zzh
/
gatk-3.8
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Made the AF prior probability distribution dynamic respect 

to the total-ploidy (added ploidy accross samples).

Changes:
--------

* Instead of calculate a fixed log10 prior array with a fix
   total likelihood we use a new component, the AFPriorProvider
   to generate the priors for different total plodies on
   demand; these are cached however so there is no unecessary
   recompute involved.
This commit is contained in:
Valentin Ruano-Rubio 2014-09-10 13:20:03 -04:00
parent 31e58ae4ec
commit ce8e93fa51
4 changed files with 315 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

112
protected/gatk-tools-protected/src/main/java/org/broadinstitute/gatk/tools/walkers/genotyper/AFPriorProvider.java 100644
View File

@ -0,0 +1,112 @@
/*
* By downloading the PROGRAM you agree to the following terms of use:
*
* BROAD INSTITUTE - SOFTWARE LICENSE AGREEMENT - FOR ACADEMIC NON-COMMERCIAL RESEARCH PURPOSES ONLY
*
* This Agreement is made between the Broad Institute, Inc. with a principal address at 7 Cambridge Center, Cambridge, MA 02142 (BROAD) and the LICENSEE and is effective at the date the downloading is completed (EFFECTIVE DATE).
*
* WHEREAS, LICENSEE desires to license the PROGRAM, as defined hereinafter, and BROAD wishes to have this PROGRAM utilized in the public interest, subject only to the royalty-free, nonexclusive, nontransferable license rights of the United States Government pursuant to 48 CFR 52.227-14; and
* WHEREAS, LICENSEE desires to license the PROGRAM and BROAD desires to grant a license on the following terms and conditions.
* NOW, THEREFORE, in consideration of the promises and covenants made herein, the parties hereto agree as follows:
*
* 1. DEFINITIONS
* 1.1 PROGRAM shall mean copyright in the object code and source code known as GATK2 and related documentation, if any, as they exist on the EFFECTIVE DATE and can be downloaded from http://www.broadinstitute/GATK on the EFFECTIVE DATE.
*
* 2. LICENSE
* 2.1 Grant. Subject to the terms of this Agreement, BROAD hereby grants to LICENSEE, solely for academic non-commercial research purposes, a non-exclusive, non-transferable license to: (a) download, execute and display the PROGRAM and (b) create bug fixes and modify the PROGRAM.
* The LICENSEE may apply the PROGRAM in a pipeline to data owned by users other than the LICENSEE and provide these users the results of the PROGRAM provided LICENSEE does so for academic non-commercial purposes only. For clarification purposes, academic sponsored research is not a commercial use under the terms of this Agreement.
* 2.2 No Sublicensing or Additional Rights. LICENSEE shall not sublicense or distribute the PROGRAM, in whole or in part, without prior written permission from BROAD. LICENSEE shall ensure that all of its users agree to the terms of this Agreement. LICENSEE further agrees that it shall not put the PROGRAM on a network, server, or other similar technology that may be accessed by anyone other than the LICENSEE and its employees and users who have agreed to the terms of this agreement.
* 2.3 License Limitations. Nothing in this Agreement shall be construed to confer any rights upon LICENSEE by implication, estoppel, or otherwise to any computer software, trademark, intellectual property, or patent rights of BROAD, or of any other entity, except as expressly granted herein. LICENSEE agrees that the PROGRAM, in whole or part, shall not be used for any commercial purpose, including without limitation, as the basis of a commercial software or hardware product or to provide services. LICENSEE further agrees that the PROGRAM shall not be copied or otherwise adapted in order to circumvent the need for obtaining a license for use of the PROGRAM.
*
* 3. OWNERSHIP OF INTELLECTUAL PROPERTY
* LICENSEE acknowledges that title to the PROGRAM shall remain with BROAD. The PROGRAM is marked with the following BROAD copyright notice and notice of attribution to contributors. LICENSEE shall retain such notice on all copies. LICENSEE agrees to include appropriate attribution if any results obtained from use of the PROGRAM are included in any publication.
* Copyright 2012 Broad Institute, Inc.
* Notice of attribution: The GATK2 program was made available through the generosity of Medical and Population Genetics program at the Broad Institute, Inc.
* LICENSEE shall not use any trademark or trade name of BROAD, or any variation, adaptation, or abbreviation, of such marks or trade names, or any names of officers, faculty, students, employees, or agents of BROAD except as states above for attribution purposes.
*
* 4. INDEMNIFICATION
* LICENSEE shall indemnify, defend, and hold harmless BROAD, and their respective officers, faculty, students, employees, associated investigators and agents, and their respective successors, heirs and assigns, (Indemnitees), against any liability, damage, loss, or expense (including reasonable attorneys fees and expenses) incurred by or imposed upon any of the Indemnitees in connection with any claims, suits, actions, demands or judgments arising out of any theory of liability (including, without limitation, actions in the form of tort, warranty, or strict liability and regardless of whether such action has any factual basis) pursuant to any right or license granted under this Agreement.
*
* 5. NO REPRESENTATIONS OR WARRANTIES
* THE PROGRAM IS DELIVERED AS IS. BROAD MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND CONCERNING THE PROGRAM OR THE COPYRIGHT, EXPRESS OR IMPLIED, INCLUDING, WITHOUT LIMITATION, WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, NONINFRINGEMENT, OR THE ABSENCE OF LATENT OR OTHER DEFECTS, WHETHER OR NOT DISCOVERABLE. BROAD EXTENDS NO WARRANTIES OF ANY KIND AS TO PROGRAM CONFORMITY WITH WHATEVER USER MANUALS OR OTHER LITERATURE MAY BE ISSUED FROM TIME TO TIME.
* IN NO EVENT SHALL BROAD OR ITS RESPECTIVE DIRECTORS, OFFICERS, EMPLOYEES, AFFILIATED INVESTIGATORS AND AFFILIATES BE LIABLE FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES OF ANY KIND, INCLUDING, WITHOUT LIMITATION, ECONOMIC DAMAGES OR INJURY TO PROPERTY AND LOST PROFITS, REGARDLESS OF WHETHER BROAD SHALL BE ADVISED, SHALL HAVE OTHER REASON TO KNOW, OR IN FACT SHALL KNOW OF THE POSSIBILITY OF THE FOREGOING.
*
* 6. ASSIGNMENT
* This Agreement is personal to LICENSEE and any rights or obligations assigned by LICENSEE without the prior written consent of BROAD shall be null and void.
*
* 7. MISCELLANEOUS
* 7.1 Export Control. LICENSEE gives assurance that it will comply with all United States export control laws and regulations controlling the export of the PROGRAM, including, without limitation, all Export Administration Regulations of the United States Department of Commerce. Among other things, these laws and regulations prohibit, or require a license for, the export of certain types of software to specified countries.
* 7.2 Termination. LICENSEE shall have the right to terminate this Agreement for any reason upon prior written notice to BROAD. If LICENSEE breaches any provision hereunder, and fails to cure such breach within thirty (30) days, BROAD may terminate this Agreement immediately. Upon termination, LICENSEE shall provide BROAD with written assurance that the original and all copies of the PROGRAM have been destroyed, except that, upon prior written authorization from BROAD, LICENSEE may retain a copy for archive purposes.
* 7.3 Survival. The following provisions shall survive the expiration or termination of this Agreement: Articles 1, 3, 4, 5 and Sections 2.2, 2.3, 7.3, and 7.4.
* 7.4 Notice. Any notices under this Agreement shall be in writing, shall specifically refer to this Agreement, and shall be sent by hand, recognized national overnight courier, confirmed facsimile transmission, confirmed electronic mail, or registered or certified mail, postage prepaid, return receipt requested. All notices under this Agreement shall be deemed effective upon receipt.
* 7.5 Amendment and Waiver; Entire Agreement. This Agreement may be amended, supplemented, or otherwise modified only by means of a written instrument signed by all parties. Any waiver of any rights or failure to act in a specific instance shall relate only to such instance and shall not be construed as an agreement to waive any rights or fail to act in any other instance, whether or not similar. This Agreement constitutes the entire agreement among the parties with respect to its subject matter and supersedes prior agreements or understandings between the parties relating to its subject matter.
* 7.6 Binding Effect; Headings. This Agreement shall be binding upon and inure to the benefit of the parties and their respective permitted successors and assigns. All headings are for convenience only and shall not affect the meaning of any provision of this Agreement.
* 7.7 Governing Law. This Agreement shall be construed, governed, interpreted and applied in accordance with the internal laws of the Commonwealth of Massachusetts, U.S.A., without regard to conflict of laws principles.
*/
package org.broadinstitute.gatk.tools.walkers.genotyper;
import java.util.Arrays;
/**
* Class that produces allele-frequency priors.
*
* <p>
* This class is thread-unsafe for performance and so should be used accordingly.
* </p>
*
* @author Valentin Ruano-Rubio &lt;valentin@broadinstitute.org&gt;
*/
public abstract class AFPriorProvider {
private double[][] priorByTotalPloidy;
protected AFPriorProvider() {
}
/**
* Returns the priors given a total-ploidy (the total number of genome copies across all samples).
*
* <p>
* For performance sake the returned value is a direct reference ot the cached prior, so the client code
* must not modify its content.
* </p>
*
* @param totalPloidy the requested total ploidy.
*
* @return never {@code null}, please do not modify its content. An array of {@code totalPloidy + 1} positions where
* the ith position is the log10(prior) probability of the an alternative allele AC to be exactly <i>i</i> copies in
* a draw of {@code totalPloidy} elements.
*/
public double[] forTotalPloidy(final int totalPloidy) {
if (totalPloidy < 0)
throw new IllegalArgumentException("the total-ploidy cannot be negative");
ensureCapacity(totalPloidy);
final double[] cachedResult = priorByTotalPloidy[totalPloidy];
if (cachedResult == null)
return priorByTotalPloidy[totalPloidy] = buildPriors(totalPloidy);
else
return cachedResult;
}
/**
* Make sure that structures have enough capacity to hold the information up to the given total-ploidy.
* @param totalPloidy
*/
protected void ensureCapacity(final int totalPloidy) {
if (priorByTotalPloidy == null)
priorByTotalPloidy = new double[totalPloidy + 1][]; // just enough for those cases in where we have a fix total-ploidy.
else if (priorByTotalPloidy.length - 1 < totalPloidy)
priorByTotalPloidy = Arrays.copyOf(priorByTotalPloidy,Math.max(priorByTotalPloidy.length << 1,totalPloidy + 1));
}
/**
* Given a total ploidy construct the allele prior probabilities array.
* @param totalPloidy the target total-ploidy. Code can assume that is a non-negative number.
*
* @return never {@code null}, an array of exactly {@code totalPloidy + 1} positions that satisifed the
* contract {@link #forTotalPloidy(int) forTotalPloidy(totalPloidy)}.
*/
protected abstract double[] buildPriors(final int totalPloidy);
}

87
protected/gatk-tools-protected/src/main/java/org/broadinstitute/gatk/tools/walkers/genotyper/CustomAFPriorProvider.java 100644
View File

@ -0,0 +1,87 @@
/*
* By downloading the PROGRAM you agree to the following terms of use:
*
* BROAD INSTITUTE - SOFTWARE LICENSE AGREEMENT - FOR ACADEMIC NON-COMMERCIAL RESEARCH PURPOSES ONLY
*
* This Agreement is made between the Broad Institute, Inc. with a principal address at 7 Cambridge Center, Cambridge, MA 02142 (BROAD) and the LICENSEE and is effective at the date the downloading is completed (EFFECTIVE DATE).
*
* WHEREAS, LICENSEE desires to license the PROGRAM, as defined hereinafter, and BROAD wishes to have this PROGRAM utilized in the public interest, subject only to the royalty-free, nonexclusive, nontransferable license rights of the United States Government pursuant to 48 CFR 52.227-14; and
* WHEREAS, LICENSEE desires to license the PROGRAM and BROAD desires to grant a license on the following terms and conditions.
* NOW, THEREFORE, in consideration of the promises and covenants made herein, the parties hereto agree as follows:
*
* 1. DEFINITIONS
* 1.1 PROGRAM shall mean copyright in the object code and source code known as GATK2 and related documentation, if any, as they exist on the EFFECTIVE DATE and can be downloaded from http://www.broadinstitute/GATK on the EFFECTIVE DATE.
*
* 2. LICENSE
* 2.1 Grant. Subject to the terms of this Agreement, BROAD hereby grants to LICENSEE, solely for academic non-commercial research purposes, a non-exclusive, non-transferable license to: (a) download, execute and display the PROGRAM and (b) create bug fixes and modify the PROGRAM.
* The LICENSEE may apply the PROGRAM in a pipeline to data owned by users other than the LICENSEE and provide these users the results of the PROGRAM provided LICENSEE does so for academic non-commercial purposes only. For clarification purposes, academic sponsored research is not a commercial use under the terms of this Agreement.
* 2.2 No Sublicensing or Additional Rights. LICENSEE shall not sublicense or distribute the PROGRAM, in whole or in part, without prior written permission from BROAD. LICENSEE shall ensure that all of its users agree to the terms of this Agreement. LICENSEE further agrees that it shall not put the PROGRAM on a network, server, or other similar technology that may be accessed by anyone other than the LICENSEE and its employees and users who have agreed to the terms of this agreement.
* 2.3 License Limitations. Nothing in this Agreement shall be construed to confer any rights upon LICENSEE by implication, estoppel, or otherwise to any computer software, trademark, intellectual property, or patent rights of BROAD, or of any other entity, except as expressly granted herein. LICENSEE agrees that the PROGRAM, in whole or part, shall not be used for any commercial purpose, including without limitation, as the basis of a commercial software or hardware product or to provide services. LICENSEE further agrees that the PROGRAM shall not be copied or otherwise adapted in order to circumvent the need for obtaining a license for use of the PROGRAM.
*
* 3. OWNERSHIP OF INTELLECTUAL PROPERTY
* LICENSEE acknowledges that title to the PROGRAM shall remain with BROAD. The PROGRAM is marked with the following BROAD copyright notice and notice of attribution to contributors. LICENSEE shall retain such notice on all copies. LICENSEE agrees to include appropriate attribution if any results obtained from use of the PROGRAM are included in any publication.
* Copyright 2012 Broad Institute, Inc.
* Notice of attribution: The GATK2 program was made available through the generosity of Medical and Population Genetics program at the Broad Institute, Inc.
* LICENSEE shall not use any trademark or trade name of BROAD, or any variation, adaptation, or abbreviation, of such marks or trade names, or any names of officers, faculty, students, employees, or agents of BROAD except as states above for attribution purposes.
*
* 4. INDEMNIFICATION
* LICENSEE shall indemnify, defend, and hold harmless BROAD, and their respective officers, faculty, students, employees, associated investigators and agents, and their respective successors, heirs and assigns, (Indemnitees), against any liability, damage, loss, or expense (including reasonable attorneys fees and expenses) incurred by or imposed upon any of the Indemnitees in connection with any claims, suits, actions, demands or judgments arising out of any theory of liability (including, without limitation, actions in the form of tort, warranty, or strict liability and regardless of whether such action has any factual basis) pursuant to any right or license granted under this Agreement.
*
* 5. NO REPRESENTATIONS OR WARRANTIES
* THE PROGRAM IS DELIVERED AS IS. BROAD MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND CONCERNING THE PROGRAM OR THE COPYRIGHT, EXPRESS OR IMPLIED, INCLUDING, WITHOUT LIMITATION, WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, NONINFRINGEMENT, OR THE ABSENCE OF LATENT OR OTHER DEFECTS, WHETHER OR NOT DISCOVERABLE. BROAD EXTENDS NO WARRANTIES OF ANY KIND AS TO PROGRAM CONFORMITY WITH WHATEVER USER MANUALS OR OTHER LITERATURE MAY BE ISSUED FROM TIME TO TIME.
* IN NO EVENT SHALL BROAD OR ITS RESPECTIVE DIRECTORS, OFFICERS, EMPLOYEES, AFFILIATED INVESTIGATORS AND AFFILIATES BE LIABLE FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES OF ANY KIND, INCLUDING, WITHOUT LIMITATION, ECONOMIC DAMAGES OR INJURY TO PROPERTY AND LOST PROFITS, REGARDLESS OF WHETHER BROAD SHALL BE ADVISED, SHALL HAVE OTHER REASON TO KNOW, OR IN FACT SHALL KNOW OF THE POSSIBILITY OF THE FOREGOING.
*
* 6. ASSIGNMENT
* This Agreement is personal to LICENSEE and any rights or obligations assigned by LICENSEE without the prior written consent of BROAD shall be null and void.
*
* 7. MISCELLANEOUS
* 7.1 Export Control. LICENSEE gives assurance that it will comply with all United States export control laws and regulations controlling the export of the PROGRAM, including, without limitation, all Export Administration Regulations of the United States Department of Commerce. Among other things, these laws and regulations prohibit, or require a license for, the export of certain types of software to specified countries.
* 7.2 Termination. LICENSEE shall have the right to terminate this Agreement for any reason upon prior written notice to BROAD. If LICENSEE breaches any provision hereunder, and fails to cure such breach within thirty (30) days, BROAD may terminate this Agreement immediately. Upon termination, LICENSEE shall provide BROAD with written assurance that the original and all copies of the PROGRAM have been destroyed, except that, upon prior written authorization from BROAD, LICENSEE may retain a copy for archive purposes.
* 7.3 Survival. The following provisions shall survive the expiration or termination of this Agreement: Articles 1, 3, 4, 5 and Sections 2.2, 2.3, 7.3, and 7.4.
* 7.4 Notice. Any notices under this Agreement shall be in writing, shall specifically refer to this Agreement, and shall be sent by hand, recognized national overnight courier, confirmed facsimile transmission, confirmed electronic mail, or registered or certified mail, postage prepaid, return receipt requested. All notices under this Agreement shall be deemed effective upon receipt.
* 7.5 Amendment and Waiver; Entire Agreement. This Agreement may be amended, supplemented, or otherwise modified only by means of a written instrument signed by all parties. Any waiver of any rights or failure to act in a specific instance shall relate only to such instance and shall not be construed as an agreement to waive any rights or fail to act in any other instance, whether or not similar. This Agreement constitutes the entire agreement among the parties with respect to its subject matter and supersedes prior agreements or understandings between the parties relating to its subject matter.
* 7.6 Binding Effect; Headings. This Agreement shall be binding upon and inure to the benefit of the parties and their respective permitted successors and assigns. All headings are for convenience only and shall not affect the meaning of any provision of this Agreement.
* 7.7 Governing Law. This Agreement shall be construed, governed, interpreted and applied in accordance with the internal laws of the Commonwealth of Massachusetts, U.S.A., without regard to conflict of laws principles.
*/
package org.broadinstitute.gatk.tools.walkers.genotyper;
import java.util.List;
/**
* Custom priors provided as an array.
*/
public class CustomAFPriorProvider extends AFPriorProvider {
private final double[] priors;
/**
*
* @param priorValues must exactly the number of genomes in the samples (the total ploidy).
*/
public CustomAFPriorProvider(final List<Double> priorValues) {
if (priorValues == null)
throw new IllegalArgumentException("the input prior values cannot be null");
priors = new double[priorValues.size() + 1];
int i = 1;
double sum = 0;
for (double value : priorValues) {
if (value <= 0 || value >= 1)
throw new IllegalArgumentException("the AF prior value "+ value + " is out of the valid interval (0,1)");
if (Double.isNaN(value))
throw new IllegalArgumentException("NaN is not a valid prior AF value");
priors[i++] = Math.log10(value);
sum += value;
}
if (sum >= 1)
throw new IllegalArgumentException("the AF prior value sum must be less than 1: " + sum);
priors[0] = Math.log10(1 - sum);
}
@Override
protected double[] buildPriors(final int totalPloidy) {
if (totalPloidy != priors.length - 1)
throw new IllegalStateException("requesting an invalid prior total ploidy " + totalPloidy + " != " + (priors.length - 1));
return priors;
}
}

55
protected/gatk-tools-protected/src/main/java/org/broadinstitute/gatk/tools/walkers/genotyper/GenotypingEngine.java
View File

@ -94,9 +94,9 @@ public abstract class GenotypingEngine<Config extends StandardCallerArgumentColl
protected final SampleList samples;
private final double[] log10AlleleFrequencyPriorsSNPs;
private final AFPriorProvider log10AlleleFrequencyPriorsSNPs;
private final double[] log10AlleleFrequencyPriorsIndels;
private final AFPriorProvider log10AlleleFrequencyPriorsIndels;
protected final GenomeLocParser genomeLocParser;
@ -133,10 +133,10 @@ public abstract class GenotypingEngine<Config extends StandardCallerArgumentColl
this.samples = samples;
numberOfGenomes = this.samples.sampleCount() * configuration.genotypeArgs.samplePloidy;
MathUtils.Log10Cache.ensureCacheContains(numberOfGenomes * 2);
log10AlleleFrequencyPriorsSNPs = computeAlleleFrequencyPriors(numberOfGenomes,
configuration.genotypeArgs.snpHeterozygosity,configuration.genotypeArgs.inputPrior);
log10AlleleFrequencyPriorsIndels = computeAlleleFrequencyPriors(numberOfGenomes,
configuration.genotypeArgs.indelHeterozygosity,configuration.genotypeArgs.inputPrior);
log10AlleleFrequencyPriorsSNPs = composeAlleleFrequencyPriorProvider(numberOfGenomes,
configuration.genotypeArgs.snpHeterozygosity, configuration.genotypeArgs.inputPrior);
log10AlleleFrequencyPriorsIndels = composeAlleleFrequencyPriorProvider(numberOfGenomes,
configuration.genotypeArgs.indelHeterozygosity, configuration.genotypeArgs.inputPrior);
this.genomeLocParser = genomeLocParser;
}
@ -496,14 +496,25 @@ public abstract class GenotypingEngine<Config extends StandardCallerArgumentColl
return new VariantCallContext(vc, QualityUtils.phredScaleLog10CorrectRate(log10POfRef) >= configuration.genotypeArgs.STANDARD_CONFIDENCE_FOR_CALLING, false);
}
protected final double[] getAlleleFrequencyPriors( final GenotypeLikelihoodsCalculationModel.Model model ) {
/**
* Returns the log10 prior probability for all possible allele counts from 0 to N where N is the total number of
* genomes (total-ploidy).
*
* @param vc the target variant-context, use to determine the total ploidy thus the possible ACs.
* @param defaultPloidy default ploidy to be assume if we do not have the ploidy for some sample in {@code vc}.
* @param model the calculation model (SNP,INDEL or MIXED) whose priors are to be retrieved.
* @throws java.lang.NullPointerException if either {@code vc} or {@code model} is {@code null}
* @return never {@code null}, an array with exactly <code>total-ploidy(vc) + 1</code> positions.
*/
protected final double[] getAlleleFrequencyPriors( final VariantContext vc, final int defaultPloidy, final GenotypeLikelihoodsCalculationModel.Model model ) {
final int totalPloidy = GATKVariantContextUtils.totalPloidy(vc,defaultPloidy);
switch (model) {
case SNP:
case GENERALPLOIDYSNP:
return log10AlleleFrequencyPriorsSNPs;
return log10AlleleFrequencyPriorsSNPs.forTotalPloidy(totalPloidy);
case INDEL:
case GENERALPLOIDYINDEL:
return log10AlleleFrequencyPriorsIndels;
return log10AlleleFrequencyPriorsIndels.forTotalPloidy(totalPloidy);
default:
throw new IllegalArgumentException("Unexpected GenotypeCalculationModel " + model);
}
@ -552,38 +563,20 @@ public abstract class GenotypingEngine<Config extends StandardCallerArgumentColl
*
* @return never {@code null}.
*/
public static double[] computeAlleleFrequencyPriors(final int N, final double heterozygosity, final List<Double> inputPriors) {
private static AFPriorProvider composeAlleleFrequencyPriorProvider(final int N, final double heterozygosity, final List<Double> inputPriors) {
final double[] priors = new double[N + 1];
double sum = 0.0;
final AFPriorProvider result;
if (!inputPriors.isEmpty()) {
// user-specified priors
if (inputPriors.size() != N)
throw new UserException.BadArgumentValue("inputPrior","Invalid length of inputPrior vector: vector length must be equal to # samples +1 ");
int idx = 1;
for (final double prior: inputPriors) {
if (prior < 0.0)
throw new UserException.BadArgumentValue("Bad argument: negative values not allowed","inputPrior");
priors[idx++] = Math.log10(prior);
sum += prior;
}
return new CustomAFPriorProvider(inputPriors);
}
else
// for each i
for (int i = 1; i <= N; i++) {
final double value = heterozygosity / (double)i;
priors[i] = Math.log10(value);
sum += value;
}
// protection against the case of heterozygosity too high or an excessive number of samples (which break population genetics assumptions)
if (sum > 1.0)
throw new UserException.BadArgumentValue("heterozygosity","The heterozygosity value is set too high relative to the number of samples to be processed, or invalid values specified if input priors were provided - try reducing heterozygosity value or correct input priors.");
// null frequency for AF=0 is (1 - sum(all other frequencies))
priors[0] = Math.log10(1.0 - sum);
return priors;
return new HeterozygosityAFPriorProvider(heterozygosity);
}
/**

92
protected/gatk-tools-protected/src/main/java/org/broadinstitute/gatk/tools/walkers/genotyper/HeterozygosityAFPriorProvider.java 100644
View File

@ -0,0 +1,92 @@
/*
* By downloading the PROGRAM you agree to the following terms of use:
*
* BROAD INSTITUTE - SOFTWARE LICENSE AGREEMENT - FOR ACADEMIC NON-COMMERCIAL RESEARCH PURPOSES ONLY
*
* This Agreement is made between the Broad Institute, Inc. with a principal address at 7 Cambridge Center, Cambridge, MA 02142 (BROAD) and the LICENSEE and is effective at the date the downloading is completed (EFFECTIVE DATE).
*
* WHEREAS, LICENSEE desires to license the PROGRAM, as defined hereinafter, and BROAD wishes to have this PROGRAM utilized in the public interest, subject only to the royalty-free, nonexclusive, nontransferable license rights of the United States Government pursuant to 48 CFR 52.227-14; and
* WHEREAS, LICENSEE desires to license the PROGRAM and BROAD desires to grant a license on the following terms and conditions.
* NOW, THEREFORE, in consideration of the promises and covenants made herein, the parties hereto agree as follows:
*
* 1. DEFINITIONS
* 1.1 PROGRAM shall mean copyright in the object code and source code known as GATK2 and related documentation, if any, as they exist on the EFFECTIVE DATE and can be downloaded from http://www.broadinstitute/GATK on the EFFECTIVE DATE.
*
* 2. LICENSE
* 2.1 Grant. Subject to the terms of this Agreement, BROAD hereby grants to LICENSEE, solely for academic non-commercial research purposes, a non-exclusive, non-transferable license to: (a) download, execute and display the PROGRAM and (b) create bug fixes and modify the PROGRAM.
* The LICENSEE may apply the PROGRAM in a pipeline to data owned by users other than the LICENSEE and provide these users the results of the PROGRAM provided LICENSEE does so for academic non-commercial purposes only. For clarification purposes, academic sponsored research is not a commercial use under the terms of this Agreement.
* 2.2 No Sublicensing or Additional Rights. LICENSEE shall not sublicense or distribute the PROGRAM, in whole or in part, without prior written permission from BROAD. LICENSEE shall ensure that all of its users agree to the terms of this Agreement. LICENSEE further agrees that it shall not put the PROGRAM on a network, server, or other similar technology that may be accessed by anyone other than the LICENSEE and its employees and users who have agreed to the terms of this agreement.
* 2.3 License Limitations. Nothing in this Agreement shall be construed to confer any rights upon LICENSEE by implication, estoppel, or otherwise to any computer software, trademark, intellectual property, or patent rights of BROAD, or of any other entity, except as expressly granted herein. LICENSEE agrees that the PROGRAM, in whole or part, shall not be used for any commercial purpose, including without limitation, as the basis of a commercial software or hardware product or to provide services. LICENSEE further agrees that the PROGRAM shall not be copied or otherwise adapted in order to circumvent the need for obtaining a license for use of the PROGRAM.
*
* 3. OWNERSHIP OF INTELLECTUAL PROPERTY
* LICENSEE acknowledges that title to the PROGRAM shall remain with BROAD. The PROGRAM is marked with the following BROAD copyright notice and notice of attribution to contributors. LICENSEE shall retain such notice on all copies. LICENSEE agrees to include appropriate attribution if any results obtained from use of the PROGRAM are included in any publication.
* Copyright 2012 Broad Institute, Inc.
* Notice of attribution: The GATK2 program was made available through the generosity of Medical and Population Genetics program at the Broad Institute, Inc.
* LICENSEE shall not use any trademark or trade name of BROAD, or any variation, adaptation, or abbreviation, of such marks or trade names, or any names of officers, faculty, students, employees, or agents of BROAD except as states above for attribution purposes.
*
* 4. INDEMNIFICATION
* LICENSEE shall indemnify, defend, and hold harmless BROAD, and their respective officers, faculty, students, employees, associated investigators and agents, and their respective successors, heirs and assigns, (Indemnitees), against any liability, damage, loss, or expense (including reasonable attorneys fees and expenses) incurred by or imposed upon any of the Indemnitees in connection with any claims, suits, actions, demands or judgments arising out of any theory of liability (including, without limitation, actions in the form of tort, warranty, or strict liability and regardless of whether such action has any factual basis) pursuant to any right or license granted under this Agreement.
*
* 5. NO REPRESENTATIONS OR WARRANTIES
* THE PROGRAM IS DELIVERED AS IS. BROAD MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND CONCERNING THE PROGRAM OR THE COPYRIGHT, EXPRESS OR IMPLIED, INCLUDING, WITHOUT LIMITATION, WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, NONINFRINGEMENT, OR THE ABSENCE OF LATENT OR OTHER DEFECTS, WHETHER OR NOT DISCOVERABLE. BROAD EXTENDS NO WARRANTIES OF ANY KIND AS TO PROGRAM CONFORMITY WITH WHATEVER USER MANUALS OR OTHER LITERATURE MAY BE ISSUED FROM TIME TO TIME.
* IN NO EVENT SHALL BROAD OR ITS RESPECTIVE DIRECTORS, OFFICERS, EMPLOYEES, AFFILIATED INVESTIGATORS AND AFFILIATES BE LIABLE FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES OF ANY KIND, INCLUDING, WITHOUT LIMITATION, ECONOMIC DAMAGES OR INJURY TO PROPERTY AND LOST PROFITS, REGARDLESS OF WHETHER BROAD SHALL BE ADVISED, SHALL HAVE OTHER REASON TO KNOW, OR IN FACT SHALL KNOW OF THE POSSIBILITY OF THE FOREGOING.
*
* 6. ASSIGNMENT
* This Agreement is personal to LICENSEE and any rights or obligations assigned by LICENSEE without the prior written consent of BROAD shall be null and void.
*
* 7. MISCELLANEOUS
* 7.1 Export Control. LICENSEE gives assurance that it will comply with all United States export control laws and regulations controlling the export of the PROGRAM, including, without limitation, all Export Administration Regulations of the United States Department of Commerce. Among other things, these laws and regulations prohibit, or require a license for, the export of certain types of software to specified countries.
* 7.2 Termination. LICENSEE shall have the right to terminate this Agreement for any reason upon prior written notice to BROAD. If LICENSEE breaches any provision hereunder, and fails to cure such breach within thirty (30) days, BROAD may terminate this Agreement immediately. Upon termination, LICENSEE shall provide BROAD with written assurance that the original and all copies of the PROGRAM have been destroyed, except that, upon prior written authorization from BROAD, LICENSEE may retain a copy for archive purposes.
* 7.3 Survival. The following provisions shall survive the expiration or termination of this Agreement: Articles 1, 3, 4, 5 and Sections 2.2, 2.3, 7.3, and 7.4.
* 7.4 Notice. Any notices under this Agreement shall be in writing, shall specifically refer to this Agreement, and shall be sent by hand, recognized national overnight courier, confirmed facsimile transmission, confirmed electronic mail, or registered or certified mail, postage prepaid, return receipt requested. All notices under this Agreement shall be deemed effective upon receipt.
* 7.5 Amendment and Waiver; Entire Agreement. This Agreement may be amended, supplemented, or otherwise modified only by means of a written instrument signed by all parties. Any waiver of any rights or failure to act in a specific instance shall relate only to such instance and shall not be construed as an agreement to waive any rights or fail to act in any other instance, whether or not similar. This Agreement constitutes the entire agreement among the parties with respect to its subject matter and supersedes prior agreements or understandings between the parties relating to its subject matter.
* 7.6 Binding Effect; Headings. This Agreement shall be binding upon and inure to the benefit of the parties and their respective permitted successors and assigns. All headings are for convenience only and shall not affect the meaning of any provision of this Agreement.
* 7.7 Governing Law. This Agreement shall be construed, governed, interpreted and applied in accordance with the internal laws of the Commonwealth of Massachusetts, U.S.A., without regard to conflict of laws principles.
*/
package org.broadinstitute.gatk.tools.walkers.genotyper;
import org.broadinstitute.gatk.utils.MathUtils;
import java.util.Arrays;
/**
* Allele frequency prior provider based on heterozygosity.
*
* @author Valentin Ruano-Rubio &lt;valentin@broadinstitute.org&gt;
*/
public class HeterozygosityAFPriorProvider extends AFPriorProvider {
private final double heterozygosity;
private final double log10Heterozygosity;
/**
* Construct a new provider given the heterozygosity value.
* @param heterozygosity must be a valid heterozygosity between larger than 0 and smaller than 1.
* @throws IllegalArgumentException if {@code heterozygosity} is not valid one in the interval (0,1).
*/
public HeterozygosityAFPriorProvider(final double heterozygosity) {
if (heterozygosity <= 0)
throw new IllegalArgumentException("the heterozygosity must be greater than 0");
if (heterozygosity >= 1)
throw new IllegalArgumentException("the heterozygosity must be less than 1");
if (Double.isNaN(heterozygosity))
throw new IllegalArgumentException("the heterozygosity cannot be a NaN");
this.heterozygosity = heterozygosity;
this.log10Heterozygosity = Math.log10(heterozygosity);
}
@Override
protected double[] buildPriors(final int totalPloidy) {
final double[] result = new double [totalPloidy + 1];
Arrays.fill(result, log10Heterozygosity);
result[0] = Double.NEGATIVE_INFINITY;
MathUtils.Log10Cache.ensureCacheContains(totalPloidy);
for (int i = 1; i <= totalPloidy; i++)
result[i] -= MathUtils.Log10Cache.get(i);
final double log10Sum = MathUtils.approximateLog10SumLog10(result);
if (log10Sum >= 0)
throw new IllegalArgumentException("heterosygosity " + heterozygosity + " is too large of total ploidy " + totalPloidy);
result[0] = MathUtils.log10OneMinusPow10(log10Sum);
return result;
}
}