation-via-hapgen2).
107 | The example files contain example haplotypes (ex.haps), legend files (ex.leg) map
108 | (ex.map) and TagSNP files (ex.tags). The following examples simulates genotype
109 | data for 100 controls and 0 cases with 1000 SNPs from chromosome 1.
110 | PhenotypeSimulator::readStandardGenotypes reads the resulting .gen and .samples
111 | files.
112 |
113 | ```{bash, eval=FALSE}
114 | # contains ex.leg file and serves as output directory
115 | mkdir -p ~/PhenotypeSimulator/inst/extdata/genotypes/hapgen
116 | cd ~/PhenotypeSimulator/inst/extdata/genotypes/hapgen
117 |
118 | # contains 1000Genomes haplotype data used for sampling
119 | hapdir=/path/to/CEU.0908.impute.files
120 |
121 | hapgen2 -m $hapdir/genetic_map_chr1_combined_b37.txt \
122 | -l ex.leg \
123 | -h $hapdir/chr1.ceu_subset.hap \
124 | -o genotypes_hapgen \
125 | -n 100 0 \
126 | -dl 45162 0 0 0 \
127 | -no_haps_output
128 | ```
129 |
130 | ## References
131 |
--------------------------------------------------------------------------------
/docs/LICENSE-text.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | License • PhenotypeSimulator
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 |
57 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
139 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 | License
148 |
149 |
150 | YEAR: 2017
151 | COPYRIGHT HOLDER: Hannah Verena Meyer
152 |
153 |
154 |
155 |
156 |
161 |
162 |
163 |
164 |
165 |
166 |
176 |
177 |
178 |
179 |
180 |
181 |
182 |
183 |
184 |
185 |
--------------------------------------------------------------------------------
/docs/404.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Page not found (404) • PhenotypeSimulator
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 |
57 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
139 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 | Page not found (404)
148 |
149 |
150 | Content not found. Please use links in the navbar.
151 |
152 |
153 |
154 |
159 |
160 |
161 |
162 |
163 |
164 |
174 |
175 |
176 |
177 |
178 |
179 |
180 |
181 |
182 |
183 |
--------------------------------------------------------------------------------
/docs/articles/index.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Articles • PhenotypeSimulator
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 |
57 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
139 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 | Articles
148 |
149 |
150 |
151 | All vignettes
152 |
153 |
154 |
164 |
165 |
166 |
167 |
168 |
169 |
179 |
180 |
181 |
182 |
183 |
184 |
185 |
186 |
187 |
188 |
--------------------------------------------------------------------------------
/docs/reference/PhenotypeSimulator.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | PhenotypeSimulator: A package for simulating phenotypes from different
10 | genetic and noise components — PhenotypeSimulator • PhenotypeSimulator
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 |
47 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
61 |
62 |
63 |
64 |
65 |
66 |
67 |
68 |
69 |
143 |
144 |
145 |
146 |
147 |
148 |
149 |
150 |
151 | PhenotypeSimulator: A package for simulating phenotypes from different
152 | genetic and noise components
153 | Source: R/PhenotypeSimulator.R
154 | PhenotypeSimulator.Rd
155 |
156 |
157 |
158 | PhenotypeSimulator: A package for simulating phenotypes from different
159 | genetic and noise components
160 |
161 |
162 |
163 |
164 |
165 |
166 |
171 |
172 |
173 |
174 |
184 |
185 |
186 |
187 |
188 |
189 |
190 |
191 |
192 |
193 |
--------------------------------------------------------------------------------
/docs/authors.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Citation and Authors • PhenotypeSimulator
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 |
57 |
58 |
59 |
60 |
61 |
62 |
63 |
195 |
196 |
197 |
198 |
199 |
200 |
201 |
202 |
203 |
--------------------------------------------------------------------------------
/docs/reference/addNonNulls.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Add all non-NULL elements of list. — addNonNulls • PhenotypeSimulator
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 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
66 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 |
148 | Add all non-NULL elements of list.
149 | Source: R/utilityFunctions.R
150 | addNonNulls.Rd
151 |
152 |
153 |
154 | Add all non-NULL elements of list.
155 |
156 |
157 | addNonNulls(compList)
158 |
159 | Arguments
160 |
161 |
163 | compList
164 | List of numeric matrices or data.frames of the equal
165 | dimensions.
167 |
168 |
169 | Value
170 |
171 | Matrix or data.frame containing sum of all list elements where
172 | is.null is FALSE.
173 |
174 |
175 |
180 |
181 |
182 |
183 |
193 |
194 |
195 |
196 |
197 |
198 |
199 |
200 |
201 |
202 |
--------------------------------------------------------------------------------
/docs/reference/read_lines.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Scan file for specific line numbers — read_lines • PhenotypeSimulator
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 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
66 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 |
152 |
153 |
154 | Scan file for specific line numbers
155 |
156 |
157 | read_lines(filename, lines, sep = "\n")
158 |
159 | Arguments
160 |
161 |
163 | filename
164 | /path/to/chromosomefile [string]
166 |
167 | lines
168 | vector of line numbers [integer] to be read
170 |
171 | sep
172 | [string] end-of-line delimiter
174 |
175 |
176 |
177 |
178 |
183 |
184 |
185 |
186 |
196 |
197 |
198 |
199 |
200 |
201 |
202 |
203 |
204 |
205 |
--------------------------------------------------------------------------------
/tests/testthat/test-createphenotypeFunctions.r:
--------------------------------------------------------------------------------
1 | context('Test createphenotype functions')
2 |
3 | noiseBg <- noiseBgEffects(N=100, P=10)
4 | noiseFixed <- noiseFixedEffects(N=100, P=10)
5 | noiseCorrelated <- correlatedBgEffects(N=100, P=10, pcorr=0.6)
6 | genotypes <- simulateGenotypes(N=100, NrSNP=20, verbose=FALSE)
7 | kinship <- getKinship(N=100, X=genotypes$genotypes, standardise=TRUE,
8 | verbose=FALSE)
9 | causalSNPs <- getCausalSNPs(N=100, genotypes=genotypes$genotypes, verbose=FALSE)
10 | geneticFixed <- geneticFixedEffects(X_causal=causalSNPs, P=10, N=100,
11 | verbose=FALSE)
12 | geneticBg <- geneticBgEffects(P=10, N=100, kinship=kinship)
13 |
14 | context("Tests rescaleVariance")
15 | test_that("rescaleVariance fails when propvar is not numeric", {
16 | expect_error(rescaleVariance(noiseBg$shared, propvar = 'a'),
17 | "propvar needs to be numeric")
18 | })
19 | test_that("rescaleVariance fails when propvar is out of range", {
20 | expect_error(rescaleVariance(noiseBg$shared, propvar = 1.1),
21 | "propvar cannot be less than 0 and or greater than 1")
22 | })
23 | test_that("rescaleVariance fails when component is not a matrix", {
24 | expect_error(rescaleVariance(data.frame(noiseBg$shared), propvar = 0.1),
25 | "component needs to be a matrix")
26 | })
27 | test_that("rescaleVariance returns NULL when no propvar is specified", {
28 | expect_equal(rescaleVariance(noiseBg$shared, propvar = 0), NULL)
29 | })
30 |
31 | test_that("rescaleVariance returns correclty scaled average column variance", {
32 | tmp <- rescaleVariance(noiseBg$shared, propvar = 0.1)
33 | expect_equal(mean(diag(var(tmp$component))), 0.1, tolerance=5e-5)
34 | })
35 |
36 |
37 | context("Tests setModel")
38 | test_that("setModel fails with input error", {
39 | expect_error(setModel(), "No variance components specified")
40 | })
41 |
42 | test_that("setModel fails with out of range error", {
43 | expect_error(setModel(genVar=0.4, h2s=1.2, theta=0.8, eta=0.8,
44 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
45 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
46 | pTraitIndependentConfounders=0.2,
47 | pIndependentGenetic=0.4,
48 | pTraitIndependentGenetic=0.2, verbose=FALSE),
49 | "Proportions have to be specified between 0 and 1:")
50 | })
51 | test_that("setModel fails with missing variance error", {
52 | expect_error(setModel(h2s=1, theta=0.8, eta=0.8,
53 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
54 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
55 | pTraitIndependentConfounders=0.2,
56 | pIndependentGenetic=0.4,
57 | pTraitIndependentGenetic=0.2, verbose=FALSE),
58 | "Neither genVar nor noiseVar are provided, thus")
59 | })
60 |
61 | test_that("setModel fails with sum of variance error", {
62 | expect_error(setModel(genVar=0.2, noiseVar=0.9, h2s=1, theta=0.8, eta=0.8,
63 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
64 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
65 | pTraitIndependentConfounders=0.2,
66 | pIndependentGenetic=0.4,
67 | pTraitIndependentGenetic=0.2, verbose=FALSE),
68 | "Sum of genetic variance and noise variance is greater than 1")
69 | })
70 |
71 | test_that("setModel fails with extra noise component error", {
72 | expect_error(setModel(noiseVar=0, h2s=1, theta=0.8, eta=0.8,
73 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
74 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
75 | pTraitIndependentConfounders=0.2,
76 | pIndependentGenetic=0.4,
77 | pTraitIndependentGenetic=0.2, verbose=FALSE),
78 | "The noise variance is set to 0 ")
79 | })
80 |
81 | test_that("setModel fails with extra genetic component error", {
82 | expect_error(setModel(genVar=0, h2s=1, theta=0.8, eta=0.8,
83 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
84 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
85 | pTraitIndependentConfounders=0.2,
86 | pIndependentGenetic=0.4,
87 | pTraitIndependentGenetic=0.2, verbose=FALSE),
88 | "The genetic variance is set to 0 ")
89 | })
90 |
91 | test_that("setModel fails with missing noise component error", {
92 | expect_error(setModel(noiseVar=0.8, h2s=1, theta=0.8, eta=0.8,
93 | pcorr=0.6, pIndependentConfounders=0.4,
94 | pTraitIndependentConfounders=0.2,
95 | pIndependentGenetic=0.4,
96 | pTraitIndependentGenetic=0.2,
97 | verbose=FALSE),
98 | "Neither delta ")
99 | })
100 |
101 | test_that("setModel fails with proportions larger than 1 error", {
102 | expect_error(setModel(noiseVar=0.8, h2s=1, theta=0.8, delta=0.8, phi=0.2,
103 | rho=0.2, eta=0.8,
104 | pcorr=0.6, pIndependentConfounders=0.4,
105 | pTraitIndependentConfounders=0.2,
106 | pIndependentGenetic=0.4,
107 | pTraitIndependentGenetic=0.2,
108 | verbose=FALSE),
109 | "Sum of the proportion of the variance of noise")
110 | })
111 |
112 | test_that("setModel fails with proportions less than 1 error", {
113 | expect_error(setModel(noiseVar=0.8, h2s=1, theta=0.8, delta=0.2, phi=0.2,
114 | rho=0.2, eta=0.8,
115 | pcorr=0.6, pIndependentConfounders=0.4,
116 | pTraitIndependentConfounders=0.2,
117 | pIndependentGenetic=0.4,
118 | pTraitIndependentGenetic=0.2,
119 | verbose=FALSE),
120 | "Sum of the proportion of the variance of noise")
121 | })
122 |
123 | test_that("setModel fails with insufficient noise component error", {
124 | expect_error(setModel(noiseVar=0.8, h2s=1, theta=0.8, eta=0.8,
125 | delta=0.8, pcorr=0.6, pIndependentConfounders=0.4,
126 | pTraitIndependentConfounders=0.2,
127 | pIndependentGenetic=0.4,
128 | pTraitIndependentGenetic=0.2,
129 | verbose=FALSE),
130 | "Not enough components provided to set proportions of")
131 | })
132 |
133 | test_that("setModel fails with insufficient genetic component error", {
134 | expect_error(setModel(genVar=0.8, theta=0.8, eta=0.8,
135 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
136 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
137 | pTraitIndependentConfounders=0.2,
138 | pIndependentGenetic=0.4,
139 | pTraitIndependentGenetic=0.2, verbose=FALSE),
140 | "Neither genetic variant effect variance")
141 | })
142 |
143 | test_that("setModel fails with sum of genetic component error", {
144 | expect_error(setModel(genVar=0.8, h2s=0.4, h2bg=0.3, theta=0.8, eta=0.8,
145 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
146 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
147 | pTraitIndependentConfounders=0.2,
148 | pIndependentGenetic=0.4,
149 | pTraitIndependentGenetic=0.2, verbose=FALSE),
150 | "Sum of the proportion of the variance of genetic")
151 | })
152 |
153 | context("Tests runSimulation")
154 | test_that("runSimulation returns correct output", {
155 | expect_equal(length(runSimulation(N=10, P=5, genVar=1, h2s=0.2)), 5)
156 | })
--------------------------------------------------------------------------------
`).
76 | getTopLevel: function($scope) {
77 | for (var i = 1; i <= 6; i++) {
78 | var $headings = this.findOrFilter($scope, 'h' + i);
79 | if ($headings.length > 1) {
80 | return i;
81 | }
82 | }
83 |
84 | return 1;
85 | },
86 |
87 | // returns the elements for the top level, and the next below it
88 | getHeadings: function($scope, topLevel) {
89 | var topSelector = 'h' + topLevel;
90 |
91 | var secondaryLevel = topLevel + 1;
92 | var secondarySelector = 'h' + secondaryLevel;
93 |
94 | return this.findOrFilter($scope, topSelector + ',' + secondarySelector);
95 | },
96 |
97 | getNavLevel: function(el) {
98 | return parseInt(el.tagName.charAt(1), 10);
99 | },
100 |
101 | populateNav: function($topContext, topLevel, $headings) {
102 | var $context = $topContext;
103 | var $prevNav;
104 |
105 | var helpers = this;
106 | $headings.each(function(i, el) {
107 | var $newNav = helpers.generateNavItem(el);
108 | var navLevel = helpers.getNavLevel(el);
109 |
110 | // determine the proper $context
111 | if (navLevel === topLevel) {
112 | // use top level
113 | $context = $topContext;
114 | } else if ($prevNav && $context === $topContext) {
115 | // create a new level of the tree and switch to it
116 | $context = helpers.createChildNavList($prevNav);
117 | } // else use the current $context
118 |
119 | $context.append($newNav);
120 |
121 | $prevNav = $newNav;
122 | });
123 | },
124 |
125 | parseOps: function(arg) {
126 | var opts;
127 | if (arg.jquery) {
128 | opts = {
129 | $nav: arg
130 | };
131 | } else {
132 | opts = arg;
133 | }
134 | opts.$scope = opts.$scope || $(document.body);
135 | return opts;
136 | }
137 | },
138 |
139 | // accepts a jQuery object, or an options object
140 | init: function(opts) {
141 | opts = this.helpers.parseOps(opts);
142 |
143 | // ensure that the data attribute is in place for styling
144 | opts.$nav.attr('data-toggle', 'toc');
145 |
146 | var $topContext = this.helpers.createChildNavList(opts.$nav);
147 | var topLevel = this.helpers.getTopLevel(opts.$scope);
148 | var $headings = this.helpers.getHeadings(opts.$scope, topLevel);
149 | this.helpers.populateNav($topContext, topLevel, $headings);
150 | }
151 | };
152 |
153 | $(function() {
154 | $('nav[data-toggle="toc"]').each(function(i, el) {
155 | var $nav = $(el);
156 | Toc.init($nav);
157 | });
158 | });
159 | })();
160 |
--------------------------------------------------------------------------------
/vignettes/SimulationAndLinearModel.bib:
--------------------------------------------------------------------------------
1 | @article{Zhou2014,
2 | abstract = {Multivariate linear mixed models (mvLMMs) are powerful tools for testing associations between single-nucleotide polymorphisms and multiple correlated phenotypes while controlling for population stratification in genome-wide association studies. We present efficient algorithms in the genome-wide efficient mixed model association (GEMMA) software for fitting mvLMMs and computing likelihood ratio tests. These algorithms offer improved computation speed, power and P-value calibration over existing methods, and can deal with more than two phenotypes.},
3 | author = {Zhou, Xiang and Stephens, Matthew},
4 | doi = {10.1038/nmeth.2848},
5 | file = {:Users/hannah/Documents/Mendeley Desktop/2014{\_}Nature Methods{\_}Zhou, Stephens(3).pdf:pdf;:Users/hannah/Documents/Mendeley Desktop/2014{\_}Nature Methods{\_}Zhou, Stephens(4).pdf:pdf},
6 | isbn = {1548-7105 (Electronic)$\backslash$r1548-7091 (Linking)},
7 | issn = {1548-7105},
8 | journal = {Nature Methods},
9 | keywords = {Algorithms,Genome,Likelihood Functions,Linear Models,Multivariate Analysis,Polymorphism,Single Nucleotide,Single Nucleotide: genetics,Software},
10 | number = {4},
11 | pages = {407--9},
12 | pmid = {24531419},
13 | title = {{Efficient multivariate linear mixed model algorithms for genome-wide association studies.}},
14 | url = {http://www.ncbi.nlm.nih.gov/pubmed/24531419},
15 | volume = {11},
16 | year = {2014}
17 | }
18 | @article{Chang2015,
19 | abstract = {PLINK 1 is a widely used open-source C/C++ toolset for genome-wide association studies (GWAS) and research in population genetics. However, the steady accumulation of data from imputation and whole-genome sequencing studies has exposed a strong need for even faster and more scalable implementations of key functions. In addition, GWAS and population-genetic data now frequently contain probabilistic calls, phase information, and/or multiallelic variants, none of which can be represented by PLINK 1's primary data format. To address these issues, we are developing a second-generation codebase for PLINK. The first major release from this codebase, PLINK 1.9, introduces extensive use of bit-level parallelism, O(sqrt(n))-time/constant-space Hardy-Weinberg equilibrium and Fisher's exact tests, and many other algorithmic improvements. In combination, these changes accelerate most operations by 1-4 orders of magnitude, and allow the program to handle datasets too large to fit in RAM. This will be followed by PLINK 2.0, which will introduce (a) a new data format capable of efficiently representing probabilities, phase, and multiallelic variants, and (b) extensions of many functions to account for the new types of information. The second-generation versions of PLINK will offer dramatic improvements in performance and compatibility. For the first time, users without access to high-end computing resources can perform several essential analyses of the feature-rich and very large genetic datasets coming into use.},
20 | author = {Chang, Christopher C and Chow, Carson C and Tellier, Laurent CAM and Vattikuti, Shashaank and Purcell, Shaun M and Lee, James J},
21 | doi = {10.1186/s13742-015-0047-8},
22 | issn = {2047-217X},
23 | journal = {GigaScience},
24 | month = {feb},
25 | number = {1},
26 | pages = {7},
27 | title = {{Second-generation PLINK: rising to the challenge of larger and richer datasets}},
28 | url = {http://arxiv.org/abs/1410.4803},
29 | volume = {4},
30 | year = {2015}
31 | }
32 | @article{Halsey2015,
33 | author = {Halsey, Lewis G and Curran-Everett, Douglas and Vowler, Sarah L and Drummond, Gordon B},
34 | doi = {10.1038/nmeth.3288},
35 | issn = {1548-7091},
36 | journal = {Nature Methods},
37 | month = {feb},
38 | number = {3},
39 | pages = {179--185},
40 | pmid = {25719825},
41 | title = {{The fickle P value generates irreproducible results}},
42 | url = {http://www.ncbi.nlm.nih.gov/pubmed/25719825 http://www.nature.com/doifinder/10.1038/nmeth.3288},
43 | volume = {12},
44 | year = {2015}
45 | }
46 | @article{Cohen1992,
47 | abstract = {One possible reason for the continued neglect of statistical power analysis in research in the behavioral sciences is the inaccessibility of or difficulty with the standard material. A convenient, although not comprehensive, presentation of required sample sizes is provided here. Effect-size indexes and conventional values for these are given for operationally defined small, medium, and large effects. The sample sizes necessary for .80 power to detect effects at these levels are tabled for eight standard statistical tests: (a) the difference between independent means, (b) the significance of a product-moment correlation, (c) the difference between independent rs, (d) the sign test, (e) the difference between independent proportions, (f) chi-square tests for goodness of fit and contingency tables, (g) one-way analysis of variance, and (h) the significance of a multiple or multiple partial correlation.},
48 | author = {Cohen, J},
49 | issn = {0033-2909},
50 | journal = {Psychological Bulletin},
51 | month = {jul},
52 | number = {1},
53 | pages = {155--9},
54 | pmid = {19565683},
55 | title = {{A power primer.}},
56 | url = {http://www.ncbi.nlm.nih.gov/pubmed/19565683},
57 | volume = {112},
58 | year = {1992}
59 | }
60 | @article{Su2011,
61 | author = {Su, Zhan and Marchini, Jonathan and Donnelly, Peter},
62 | doi = {10.1093/bioinformatics/btr341},
63 | file = {:Users/hannah/Documents/Mendeley Desktop/2011{\_}Bioinformatics{\_}Su, Marchini, Donnelly.pdf:pdf},
64 | issn = {1460-2059},
65 | journal = {Bioinformatics},
66 | month = {aug},
67 | number = {16},
68 | pages = {2304--2305},
69 | publisher = {Oxford University Press},
70 | title = {{HAPGEN2: simulation of multiple disease SNPs}},
71 | url = {https://academic.oup.com/bioinformatics/article-lookup/doi/10.1093/bioinformatics/btr341},
72 | volume = {27},
73 | year = {2011}
74 | }
75 |
--------------------------------------------------------------------------------
/tests/testthat/test-utility.r:
--------------------------------------------------------------------------------
1 | context('Test utility functions')
2 |
3 | test_that("vmessage fails when non-character specified as separator", {
4 | expect_error(vmessage(c("hello", "world"), sep=9))
5 | })
6 |
7 | test_that("vmessage returns correct messsage", {
8 | expect_that(vmessage(c("Hello", "World"), sep="Little"),
9 | shows_message("HelloLittleWorld"))
10 | })
11 |
12 | test_that('vmessage returns NULL when verbose set to FALSE', {
13 | expect_equal(vmessage("Hello world", verbose=FALSE), NULL)
14 | })
15 |
16 | test_that('commaList2vector returns vector from comma-separated inputlist', {
17 | expect_equal(commaList2vector("1,2,3"), c(1,2,3))
18 | })
19 |
20 | test_that('commaList2vector returns NULL', {
21 | expect_equal(commaList2vector(), NULL)
22 | })
23 |
24 | test_that('commaList2vector fails with unknown type', {
25 | expect_error(commaList2vector(type='matrix', commastring="1,2,2,3,4"),
26 | "Unknown type of comma-separated list elements")
27 | })
28 | test_that('addNonNulls fails when dimensions of list elements are not the same',
29 | {
30 | expect_error(addNonNulls(list(matrix(1:10, ncol=2), matrix(11:20, ncol=2),
31 | matrix(21:30, ncol=5), NULL, NULL)),
32 | "Column dimensions of list elements are different")
33 | expect_error(addNonNulls(list(matrix(1:10, ncol=2), matrix(11:20, ncol=2),
34 | matrix(21:40, ncol=2), NULL, NULL)),
35 | "Row dimensions of list elements are different")
36 | })
37 |
38 | test_that('addNonNulls fails when non-list supplied as input', {
39 | expect_error(addNonNulls(matrix(1:10, ncol=2)),
40 | "addNonNulls expects input of type list")
41 | })
42 |
43 | test_that('simulateDist fails when distr provided is not one of "unif", "norm",
44 | "bin", "cat_norm" or "cat_unif" ', {
45 | expect_error(simulateDist(x=10, dist="gamma"),"Unknown distribution")
46 | })
47 |
48 | test_that('simulateDist fails when number of observations is less than 0', {
49 | expect_error(simulateDist(x=-10, dist="bin"),
50 | "The number of observations to simulate")
51 | })
52 |
53 | test_that('simulateDist fails when standard deviation for normal distribution
54 | less than 0', {
55 | expect_error(simulateDist(x=10, dist="norm", std=-0.3),
56 | "Simulating normal distribution")
57 | })
58 |
59 | test_that('simulateDist fails when probability for binomial dist is not
60 | provided', {
61 | expect_error(simulateDist(x=10, dist="bin"),
62 | "Simulating binomial distribution")
63 | })
64 |
65 | test_that('simulateDist fails when probability for binomial dist is less
66 | than zero', {
67 | expect_error(simulateDist(x=10, dist="bin", prob=-0.02),
68 | "Simulating binomial distribution")
69 | })
70 |
71 | test_that('simulateDist fails with unknown distribution', {
72 | expect_error(simulateDist(x=10, dist="beta"), "Unknown distribution")
73 | })
74 |
75 | test_that('simulateDist fails with length of distribution argument', {
76 | dist=c("unif", "norm", "bin", "cat_norm", "cat_unif")
77 | expect_error(simulateDist(x=10),
78 | paste("Please specify exactly one distribution to sample from,",
79 | "currently ", length(dist), " provided.", sep=""))
80 | })
81 |
82 | test_that('simulateDist fails when number of categories for categorical dist
83 | is not provided', {
84 | expect_error(simulateDist(x=10, dist="cat_unif"),
85 | "Simulating categorical distribution")
86 | })
87 |
88 | test_that('simulateDist fails when number of categories for categorical dist
89 | is less than zero', {
90 | expect_error(simulateDist(x=10, dist="cat_unif", categories=-2),
91 | "Simulating categorical distribution")
92 | })
93 |
94 | test_that("probGen2expGen fails when input is not a numeric vector", {
95 | expect_error( probGen2expGen(list(c(0.1, 0.9, 0))),
96 | "probGen2expGen takes")
97 | })
98 |
99 | test_that("probGen2expGen fails when input vector is not a multiple of three", {
100 | expect_error( probGen2expGen(c(0.1, 0.9)),
101 | "Length of genotype probabilty vector")
102 | })
103 |
104 | test_that("probGen2expGen fails with NA", {
105 | expect_error( probGen2expGen(c(0.1, 0.8, NA)),
106 | "Samples need to be fully genotyped")
107 | })
108 |
109 | test_that("probGen2expGen fails with incorrect probabilities", {
110 | expect_error( probGen2expGen(c(0.1, 0.8, 0)),
111 | "Genotype probabilities do not sum to 1")
112 | })
113 |
114 | test_that('expGen2probGen fails when wrong encoding is given', {
115 | expect_error(expGen2probGen(c(0,2,3)),
116 | "Genotypes can only be encoded as 0,1,2")
117 | })
118 |
119 | test_that('expGen2probGen fails when wrong input format is given', {
120 | expect_error(expGen2probGen(list(c(0,2,1))),
121 | "expGen2probGen takes a vector of ")
122 | })
123 |
124 | test_that('expGen2probGen returns triple NA', {
125 | expect_equal(expGen2probGen(c(NA)), c(NA,NA,NA))
126 | })
127 |
128 | test_that('expGen2probGen returns expected output', {
129 | expect_equal(expGen2probGen(c(2)), c(0,0,1))
130 | })
131 | test_that('probGen2expGen and expGen2probGen conversions work',{
132 | nrSamples <- 10
133 | genotype_prob <- rep(0, 3*nrSamples)
134 | genotype_prob[seq(1, nrSamples*3, 3) + sample(0:2, 10, replace=TRUE)] <- 1
135 | expect_equal(genotype_prob, expGen2probGen(probGen2expGen(genotype_prob)))
136 | })
137 |
138 |
--------------------------------------------------------------------------------
/man/getCausalSNPs.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/genotypeFunctions.R
3 | \name{getCausalSNPs}
4 | \alias{getCausalSNPs}
5 | \title{Draw random SNPs from genotypes.}
6 | \usage{
7 | getCausalSNPs(
8 | N,
9 | NrCausalSNPs = 20,
10 | genotypes = NULL,
11 | chr = NULL,
12 | NrSNPsOnChromosome = NULL,
13 | NrChrCausal = NULL,
14 | genoFilePrefix = NULL,
15 | genoFileSuffix = NULL,
16 | format = "delim",
17 | delimiter = ",",
18 | header = FALSE,
19 | skipFields = NULL,
20 | probabilities = FALSE,
21 | sampleID = "ID_",
22 | verbose = TRUE
23 | )
24 | }
25 | \arguments{
26 | \item{N}{Number [integer] of samples to simulate.}
27 |
28 | \item{NrCausalSNPs}{Number [integer] of SNPs to chose at random.}
29 |
30 | \item{genotypes}{[NrSamples x totalNrSNPs] Matrix of genotypes [integer]/
31 | [double].}
32 |
33 | \item{chr}{Vector of chromosome(s) [integer] to chose NrCausalSNPs from; only
34 | used when external genotype data is provided i.e. !is.null(genoFilePrefix).}
35 |
36 | \item{NrSNPsOnChromosome}{Vector of number(s) of SNPs [integer] per entry in
37 | chr (see above); has to be the same length as chr. If not provided, number of
38 | SNPS in file will be determined from line count (which can be slow for large
39 | files); (optional) header lines will be ignored, so accurate number of SNPs
40 | not lines in file should be specified.}
41 |
42 | \item{NrChrCausal}{Number [integer] of causal chromosomes to sample
43 | NrCausalSNPs from (as opposed to the actual chromosomes to chose from via chr
44 | ); only used when external genotype data is provided i.e.
45 | !is.null(genoFilePrefix).}
46 |
47 | \item{genoFilePrefix}{full path/to/chromosome-wise-genotype-file-ending-
48 | before-"chrChromosomeNumber" (no '~' expansion!) [string].}
49 |
50 | \item{genoFileSuffix}{[string] Following chromosome number including
51 | .fileformat (e.g. ".csv"); File described by genoFilePrefix-genoFileSuffix
52 | has to be a text format i.e. comma/tab/space separated.}
53 |
54 | \item{format}{Name [string] of genotype file format. Options are:
55 | "oxgen", "bimbam" or "delim". See \link{readStandardGenotypes} for details.}
56 |
57 | \item{delimiter}{Field separator [string] of genotypefile or
58 | genoFilePrefix-genoFileSuffix file if format == 'delim'.}
59 |
60 | \item{header}{[logical] Can be set to indicate if
61 | genoFilePrefix-genoFileSuffix file has a header for format == 'delim'.
62 | See details.}
63 |
64 | \item{skipFields}{Number [integer] of fields (columns) to skip in
65 | genoFilePrefix-genoFileSuffix file if format == 'delim'. See details.}
66 |
67 | \item{probabilities}{[boolean]. If set to TRUE, the genotypes in the files
68 | described by genoFilePrefix-genoFileSuffix are provided as triplets of
69 | probabilities (p(AA), p(Aa), p(aa)) and are converted into their expected
70 | genotype frequencies by 0*p(AA) + p(Aa) + 2p(aa) via \link{probGen2expGen}.}
71 |
72 | \item{sampleID}{Prefix [string] for naming samples (will be followed by
73 | sample number from 1 to N when constructing id_samples)}
74 |
75 | \item{verbose}{[boolean] If TRUE, progress info is printed to standard out}
76 | }
77 | \value{
78 | [N x NrCausalSNPs] Matrix of randomly drawn genotypes [integer]/
79 | [double]
80 | }
81 | \description{
82 | Draw random SNPs from genotypes provided or external genotype files.
83 | When drawing from external genotype files, only lines of randomly
84 | chosen SNPs are read, which is recommended for large genotype files. See
85 | details for more information. The latter option currently supports file in
86 | simple delim-formats (with specified delimiter and optional number of fields
87 | to skip) and the bimbam and the oxgen format.
88 | }
89 | \details{
90 | In order to chose SNPs from external genotype files without reading
91 | them into memory, genotypes for each chromosome need to be accesible as
92 | [SNPs x samples] in a separate file, containing "chrChromosomenumber" (e.g
93 | chr22) in the file name (e.g. /path/to/dir/related_nopopstructure_chr22.csv).
94 | All genotype files need to be saved in the same directory. genoFilePrefix
95 | (/path/to/dir/related_nopopstructure_) and genoFileSuffix (.csv) specify the
96 | strings leading and following the "chrChromosomenumber". If format== delim,
97 | the first column in each file needs to be the SNP_ID, the first row can
98 | either contain sample IDs or the first row of genotypes (specified with
99 | header). Subsequent columns containing additional SNP information can be
100 | skipped by setting skipFields. If format==oxgen or bimbam, files need to be
101 | in the oxgen or bimbam format (see \link{readStandardGenotypes} for details)
102 | and no additional information about delim, header or skipFields will be
103 | considered.
104 | getCausalSNPs generates a vector of chromosomes from which to sample the
105 | SNPs. For each of the chromosomes, it counts the number of SNPs in the
106 | chromosome file and creates vectors of random numbers ranging
107 | from 1:NrSNPSinFile. Only the lines corresponding to these numbers are then
108 | read into R. The example data provided for chromosome 22 contains genotypes
109 | (50 samples) of the first 500 SNPs on chromosome 22 with a minor allele
110 | frequency of greater than 2% from the European populations of the the 1000
111 | Genomes project.
112 | }
113 | \examples{
114 | # get causal SNPs from genotypes simulated within PhenotypeSimulator
115 | geno <- simulateGenotypes(N=10, NrSNP=10)
116 | causalSNPsFromSimulatedGenoStandardised <- getCausalSNPs(N=10,
117 | NrCausalSNPs=10, genotypes=geno$genotypes)
118 |
119 | # Get causal SNPs by sampling lines from large SNP files
120 | genotypeFile <- system.file("extdata/genotypes/",
121 | "genotypes_chr22.csv",
122 | package = "PhenotypeSimulator")
123 | genoFilePrefix <- gsub("chr.*", "", genotypeFile)
124 | genoFileSuffix <- ".csv"
125 | causalSNPsFromLines <- getCausalSNPs(N=50, NrCausalSNPs=10, chr=22,
126 | genoFilePrefix=genoFilePrefix,
127 | genoFileSuffix=genoFileSuffix)
128 | }
129 | \seealso{
130 | \code{\link{standardiseGenotypes}}
131 | }
132 |
--------------------------------------------------------------------------------
/man/noiseFixedEffects.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/variancecomponentFunctions.R
3 | \name{noiseFixedEffects}
4 | \alias{noiseFixedEffects}
5 | \title{Simulate noise fixed effects.}
6 | \usage{
7 | noiseFixedEffects(
8 | N,
9 | P,
10 | NrConfounders = 10,
11 | sampleID = "ID_",
12 | phenoID = "Trait_",
13 | id_samples = NULL,
14 | id_phenos = NULL,
15 | pTraitsAffected = 1,
16 | NrFixedEffects = 1,
17 | pIndependentConfounders = 0.4,
18 | pTraitIndependentConfounders = 0.2,
19 | keepSameIndependent = FALSE,
20 | distConfounders = "norm",
21 | mConfounders = 0,
22 | sdConfounders = 1,
23 | catConfounders = NULL,
24 | probConfounders = NULL,
25 | distBeta = "norm",
26 | mBeta = 0,
27 | sdBeta = 1,
28 | verbose = FALSE
29 | )
30 | }
31 | \arguments{
32 | \item{N}{Number [integer] of samples to simulate.}
33 |
34 | \item{P}{Number [integer] of phenotypes to simulate.}
35 |
36 | \item{NrConfounders}{Vector of number(s) [integer] of confounders from a
37 | specified distribution to simulate.}
38 |
39 | \item{sampleID}{Prefix [string] for naming samples.}
40 |
41 | \item{phenoID}{Prefix [string] for naming traits.}
42 |
43 | \item{id_samples}{Vector of [NrSamples] sample IDs [string]; if not provided
44 | constructed by paste(sampleID, 1:N, sep="").}
45 |
46 | \item{id_phenos}{Vector of [NrTraits] phenotype IDs [string]; if not provided
47 | constructed by paste(phenoID, 1:P, sep="").}
48 |
49 | \item{pTraitsAffected}{Vector of proportion(s) [double] of traits affected by
50 | the confounders. For non-integer results of pTraitsAffected*P, the ceiling of
51 | the result is used.}
52 |
53 | \item{NrFixedEffects}{Number [integer] of different confounder effects to
54 | simulate; allows to simulate fixed effects from different distributions or
55 | with different parameters; if only one type of confounder distribution is
56 | wanted, set NrFixedEffects=1 and choose the number of confounders with eg
57 | NrConfounders=10.}
58 |
59 | \item{pIndependentConfounders}{Vector of proportion(s) [double] of
60 | confounders to have a trait-independent effect.}
61 |
62 | \item{pTraitIndependentConfounders}{Vector of proportion(s) [double] of
63 | traits influenced by independent confounder effects.}
64 |
65 | \item{keepSameIndependent}{[boolean] If set to TRUE, the independent genetic
66 | effects always influence the same subset of traits.}
67 |
68 | \item{distConfounders}{Vector of name(s) [string] of distribution to use to
69 | simulate confounders; one of "unif", "norm", "bin", "cat_norm", "cat_unif".}
70 |
71 | \item{mConfounders}{Vector of mean/midpoint(s) [double] of normal/uniform
72 | distribution for confounders.}
73 |
74 | \item{sdConfounders}{Vector of standard deviation(s)/distance from
75 | midpoint(s) [double] of normal/uniform distribution for confounders.}
76 |
77 | \item{catConfounders}{Vector of number(s) of confounder categories [integer];
78 | required if distConfounders "cat_norm" or "cat_unif".}
79 |
80 | \item{probConfounders}{Vector of probability(s) [double] of binomial
81 | confounders (0/1); required if distConfounders "bin".}
82 |
83 | \item{distBeta}{Vector of name(s) [string] of distribution to use to simulate
84 | effect sizes of confounders; one of "unif" or "norm".}
85 |
86 | \item{mBeta}{Vector of mean/midpoint [double] of normal/uniform distribution
87 | for effect sizes of confounders.}
88 |
89 | \item{sdBeta}{Vector of standard deviation/distance from midpoint [double]
90 | of normal/uniform distribution for effect sizes of confounders.}
91 |
92 | \item{verbose}{[boolean] If TRUE, progress info is printed to standard out}
93 | }
94 | \value{
95 | Named list of shared confounder effects (shared: [N x P] matrix),
96 | independent confoudner effects (independent: [N x P] matrix),
97 | the confounders labeled as shared or independent effect
98 | (cov: [NrConfounders x N] matrix) and the simulated effect sizes of the
99 | confounders (cov_effect: [P x NrConfounders] dataframe).
100 | }
101 | \description{
102 | noiseFixedEffects simulates a number of non-genetic covariate effects
103 | (confounders). Confounders can have effects across all traits (shared)
104 | or to a number of traits only (independent); in addition, only a certain
105 | proportion of traits can be affected by the confounders.
106 | Confounders can be simulated as categorical variables or following a binomial
107 | , uniform or normal distribution. Effect sizes for the noise effects can be
108 | simulated from a uniform or normal distribution. Multiple confounder sets
109 | drawn from different distributions/different parameters of the same
110 | distribution can be simulated by specifying NrFixedEffects and supplying the
111 | respective distribution parameters.
112 | }
113 | \examples{
114 | # fixed noise effect with default setting
115 | noiseFE <- noiseFixedEffects(P=5, N=20)
116 |
117 | # 1 categorical fixed noise effect with uniform distribution of the
118 | # categories
119 | noiseFE_catUnif <- noiseFixedEffects(P=10, N=20, NrConfounders=1,
120 | distConfounders="cat_unif", catConfounders=3)
121 |
122 | # 10 fixed noise effect with uniform distribution between 1 and 5 (3 +/- 2)
123 | # categories
124 | noiseFE_uniformConfounders_normBetas <- noiseFixedEffects(P=10, N=20,
125 | NrConfounders=10, distConfounders="unif", mConfounders=3, sdConfounders=2,
126 | distBeta="norm", sdBeta=2)
127 |
128 | # 4 fixed noise effect with binomial distribution with p=0.2
129 | noiseFE_binomialConfounders_uniformBetas <- noiseFixedEffects(P=10, N=20,
130 | NrConfounders=4, distConfounders="bin", probConfounders=0.2, distBeta="norm",
131 | sdBeta=2)
132 |
133 | # 2 fixed noise effect with 1 binomial confounders and 1 normally
134 | # distributed confounder; the latter only affects 2 traits
135 | noiseFE_binomialandNormalConfounders <- noiseFixedEffects(P=10, N=20,
136 | NrFixedEffects=2, pTraitsAffected =c (1,0.2), NrConfounders=c(2,2),
137 | distConfounders=c("bin", "norm"), probConfounders=0.2)
138 | }
139 | \seealso{
140 | \code{\link{simulateDist}}
141 | }
142 |
--------------------------------------------------------------------------------
/NEWS.md:
--------------------------------------------------------------------------------
1 | # PhenotypeSimulator 0.3.4
2 | ## Minor changes
3 | 1. Fixed missing --genotypefile flag [issue 27](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/27)
4 | 2. Update vignettes with new location for impute files and commands to get the
5 | CEU samples [issue 24](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/24),
6 | thanks to @zfuller5280 for the suggestion!)
7 | 3. Standardise genotypes on row with major alleles [issue 21](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/21).
8 | Thank you for the detailed bug report by @alanw1!
9 | 4. Add option to imput missing genotypes to standardise genotype function;
10 | otherwise, if genotypes are missing, function will fail
11 | [issue 17](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/17)
12 | 5. Fix function description and passing of SNP IDs in readStandardGenotypes with
13 | delim option [issue 25](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/25),
14 | thanks @BSchmidt1.
15 |
16 | # PhenotypeSimulator 0.3.3
17 | ## Minor changes
18 | 1. Fixed bug that failed to return causal SNP name when only one SNP was chosen
19 | to be causal [issue
20 | 13](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/13).
21 |
22 | # PhenotypeSimulator 0.3.2
23 | ## Minor changes
24 | 1. Option for external, delimited genotype file to contain a header;
25 | additional checks to make sure the right data is received when
26 | sampling from the genotypes file [issue 10](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/10)
27 | 1. Fixed bug for reading external genotypes file [issue
28 | 9](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/9)
29 |
30 | # PhenotypeSimulator 0.3.1
31 | ## Minor changes
32 | 1. Adapted output file names for genotypes consistent with other
33 | filenames from genotypes.txt to Genotypes.txt
34 |
35 | # PhenotypeSimulator 0.3.0
36 | ## Major changes
37 | 1. Add option for non-linear transformation of simulated phenotypes: function (transformNonlinear)[https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/createphenotypeFunctions.R], accessible from [runSimulation](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/createphenotypeFunctions.R). Both transformed and original phenotypes are automatically returned with [savePheno](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R)
38 | 1. Replace parameter 'oxgen' in [readStandardGenotypes](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/genotypeFunctions.R) and [getCausalSNPs](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/genotypeFunctions.R) with 'format' - ensures proper
39 | specification of genotype format for all cases.
40 | ## Minor changes
41 | 1. In addition to full kinship, [savePheno](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R) and [writeStandardOutput](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R) write eigenvalues and eigenvalues of kinship matrix.
42 | 1. Output file names have been made more consistent in [savePheno](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R) and [writeStandardOutput](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R).
43 | 1. Causal SNPs are now also saved in specified standard output format.
44 | 1. LiMMBo has been added as output format in [savePheno](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R) and [writeStandardOutput](https://github.com/HannahVMeyer/PhenotypeSimulator/blob/master/R/outputFunctions.R)([LiMMBo format](https://limmbo.readthedocs.io/en/latest/))
45 |
46 | # PhenotypeSimulator 0.2.2
47 | ## Minor changes
48 | 1. Update readStandardGenotypes to be compatible with latest
49 | release of data.table (v1.11.2), see
50 | [here](https://github.com/HannahVMeyer/PhenotypeSimulator/issues/7)
51 |
52 | # PhenotypeSimulator 0.2.1
53 | ## Minor changes
54 | 1. Additional tests for compatibility of input parameters with variance
55 | components functions, genotype functions and output functions.
56 | 1. Bug fix in output function: savePheno now properly saves kinship matrix as
57 | .rds.
58 |
59 | # PhenotypeSimulator 0.2.0
60 | ## Major changes
61 | **Input**
62 | 1. *PhenotypeSimulator* now includes readStandardGenotypes which can read externally simulated or user-provided genotypes in plink, genome, oxgen (hapgen/impute2), bimbam or simple delimited format.
63 | 1. A user-specified correlation matrix can be provided for the simulation of the correlatedBdEffects.
64 | 1. Short option flags for command-line use of *PhenotypeSimulator* were removed.
65 |
66 | **Output**
67 | 1. *PhenotypeSimulator* provides the option to save the simulated phenotypes and genotypes in formats compatible with a number of commonly used genetic association software (gemma, bimbam, plink, snptest) via writeStandardOutput.
68 | 1. Intermediate phenotype components are now saved per default.
69 | 1. Saving additional subsets of the simulated data has been removed.
70 |
71 | **Variance components**
72 | 1. Genotype simulation and kinship estimation: functions for genotype simulation and kinship estimation have been rewritten for
73 | significant speed-ups of the computation time [benchmarking](https://github.com/HannahVMeyer/PhenotypeSimulator-profiling).
74 |
75 | 1. geneticFixedEffects and noiseFixedEffects:
76 | 1. The effect size distributions of the shared effects are now modelled as the product of two exponential distributions
77 | (to yield an approximately uniform distributions) or the product of a normal distribution with user-specified
78 | parameters and a standard normal distribution.
79 |
80 | 1. The independent effects can now be specified to affect the same subset or different subsets of traits (via
81 | keepSameIndependent).
82 |
83 | 1. The overall number of traits affected by the effects can now be specified via pTraitsAffected.
84 |
85 |
86 | 1. correlatedBgEffects: the additional correlation between the traits can be specified by the user by providing an external
87 | correlation matrix.
88 |
--------------------------------------------------------------------------------
/vignettes/sample-scripts-external-genotype-simulation.Rmd:
--------------------------------------------------------------------------------
1 | ---
2 | title: "Genotype simulation supported by *PhenotypeSimulator*"
3 | author: "Hannah Meyer"
4 | date: "`r Sys.Date()`"
5 | output: pdf_document
6 | bibliography: genotype-simulation.bib
7 | csl: plos-genetics.csl
8 | vignette: >
9 | %\VignetteIndexEntry{Supported Genotype Simulation}
10 | %\VignetteEngine{knitr::rmarkdown}
11 | %\VignetteEncoding{UTF-8}
12 | ---
13 | ```{r, echo = FALSE, message=FALSE}
14 | library(knitr)
15 | knitr::opts_chunk$set(collapse = TRUE, comment = "#>")
16 | ```
17 |
18 | There are a number of different strategies to generate genotype data for genetic
19 | association studies.
20 |
21 | 1. **Simple bi-allelic SNPs without structure**:
22 | In the most simple case and assuming bi-allelic SNPs, each SNP is simulated
23 | from a binomial distribution with two trials and probability equal to the
24 | given allele frequencies. This simple approach does not simulate any
25 | dependency between the genotypes as is observed with LD structure in the
26 | genome. Simple bi-allelic SNPs can be simulated with
27 | PhenotypeSimulator::simulateGenotypes or via PLINK @Chang2015. For large
28 | datasets (~ 1 million SNPs), simulation via PLINK is recommended.
29 |
30 | 1. **Coalescent approaches**:
31 | Coalescent methods simulate genealogical events backward in time. These
32 | events typically include the coalescence of two sequences into a single
33 | ancestral lineage, recombination within a sequence or migration between
34 | populations. A coalescent-based approach to simulate whole genome data is
35 | implemented in GENOME @Liang2007 and its output format is supported as an
36 | input format for *PhenotypeSimulator*.
37 |
38 | 1. **Forward-time approaches**:
39 | Forward-time simulation methods evolve a population forward in time, subject
40 | to arbitrary genetic and demographic factors @Peng2010. Several algorithms
41 | are available, many of which allow customisation by building the simulation
42 | scheme in R or python, hence the output format of the genotypes can be
43 | specified by the user. *PhenotypeSimulator* offers reading genotypes from
44 | delimited-files and as such, any genotypes generated by these programmes can
45 | be read (e.g. MetaSim @Strand2002 or simuPop @Peng2005).
46 |
47 | 1. **Resampling approaches**:
48 | Resampling-based approaches combine existing genotype data into the
49 | genotypes of the simulated samples, thereby retaining allele frequency and
50 | LD patterns @Wright2007. A standard resampling-based approach that uses
51 | common genotype formats (common for the
52 | [oxford genetics format](http://www.stats.ox.ac.uk/~marchini/software/gwas/file_format.html))
53 | is Hapgen2 and an example usage is given below.
54 |
55 | Examples for genotype simulation via PLINK, GENOME and Hapgen2 are given below.
56 | The corresponding data can be found in the extdata folder.
57 |
58 | ## Simple bi-allelic SNPs without structure via PLINK
59 | Download [PLINK](https://www.cog-genomics.org/plink/1.9/) and create a folder
60 | for the PLINK output files. The example below uses PLINK to simulate 1000 SNPs,
61 | with allele frequencies between 0 and 1 for 100 controls and 0 cases.
62 | PhenotypeSimulator::readStandardGenotypes reads the resulting .bim, .fam, .bed
63 | files.
64 |
65 | ```{bash, eval=FALSE}
66 | # Serves as output directory for simulation and parameter file
67 | mkdir -p ~/PhenotypeSimulator/inst/extdata/genotypes/plink
68 | cd ~/PhenotypeSimulator/inst/extdata/genotypes/plink
69 |
70 | # Write paramter file to simulate 1000 SNPs, with allele frequencies between 0
71 | # and 1 for 100 controls and 0 cases
72 | echo -e "1000\tSNP\t0.00\t1.00\t1.00\t1.00" > plink_sim_par.txt
73 |
74 | plink --simulate plink_sim_par.txt \
75 | --simulate-ncases 0 \
76 | --simulate-ncontrols 100 \
77 | --out genotypes_plink
78 | ```
79 |
80 |
81 |
82 | ## Coalescent simulation via Genome
83 | Download [GENOME](http://csg.sph.umich.edu/liang/genome/download.html) and
84 | create a folder for the GENOME output files. The example below uses GENOME to
85 | simulate genetic data for a population comprised of three sub-populations with
86 | 30, 30 and 40 samples. The simulated POPULATION PROFILE, SNP positions and the
87 | genotypes are all saved in genotypes_genome.txt.
88 | PhenotypeSimulator::readStandardGenotypes reads genotype information by parsing
89 | this output file and extracting the samples information following the line
90 | 'Samples:'
91 |
92 | ```{bash, eval=FALSE}
93 | mkdir -p ~/PhenotypeSimulator/inst/extdata/genotypes/genome
94 | cd ~/PhenotypeSimulator/inst/extdata/genotypes/genome
95 |
96 | # subpopulation with 30, 30 and 40 individuals each
97 | genome -pop 3 30 30 40 > genotypes_genome.txt
98 | ```
99 |
100 | ## Resampling-based simulation via Hapgen2
101 | Download [hapgen2](http://mathgen.stats.ox.ac.uk/genetics_software/hapgen/hapgen2.html),
102 | [hapgen example files](http://mathgen.stats.ox.ac.uk/genetics_software/hapgen/download/example/hapgen2.example.tgz)
103 | and the 1000Genomes data from the
104 | [impute2 webpage](https://mathgen.stats.ox.ac.uk/impute/impute_v1.html#Using_IMPUTE_with_the_HapMap_Data)
105 | as starting point for the resampling-based genotype simulation with hapgen2.
106 | For formating of the 1000Genomes data, have a look at this [vignette](https://hannahvmeyer.github.io/PhenotypeSimulator/articles/Simulation-and-LinearModel.html#genotype-simulation-via-hapgen2).
107 | The example files contain example haplotypes (ex.haps), legend files (ex.leg) map
108 | (ex.map) and TagSNP files (ex.tags). The following examples simulates genotype
109 | data for 100 controls and 0 cases with 1000 SNPs from chromosome 1.
110 | PhenotypeSimulator::readStandardGenotypes reads the resulting .gen and .samples
111 | files.
112 |
113 | ```{bash, eval=FALSE}
114 | # contains ex.leg file and serves as output directory
115 | mkdir -p ~/PhenotypeSimulator/inst/extdata/genotypes/hapgen
116 | cd ~/PhenotypeSimulator/inst/extdata/genotypes/hapgen
117 |
118 | # contains 1000Genomes haplotype data used for sampling
119 | hapdir=/path/to/CEU.0908.impute.files
120 |
121 | hapgen2 -m $hapdir/genetic_map_chr1_combined_b37.txt \
122 | -l ex.leg \
123 | -h $hapdir/chr1.ceu_subset.hap \
124 | -o genotypes_hapgen \
125 | -n 100 0 \
126 | -dl 45162 0 0 0 \
127 | -no_haps_output
128 | ```
129 |
130 | ## References
131 |
--------------------------------------------------------------------------------
/docs/LICENSE-text.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | License • PhenotypeSimulator
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 |
57 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
139 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 | License
148 |
149 |
150 | YEAR: 2017
151 | COPYRIGHT HOLDER: Hannah Verena Meyer
152 |
153 |
154 |
155 |
156 |
161 |
162 |
163 |
164 |
165 |
166 |
176 |
177 |
178 |
179 |
180 |
181 |
182 |
183 |
184 |
185 |
--------------------------------------------------------------------------------
/docs/404.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Page not found (404) • PhenotypeSimulator
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 |
57 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
139 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 | Page not found (404)
148 |
149 |
150 | Content not found. Please use links in the navbar.
151 |
152 |
153 |
154 |
159 |
160 |
161 |
162 |
163 |
164 |
174 |
175 |
176 |
177 |
178 |
179 |
180 |
181 |
182 |
183 |
--------------------------------------------------------------------------------
/docs/articles/index.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Articles • PhenotypeSimulator
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 |
57 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
139 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 | Articles
148 |
149 |
150 |
151 | All vignettes
152 |
153 |
154 |
164 |
165 |
166 |
167 |
168 |
169 |
179 |
180 |
181 |
182 |
183 |
184 |
185 |
186 |
187 |
188 |
--------------------------------------------------------------------------------
/docs/reference/PhenotypeSimulator.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | PhenotypeSimulator: A package for simulating phenotypes from different
10 | genetic and noise components — PhenotypeSimulator • PhenotypeSimulator
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 |
47 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
61 |
62 |
63 |
64 |
65 |
66 |
67 |
68 |
69 |
143 |
144 |
145 |
146 |
147 |
148 |
149 |
150 |
151 | PhenotypeSimulator: A package for simulating phenotypes from different
152 | genetic and noise components
153 | Source: R/PhenotypeSimulator.R
154 | PhenotypeSimulator.Rd
155 |
156 |
157 |
158 | PhenotypeSimulator: A package for simulating phenotypes from different
159 | genetic and noise components
160 |
161 |
162 |
163 |
164 |
165 |
166 |
171 |
172 |
173 |
174 |
184 |
185 |
186 |
187 |
188 |
189 |
190 |
191 |
192 |
193 |
--------------------------------------------------------------------------------
/docs/authors.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Citation and Authors • PhenotypeSimulator
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 |
57 |
58 |
59 |
60 |
61 |
62 |
63 |
195 |
196 |
197 |
198 |
199 |
200 |
201 |
202 |
203 |
--------------------------------------------------------------------------------
/docs/reference/addNonNulls.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Add all non-NULL elements of list. — addNonNulls • PhenotypeSimulator
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 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
66 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 |
148 | Add all non-NULL elements of list.
149 | Source: R/utilityFunctions.R
150 | addNonNulls.Rd
151 |
152 |
153 |
154 | Add all non-NULL elements of list.
155 |
156 |
157 | addNonNulls(compList)
158 |
159 | Arguments
160 |
161 |
163 | compList
164 | List of numeric matrices or data.frames of the equal
165 | dimensions.
167 |
168 |
169 | Value
170 |
171 | Matrix or data.frame containing sum of all list elements where
172 | is.null is FALSE.
173 |
174 |
175 |
180 |
181 |
182 |
183 |
193 |
194 |
195 |
196 |
197 |
198 |
199 |
200 |
201 |
202 |
--------------------------------------------------------------------------------
/docs/reference/read_lines.html:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 | Scan file for specific line numbers — read_lines • PhenotypeSimulator
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 |
58 |
59 |
60 |
61 |
62 |
63 |
64 |
65 |
66 |
140 |
141 |
142 |
143 |
144 |
145 |
146 |
147 |
152 |
153 |
154 | Scan file for specific line numbers
155 |
156 |
157 | read_lines(filename, lines, sep = "\n")
158 |
159 | Arguments
160 |
161 |
163 | filename
164 | /path/to/chromosomefile [string]
166 |
167 | lines
168 | vector of line numbers [integer] to be read
170 |
171 | sep
172 | [string] end-of-line delimiter
174 |
175 |
176 |
177 |
178 |
183 |
184 |
185 |
186 |
196 |
197 |
198 |
199 |
200 |
201 |
202 |
203 |
204 |
205 |
--------------------------------------------------------------------------------
/tests/testthat/test-createphenotypeFunctions.r:
--------------------------------------------------------------------------------
1 | context('Test createphenotype functions')
2 |
3 | noiseBg <- noiseBgEffects(N=100, P=10)
4 | noiseFixed <- noiseFixedEffects(N=100, P=10)
5 | noiseCorrelated <- correlatedBgEffects(N=100, P=10, pcorr=0.6)
6 | genotypes <- simulateGenotypes(N=100, NrSNP=20, verbose=FALSE)
7 | kinship <- getKinship(N=100, X=genotypes$genotypes, standardise=TRUE,
8 | verbose=FALSE)
9 | causalSNPs <- getCausalSNPs(N=100, genotypes=genotypes$genotypes, verbose=FALSE)
10 | geneticFixed <- geneticFixedEffects(X_causal=causalSNPs, P=10, N=100,
11 | verbose=FALSE)
12 | geneticBg <- geneticBgEffects(P=10, N=100, kinship=kinship)
13 |
14 | context("Tests rescaleVariance")
15 | test_that("rescaleVariance fails when propvar is not numeric", {
16 | expect_error(rescaleVariance(noiseBg$shared, propvar = 'a'),
17 | "propvar needs to be numeric")
18 | })
19 | test_that("rescaleVariance fails when propvar is out of range", {
20 | expect_error(rescaleVariance(noiseBg$shared, propvar = 1.1),
21 | "propvar cannot be less than 0 and or greater than 1")
22 | })
23 | test_that("rescaleVariance fails when component is not a matrix", {
24 | expect_error(rescaleVariance(data.frame(noiseBg$shared), propvar = 0.1),
25 | "component needs to be a matrix")
26 | })
27 | test_that("rescaleVariance returns NULL when no propvar is specified", {
28 | expect_equal(rescaleVariance(noiseBg$shared, propvar = 0), NULL)
29 | })
30 |
31 | test_that("rescaleVariance returns correclty scaled average column variance", {
32 | tmp <- rescaleVariance(noiseBg$shared, propvar = 0.1)
33 | expect_equal(mean(diag(var(tmp$component))), 0.1, tolerance=5e-5)
34 | })
35 |
36 |
37 | context("Tests setModel")
38 | test_that("setModel fails with input error", {
39 | expect_error(setModel(), "No variance components specified")
40 | })
41 |
42 | test_that("setModel fails with out of range error", {
43 | expect_error(setModel(genVar=0.4, h2s=1.2, theta=0.8, eta=0.8,
44 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
45 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
46 | pTraitIndependentConfounders=0.2,
47 | pIndependentGenetic=0.4,
48 | pTraitIndependentGenetic=0.2, verbose=FALSE),
49 | "Proportions have to be specified between 0 and 1:")
50 | })
51 | test_that("setModel fails with missing variance error", {
52 | expect_error(setModel(h2s=1, theta=0.8, eta=0.8,
53 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
54 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
55 | pTraitIndependentConfounders=0.2,
56 | pIndependentGenetic=0.4,
57 | pTraitIndependentGenetic=0.2, verbose=FALSE),
58 | "Neither genVar nor noiseVar are provided, thus")
59 | })
60 |
61 | test_that("setModel fails with sum of variance error", {
62 | expect_error(setModel(genVar=0.2, noiseVar=0.9, h2s=1, theta=0.8, eta=0.8,
63 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
64 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
65 | pTraitIndependentConfounders=0.2,
66 | pIndependentGenetic=0.4,
67 | pTraitIndependentGenetic=0.2, verbose=FALSE),
68 | "Sum of genetic variance and noise variance is greater than 1")
69 | })
70 |
71 | test_that("setModel fails with extra noise component error", {
72 | expect_error(setModel(noiseVar=0, h2s=1, theta=0.8, eta=0.8,
73 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
74 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
75 | pTraitIndependentConfounders=0.2,
76 | pIndependentGenetic=0.4,
77 | pTraitIndependentGenetic=0.2, verbose=FALSE),
78 | "The noise variance is set to 0 ")
79 | })
80 |
81 | test_that("setModel fails with extra genetic component error", {
82 | expect_error(setModel(genVar=0, h2s=1, theta=0.8, eta=0.8,
83 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
84 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
85 | pTraitIndependentConfounders=0.2,
86 | pIndependentGenetic=0.4,
87 | pTraitIndependentGenetic=0.2, verbose=FALSE),
88 | "The genetic variance is set to 0 ")
89 | })
90 |
91 | test_that("setModel fails with missing noise component error", {
92 | expect_error(setModel(noiseVar=0.8, h2s=1, theta=0.8, eta=0.8,
93 | pcorr=0.6, pIndependentConfounders=0.4,
94 | pTraitIndependentConfounders=0.2,
95 | pIndependentGenetic=0.4,
96 | pTraitIndependentGenetic=0.2,
97 | verbose=FALSE),
98 | "Neither delta ")
99 | })
100 |
101 | test_that("setModel fails with proportions larger than 1 error", {
102 | expect_error(setModel(noiseVar=0.8, h2s=1, theta=0.8, delta=0.8, phi=0.2,
103 | rho=0.2, eta=0.8,
104 | pcorr=0.6, pIndependentConfounders=0.4,
105 | pTraitIndependentConfounders=0.2,
106 | pIndependentGenetic=0.4,
107 | pTraitIndependentGenetic=0.2,
108 | verbose=FALSE),
109 | "Sum of the proportion of the variance of noise")
110 | })
111 |
112 | test_that("setModel fails with proportions less than 1 error", {
113 | expect_error(setModel(noiseVar=0.8, h2s=1, theta=0.8, delta=0.2, phi=0.2,
114 | rho=0.2, eta=0.8,
115 | pcorr=0.6, pIndependentConfounders=0.4,
116 | pTraitIndependentConfounders=0.2,
117 | pIndependentGenetic=0.4,
118 | pTraitIndependentGenetic=0.2,
119 | verbose=FALSE),
120 | "Sum of the proportion of the variance of noise")
121 | })
122 |
123 | test_that("setModel fails with insufficient noise component error", {
124 | expect_error(setModel(noiseVar=0.8, h2s=1, theta=0.8, eta=0.8,
125 | delta=0.8, pcorr=0.6, pIndependentConfounders=0.4,
126 | pTraitIndependentConfounders=0.2,
127 | pIndependentGenetic=0.4,
128 | pTraitIndependentGenetic=0.2,
129 | verbose=FALSE),
130 | "Not enough components provided to set proportions of")
131 | })
132 |
133 | test_that("setModel fails with insufficient genetic component error", {
134 | expect_error(setModel(genVar=0.8, theta=0.8, eta=0.8,
135 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
136 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
137 | pTraitIndependentConfounders=0.2,
138 | pIndependentGenetic=0.4,
139 | pTraitIndependentGenetic=0.2, verbose=FALSE),
140 | "Neither genetic variant effect variance")
141 | })
142 |
143 | test_that("setModel fails with sum of genetic component error", {
144 | expect_error(setModel(genVar=0.8, h2s=0.4, h2bg=0.3, theta=0.8, eta=0.8,
145 | delta=0.2, gamma=0.8, rho=0.2, phi=0.6,
146 | alpha=0.8, pcorr=0.6, pIndependentConfounders=0.4,
147 | pTraitIndependentConfounders=0.2,
148 | pIndependentGenetic=0.4,
149 | pTraitIndependentGenetic=0.2, verbose=FALSE),
150 | "Sum of the proportion of the variance of genetic")
151 | })
152 |
153 | context("Tests runSimulation")
154 | test_that("runSimulation returns correct output", {
155 | expect_equal(length(runSimulation(N=10, P=5, genVar=1, h2s=0.2)), 5)
156 | })
--------------------------------------------------------------------------------
145 |
163 |
164 |
165 |
166 |
176 |
146 |
155 |
156 |
161 |
162 |
147 |
149 |
150 | License
148 |YEAR: 2017 151 | COPYRIGHT HOLDER: Hannah Verena Meyer 152 |153 | 154 |
145 |
161 |
162 |
163 |
164 |
174 |
146 |
153 |
154 |
159 |
160 |
147 |
149 |
150 | Content not found. Please use links in the navbar.
151 |
152 | Page not found (404)
148 |
145 |
167 |
168 |
169 |
179 |
146 |
166 |
147 |
149 |
150 | Articles
148 |
151 |
165 | All vignettes
152 | 153 | 154 | 164 |
149 |
172 |
173 |
174 |
184 |
150 |
166 |
171 |
151 |
155 |
156 |
157 | PhenotypeSimulator: A package for simulating phenotypes from different 152 | genetic and noise components
153 | Source:R/PhenotypeSimulator.R
154 | PhenotypeSimulator.Rd
158 |
161 |
162 |
163 |
164 |
165 | PhenotypeSimulator: A package for simulating phenotypes from different 159 | genetic and noise components
160 |
146 |
181 |
182 |
183 |
193 |
147 |
161 |
168 |
169 |
175 |
180 |
148 |
151 |
152 |
153 | Add all non-NULL elements of list.
149 | Source:R/utilityFunctions.R
150 | addNonNulls.Rd
154 |
156 |
157 | Add all non-NULL elements of list.
155 |addNonNulls(compList)158 | 159 |
Arguments
160 || compList | 164 |List of numeric matrices or data.frames of the equal 165 | dimensions. |
166 |
|---|
Value
170 | 171 |Matrix or data.frame containing sum of all list elements where
172 | is.null is FALSE.
146 |
184 |
185 |
186 |
196 |
147 |
152 |
153 |
161 |
175 |
176 |
177 |
178 |
183 |
154 |
156 |
157 | Scan file for specific line numbers
155 |read_lines(filename, lines, sep = "\n")158 | 159 |
Arguments
160 || filename | 164 |/path/to/chromosomefile [string] |
165 |
|---|---|
| lines | 168 |vector of line numbers [integer] to be read |
169 |
| sep | 172 |[string] end-of-line delimiter |
173 |