EVOLUTION-MANAGER
Edit File: calc.penalties.html
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><title>R: Calculate LOD penalties</title> <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> <link rel="stylesheet" type="text/css" href="R.css" /> </head><body> <table width="100%" summary="page for calc.penalties {qtl}"><tr><td>calc.penalties {qtl}</td><td style="text-align: right;">R Documentation</td></tr></table> <h2>Calculate LOD penalties</h2> <h3>Description</h3> <p>Derive penalties for the penalized LOD scores (used by <code><a href="stepwiseqtl.html">stepwiseqtl</a></code>) on the basis of permutation results from a two-dimensional, two-QTL scan (obtained by <code><a href="scantwo.html">scantwo</a></code>). </p> <h3>Usage</h3> <pre>calc.penalties(perms, alpha=0.05, lodcolumn)</pre> <h3>Arguments</h3> <table summary="R argblock"> <tr valign="top"><td><code>perms</code></td> <td> <p>Permutation results from <code><a href="scantwo.html">scantwo</a></code>.</p> </td></tr> <tr valign="top"><td><code>alpha</code></td> <td> <p>Significance level.</p> </td></tr> <tr valign="top"><td><code>lodcolumn</code></td> <td> <p>If the scantwo permutation results contain LOD scores for multiple phenotypes, this argument indicates which to use in the summary. This may be a vector. If missing, penalties for all phenotypes are calculated.</p> </td></tr> </table> <h3>Details</h3> <p>Thresholds derived from <code><a href="scantwo.html">scantwo</a></code> permutations (that is, for a two-dimensional, two-QTL genome scan) are used to calculate penalties on main effects and interactions. </p> <p>The main effect penalty is the 1-<code>alpha</code> quantile of the null distribution of the genome-wide maximum LOD score from a single-QTL genome scan (as with <code><a href="scanone.html">scanone</a></code>). </p> <p>The "heavy" interaction penalty is the 1-<code>alpha</code> quantile of the null distribution of the maximum interaction LOD score (that is, the <i>log10</i> likelihood ratio comparing the best model with two interacting QTL to the best model with two additive QTL) from a two-dimensional, two-QTL genome scan (as with <code><a href="scantwo.html">scantwo</a></code>). </p> <p>The "light" interaction penality is the difference between the <code>"fv1"</code> threshold from the <code><a href="scantwo.html">scantwo</a></code> permutations (that is, the 1-<code>alpha</code> quantile of the LOD score comparing the best model with two interacting QTL to the best single-QTL model) and the main effect penalty. </p> <p>If the permutations results were obtained with <code>perm.Xsp=TRUE</code>, to give X-chr-specific results, six penalties are calculated: main effect for autosomes, main effect for X chr, heavy penalty on A:A interactions, light penalty on A:A interactions, penalty on A:X interactions, and penalty on X:X interactions. </p> <h3>Value</h3> <p>Vector of three values indicating the penalty on main effects and heavy and light penalties on interactions, or a matrix of such results, with each row corresponding to a different phenotype. </p> <p>If the input permutations are X-chromosome-specific, the result has six values: main effect for autosomes, main effect for X chr, heavy penalty on A:A interactions, light penalty on A:A interactions, penalty on A:X interactions, and penalty on X:X interactions. </p> <h3>Author(s)</h3> <p>Karl W Broman, <a href="mailto:broman@wisc.edu">broman@wisc.edu</a> </p> <h3>References</h3> <p>Manichaikul, A., Moon, J. Y., Sen, Ś, Yandell, B. S. and Broman, K. W. (2009) A model selection approach for the identification of quantitative trait loci in experimental crosses, allowing epistasis. <em>Genetics</em>, <b>181</b>, 1077–1086. </p> <h3>See Also</h3> <p><code><a href="scantwo.html">scantwo</a></code>, <code><a href="stepwiseqtl.html">stepwiseqtl</a></code> </p> <h3>Examples</h3> <pre> data(fake.f2) fake.f2 <- calc.genoprob(fake.f2, step=5) out.2dim <- scantwo(fake.f2, method="hk") # permutations ## Not run: permo.2dim <- scantwo(fake.f2, method="hk", n.perm=1000) summary(permo.2dim, alpha=0.05) # penalties calc.penalties(permo.2dim) </pre> <hr /><div style="text-align: center;">[Package <em>qtl</em> version 1.46-2 <a href="00Index.html">Index</a>]</div> </body></html>