From 9a29d6d4be4b929313a5e59b39c02f6a57d10daa Mon Sep 17 00:00:00 2001 From: Mark DePristo Date: Mon, 11 Feb 2013 10:42:37 -0800 Subject: [PATCH] Fix an catastrophic bug (WoW!) in the reference calculation of the UG -- The UG was using MathUtils binomial probability backward, so that the estimated confidence was always NaN, and was as a side effect other utils converted this to a meaningless 0.0. This is all because there wasn't a unit test. -- I've fixed the calculation, so it's now log10 based, uses robust MathUtils and QualityUtils functions to compute probabilities, and added a unit test. --- .../genotyper/UnifiedGenotyperEngine.java | 49 +++++--- .../UnifiedGenotyperEngineUnitTest.java | 105 ++++++++++++++++++ .../sting/utils/QualityUtils.java | 18 ++- 3 files changed, 154 insertions(+), 18 deletions(-) create mode 100644 protected/java/test/org/broadinstitute/sting/gatk/walkers/genotyper/UnifiedGenotyperEngineUnitTest.java diff --git a/protected/java/src/org/broadinstitute/sting/gatk/walkers/genotyper/UnifiedGenotyperEngine.java b/protected/java/src/org/broadinstitute/sting/gatk/walkers/genotyper/UnifiedGenotyperEngine.java index 4cfd8c7bc..ede0741ff 100644 --- a/protected/java/src/org/broadinstitute/sting/gatk/walkers/genotyper/UnifiedGenotyperEngine.java +++ b/protected/java/src/org/broadinstitute/sting/gatk/walkers/genotyper/UnifiedGenotyperEngine.java @@ -46,6 +46,7 @@ package org.broadinstitute.sting.gatk.walkers.genotyper; +import com.google.java.contract.Ensures; import com.google.java.contract.Requires; import org.apache.log4j.Logger; import org.broadinstitute.sting.commandline.RodBinding; @@ -138,6 +139,10 @@ public class UnifiedGenotyperEngine { this(toolkit, UAC, Logger.getLogger(UnifiedGenotyperEngine.class), null, null, SampleUtils.getSAMFileSamples(toolkit.getSAMFileHeader()), GATKVariantContextUtils.DEFAULT_PLOIDY); } + protected UnifiedGenotyperEngine(GenomeAnalysisEngine toolkit, Set samples, UnifiedArgumentCollection UAC) { + this(toolkit, UAC, Logger.getLogger(UnifiedGenotyperEngine.class), null, null, samples, GATKVariantContextUtils.DEFAULT_PLOIDY); + } + @Requires({"toolkit != null", "UAC != null", "logger != null", "samples != null && samples.size() > 0","ploidy>0"}) public UnifiedGenotyperEngine(GenomeAnalysisEngine toolkit, UnifiedArgumentCollection UAC, Logger logger, PrintStream verboseWriter, VariantAnnotatorEngine engine, Set samples, int ploidy) { this.BAQEnabledOnCMDLine = toolkit.getArguments().BAQMode != BAQ.CalculationMode.OFF; @@ -577,43 +582,53 @@ public class UnifiedGenotyperEngine { } private final static double[] binomialProbabilityDepthCache = new double[10000]; + private final static double REF_BINOMIAL_PROB_LOG10_0_5 = Math.log10(0.5); + static { for ( int i = 1; i < binomialProbabilityDepthCache.length; i++ ) { - binomialProbabilityDepthCache[i] = MathUtils.binomialProbability(0, i, 0.5); + binomialProbabilityDepthCache[i] = MathUtils.log10BinomialProbability(i, 0, REF_BINOMIAL_PROB_LOG10_0_5); } } - private final double getRefBinomialProb(final int depth) { + private final double getRefBinomialProbLog10(final int depth) { if ( depth < binomialProbabilityDepthCache.length ) return binomialProbabilityDepthCache[depth]; else - return MathUtils.binomialProbability(0, depth, 0.5); + return MathUtils.log10BinomialProbability(depth, 0, REF_BINOMIAL_PROB_LOG10_0_5); } - private VariantCallContext estimateReferenceConfidence(VariantContext vc, Map contexts, double theta, boolean ignoreCoveredSamples, double initialPofRef) { if ( contexts == null ) return null; - double P_of_ref = initialPofRef; + double log10POfRef = Math.log10(initialPofRef); // for each sample that we haven't examined yet for ( String sample : samples ) { - boolean isCovered = contexts.containsKey(sample); - if ( ignoreCoveredSamples && isCovered ) + final AlignmentContext context = contexts.get(sample); + if ( ignoreCoveredSamples && context != null ) continue; - - - int depth = 0; - - if ( isCovered ) { - depth = contexts.get(sample).getBasePileup().depthOfCoverage(); - } - - P_of_ref *= 1.0 - (theta / 2.0) * getRefBinomialProb(depth); + final int depth = context == null ? 0 : context.getBasePileup().depthOfCoverage(); + log10POfRef += estimateLog10ReferenceConfidenceForOneSample(depth, theta); } - return new VariantCallContext(vc, QualityUtils.phredScaleCorrectRate(P_of_ref) >= UAC.STANDARD_CONFIDENCE_FOR_CALLING, false); + return new VariantCallContext(vc, QualityUtils.phredScaleLog10CorrectRate(log10POfRef) >= UAC.STANDARD_CONFIDENCE_FOR_CALLING, false); + } + + /** + * Compute the log10 probability of a sample with sequencing depth and no alt allele is actually truly homozygous reference + * + * Assumes the sample is diploid + * + * @param depth the depth of the sample + * @param theta the heterozygosity of this species (between 0 and 1) + * @return a valid log10 probability of the sample being hom-ref + */ + @Requires({"depth >= 0", "theta >= 0.0 && theta <= 1.0"}) + @Ensures("MathUtils.goodLog10Probability(result)") + protected double estimateLog10ReferenceConfidenceForOneSample(final int depth, final double theta) { + final double log10PofNonRef = Math.log10(theta / 2.0) + getRefBinomialProbLog10(depth); + return MathUtils.log10OneMinusX(Math.pow(10.0, log10PofNonRef)); } protected void printVerboseData(String pos, VariantContext vc, double PofF, double phredScaledConfidence, final GenotypeLikelihoodsCalculationModel.Model model) { diff --git a/protected/java/test/org/broadinstitute/sting/gatk/walkers/genotyper/UnifiedGenotyperEngineUnitTest.java b/protected/java/test/org/broadinstitute/sting/gatk/walkers/genotyper/UnifiedGenotyperEngineUnitTest.java new file mode 100644 index 000000000..23596db83 --- /dev/null +++ b/protected/java/test/org/broadinstitute/sting/gatk/walkers/genotyper/UnifiedGenotyperEngineUnitTest.java @@ -0,0 +1,105 @@ +/* +* 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.sting.gatk.walkers.genotyper; + + +// the imports for unit testing. + + +import org.broadinstitute.sting.BaseTest; +import org.broadinstitute.sting.gatk.GenomeAnalysisEngine; +import org.broadinstitute.sting.gatk.arguments.GATKArgumentCollection; +import org.broadinstitute.sting.utils.MathUtils; +import org.testng.Assert; +import org.testng.annotations.BeforeClass; +import org.testng.annotations.BeforeMethod; +import org.testng.annotations.DataProvider; +import org.testng.annotations.Test; + +import java.util.*; + +public class UnifiedGenotyperEngineUnitTest extends BaseTest { + private final static double TOLERANCE = 1e-5; + private UnifiedGenotyperEngine ugEngine; + + @BeforeClass + public void setUp() throws Exception { + final GenomeAnalysisEngine engine = new GenomeAnalysisEngine(); + engine.setArguments(new GATKArgumentCollection()); + final UnifiedArgumentCollection args = new UnifiedArgumentCollection(); + final Set fakeSamples = Collections.singleton("fake"); + ugEngine = new UnifiedGenotyperEngine(engine, fakeSamples, args); + } + + private UnifiedGenotyperEngine getEngine() { + return ugEngine; + } + + @DataProvider(name = "ReferenceQualityCalculation") + public Object[][] makeReferenceQualityCalculation() { + List tests = new ArrayList(); + + // this functionality can be adapted to provide input data for whatever you might want in your data + final double p = Math.log10(0.5); + for ( final double theta : Arrays.asList(0.1, 0.01, 0.001) ) { + for ( final int depth : Arrays.asList(0, 1, 2, 10, 100, 1000, 10000) ) { + final double log10PofNonRef = Math.log10(theta / 2.0) + MathUtils.log10BinomialProbability(depth, 0, p); + final double log10POfRef = MathUtils.log10OneMinusX(Math.pow(10.0, log10PofNonRef)); + tests.add(new Object[]{depth, theta, log10POfRef}); + } + } + + return tests.toArray(new Object[][]{}); + } + + @Test(dataProvider = "ReferenceQualityCalculation") + public void testReferenceQualityCalculation(final int depth, final double theta, final double expected) { + final double ref = getEngine().estimateLog10ReferenceConfidenceForOneSample(depth, theta); + Assert.assertTrue(MathUtils.goodLog10Probability(ref), "Reference calculation wasn't a well formed log10 prob " + ref); + Assert.assertEquals(ref, expected, TOLERANCE, "Failed reference confidence for single sample"); + } +} \ No newline at end of file diff --git a/public/java/src/org/broadinstitute/sting/utils/QualityUtils.java b/public/java/src/org/broadinstitute/sting/utils/QualityUtils.java index 9dd9b735d..a7552ca9c 100644 --- a/public/java/src/org/broadinstitute/sting/utils/QualityUtils.java +++ b/public/java/src/org/broadinstitute/sting/utils/QualityUtils.java @@ -311,6 +311,21 @@ public class QualityUtils { return phredScaleLog10ErrorRate(MathUtils.log10OneMinusX(trueRate)); } + /** + * Convert a log10 probability of being right to a phred-scaled quality score of being wrong as a double + * + * This is a very generic method, that simply computes a phred-scaled double quality + * score given an error rate. It has the same precision as a normal double operation + * + * @param trueRateLog10 the probability of being right (0.0-1.0) + * @return a phred-scaled version of the error rate implied by trueRate + */ + @Requires("MathUtils.goodLog10Probability(trueRateLog10)") + @Ensures("result >= 0.0") + public static double phredScaleLog10CorrectRate(final double trueRateLog10) { + return phredScaleCorrectRate(Math.pow(10.0, trueRateLog10)); + } + /** * Convert a probability of being wrong to a phred-scaled quality score of being wrong as a double * @@ -337,7 +352,8 @@ public class QualityUtils { */ @Ensures("result >= 0.0") public static double phredScaleLog10ErrorRate(final double errorRateLog10) { - return -10.0 * Math.max(errorRateLog10, RAW_MIN_PHRED_SCALED_QUAL); + // abs is necessary for edge base with errorRateLog10 = 0 producing -0.0 doubles + return Math.abs(-10.0 * Math.max(errorRateLog10, RAW_MIN_PHRED_SCALED_QUAL)); } // ----------------------------------------------------------------------