From f726d8130a076202cf2246a9fdb5123e0e91af30 Mon Sep 17 00:00:00 2001 From: Mark DePristo Date: Thu, 20 Jun 2013 15:15:17 -0400 Subject: [PATCH 1/2] VariantRecalibrator bugfix for bad log10sumlog10 values -- The VR, when the model is bad, may evaluate log10sumlog10 where some of the values in the vector are NaN. This case is now trapped in VR and handled as previously -- indicating that the model has failed and evaluation continues. --- .../GaussianMixtureModel.java | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/protected/java/src/org/broadinstitute/sting/gatk/walkers/variantrecalibration/GaussianMixtureModel.java b/protected/java/src/org/broadinstitute/sting/gatk/walkers/variantrecalibration/GaussianMixtureModel.java index 92b0d4df2..efc24d5f9 100644 --- a/protected/java/src/org/broadinstitute/sting/gatk/walkers/variantrecalibration/GaussianMixtureModel.java +++ b/protected/java/src/org/broadinstitute/sting/gatk/walkers/variantrecalibration/GaussianMixtureModel.java @@ -225,6 +225,20 @@ public class GaussianMixtureModel { isModelReadyForEvaluation = true; } + /** + * A version of Log10SumLog10 that tolerates NaN values in the array + * + * In the case where one or more of the values are NaN, this function returns NaN + * + * @param values a non-null vector of doubles + * @return log10 of the sum of the log10 values, or NaN + */ + private double nanTolerantLog10SumLog10(final double[] values) { + for ( final double value : values ) + if ( Double.isNaN(value) ) return Double.NaN; + return MathUtils.log10sumLog10(values); + } + public double evaluateDatum( final VariantDatum datum ) { for( final boolean isNull : datum.isNull ) { if( isNull ) { return evaluateDatumMarginalized( datum ); } @@ -235,7 +249,7 @@ public class GaussianMixtureModel { for( final MultivariateGaussian gaussian : gaussians ) { pVarInGaussianLog10[gaussianIndex++] = gaussian.pMixtureLog10 + gaussian.evaluateDatumLog10( datum ); } - return MathUtils.log10sumLog10(pVarInGaussianLog10); // Sum(pi_k * p(v|n,k)) + return nanTolerantLog10SumLog10(pVarInGaussianLog10); // Sum(pi_k * p(v|n,k)) } // Used only to decide which covariate dimension is most divergent in order to report in the culprit info field annotation @@ -247,7 +261,7 @@ public class GaussianMixtureModel { for( final MultivariateGaussian gaussian : gaussians ) { pVarInGaussianLog10[gaussianIndex++] = gaussian.pMixtureLog10 + MathUtils.normalDistributionLog10(gaussian.mu[iii], gaussian.sigma.get(iii, iii), datum.annotations[iii]); } - return MathUtils.log10sumLog10(pVarInGaussianLog10); // Sum(pi_k * p(v|n,k)) + return nanTolerantLog10SumLog10(pVarInGaussianLog10); // Sum(pi_k * p(v|n,k)) } public double evaluateDatumMarginalized( final VariantDatum datum ) { From 8caf39cb657ac341539eb860ddf5507a427164b5 Mon Sep 17 00:00:00 2001 From: Mark DePristo Date: Thu, 20 Jun 2013 12:57:14 -0400 Subject: [PATCH 2/2] Experimental LikelihoodRankSum annotation -- Added experimental LikelihoodRankSum, which required slightly more detailed access to the information managed by the base class, so added an overloaded getElementForRead also provides access to the MostLikelyAllele class -- Added base class default implementation of getElementForPileupElement() which returns null, indicating that the pileup version isn't supported. -- Added @Override to many of the RankSum classes for safety's sake -- Updates to GeneralCallingPipeline: annotate sites with dbSNP IDs, -- R script to assess the value of annotations for VQSR --- .../annotator/BaseQualityRankSumTest.java | 4 + .../annotator/ClippingRankSumTest.java | 9 +-- .../annotator/LikelihoodRankSumTest.java | 79 +++++++++++++++++++ .../annotator/MappingQualityRankSumTest.java | 5 +- .../gatk/walkers/annotator/RankSumTest.java | 21 ++++- .../walkers/annotator/ReadPosRankSumTest.java | 8 +- 6 files changed, 114 insertions(+), 12 deletions(-) create mode 100644 protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/LikelihoodRankSumTest.java diff --git a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/BaseQualityRankSumTest.java b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/BaseQualityRankSumTest.java index 534834d0e..9ba468191 100644 --- a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/BaseQualityRankSumTest.java +++ b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/BaseQualityRankSumTest.java @@ -65,14 +65,18 @@ import java.util.*; *

The base quality rank sum test can not be calculated for sites without a mixture of reads showing both the reference and alternate alleles.

*/ public class BaseQualityRankSumTest extends RankSumTest implements StandardAnnotation { + @Override public List getKeyNames() { return Arrays.asList("BaseQRankSum"); } + @Override public List getDescriptions() { return Arrays.asList(new VCFInfoHeaderLine("BaseQRankSum", 1, VCFHeaderLineType.Float, "Z-score from Wilcoxon rank sum test of Alt Vs. Ref base qualities")); } + @Override protected Double getElementForRead(final GATKSAMRecord read, final int refLoc) { return (double)read.getBaseQualities()[ReadUtils.getReadCoordinateForReferenceCoordinateUpToEndOfRead(read, refLoc, ReadUtils.ClippingTail.RIGHT_TAIL)]; } + @Override protected Double getElementForPileupElement(final PileupElement p) { return (double)p.getQual(); } diff --git a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/ClippingRankSumTest.java b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/ClippingRankSumTest.java index 68e983bb8..eaa9df128 100644 --- a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/ClippingRankSumTest.java +++ b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/ClippingRankSumTest.java @@ -66,17 +66,14 @@ import java.util.*; * @since 6/28/12 */ public class ClippingRankSumTest extends RankSumTest { - + @Override public List getKeyNames() { return Arrays.asList("ClippingRankSum"); } + @Override public List getDescriptions() { return Arrays.asList(new VCFInfoHeaderLine("ClippingRankSum", 1, VCFHeaderLineType.Float, "Z-score From Wilcoxon rank sum test of Alt vs. Ref number of hard clipped bases")); } + @Override protected Double getElementForRead(final GATKSAMRecord read, final int refLoc) { return (double)AlignmentUtils.getNumHardClippedBases(read); } - - protected Double getElementForPileupElement(final PileupElement p) { - // TODO - we only support the non-pileup case for now, e.g. an active-region based version - return null; - } } diff --git a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/LikelihoodRankSumTest.java b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/LikelihoodRankSumTest.java new file mode 100644 index 000000000..c7fff5a7f --- /dev/null +++ b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/LikelihoodRankSumTest.java @@ -0,0 +1,79 @@ +/* +* 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.annotator; + +import org.broadinstitute.sting.utils.genotyper.MostLikelyAllele; +import org.broadinstitute.sting.utils.sam.GATKSAMRecord; +import org.broadinstitute.variant.vcf.VCFHeaderLineType; +import org.broadinstitute.variant.vcf.VCFInfoHeaderLine; + +import java.util.Arrays; +import java.util.List; + +/** + * U-based z-approximation from the Mann-Whitney Rank Sum Test contrasting the likelihoods of reads to their + * most likely haplotypes. This is effectively testing for a differentiate quality in the modeling of the alt + * allele than the reference allele. + */ +public class LikelihoodRankSumTest extends RankSumTest { + @Override + public List getKeyNames() { return Arrays.asList("LikelihoodRankSum"); } + + @Override + public List getDescriptions() { return Arrays.asList(new VCFInfoHeaderLine("LikelihoodRankSum", 1, VCFHeaderLineType.Float, "Z-score from Wilcoxon rank sum test of Alt Vs. Ref haplotype likelihoods")); } + + @Override + protected Double getElementForRead(final GATKSAMRecord read, final int refLoc, final MostLikelyAllele mostLikelyAllele) { + if ( ! mostLikelyAllele.isInformative() ) throw new IllegalStateException("Should never have seen non-informative read " + read + " MostLikelyAllele " + mostLikelyAllele); + return mostLikelyAllele.getLog10LikelihoodOfMostLikely(); + } + + @Override + protected Double getElementForRead(GATKSAMRecord read, int refLoc) { + throw new IllegalStateException("This method should never have been called as getElementForRead(read,refloc,mostLikelyAllele) was overloaded"); + } +} \ No newline at end of file diff --git a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/MappingQualityRankSumTest.java b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/MappingQualityRankSumTest.java index 0ebb09961..b2a504eb2 100644 --- a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/MappingQualityRankSumTest.java +++ b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/MappingQualityRankSumTest.java @@ -64,15 +64,18 @@ import java.util.*; *

The mapping quality rank sum test can not be calculated for sites without a mixture of reads showing both the reference and alternate alleles.

*/ public class MappingQualityRankSumTest extends RankSumTest implements StandardAnnotation { - + @Override public List getKeyNames() { return Arrays.asList("MQRankSum"); } + @Override public List getDescriptions() { return Arrays.asList(new VCFInfoHeaderLine("MQRankSum", 1, VCFHeaderLineType.Float, "Z-score From Wilcoxon rank sum test of Alt vs. Ref read mapping qualities")); } + @Override protected Double getElementForRead(final GATKSAMRecord read, final int refLoc) { return (double)read.getMappingQuality(); } + @Override protected Double getElementForPileupElement(final PileupElement p) { return (double)p.getRead().getMappingQuality(); } diff --git a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/RankSumTest.java b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/RankSumTest.java index 37508fc06..1ba13afa1 100644 --- a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/RankSumTest.java +++ b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/RankSumTest.java @@ -176,7 +176,7 @@ public abstract class RankSumTest extends InfoFieldAnnotation implements ActiveR final GATKSAMRecord read = el.getKey(); if ( isUsableRead(read, refLoc) ) { - final Double value = getElementForRead(read, refLoc); + final Double value = getElementForRead(read, refLoc, a); if ( value == null ) continue; @@ -188,6 +188,18 @@ public abstract class RankSumTest extends InfoFieldAnnotation implements ActiveR } } + /** + * Get the element for the given read at the given reference position + * + * @param read the read + * @param refLoc the reference position + * @param mostLikelyAllele the most likely allele for this read + * @return a Double representing the element to be used in the rank sum test, or null if it should not be used + */ + protected Double getElementForRead(final GATKSAMRecord read, final int refLoc, final MostLikelyAllele mostLikelyAllele) { + return getElementForRead(read, refLoc); + } + /** * Get the element for the given read at the given reference position * @@ -202,10 +214,15 @@ public abstract class RankSumTest extends InfoFieldAnnotation implements ActiveR /** * Get the element for the given read at the given reference position * + * By default this function returns null, indicating that the test doesn't support the old style of pileup calculations + * * @param p the pileup element * @return a Double representing the element to be used in the rank sum test, or null if it should not be used */ - protected abstract Double getElementForPileupElement(final PileupElement p); + protected Double getElementForPileupElement(final PileupElement p) { + // does not work in pileup mode + return null; + } /** * Can the base in this pileup element be used in comparative tests between ref / alt bases? diff --git a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/ReadPosRankSumTest.java b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/ReadPosRankSumTest.java index 37faaed22..f4528b16f 100644 --- a/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/ReadPosRankSumTest.java +++ b/protected/java/src/org/broadinstitute/sting/gatk/walkers/annotator/ReadPosRankSumTest.java @@ -71,14 +71,15 @@ import java.util.*; */ public class ReadPosRankSumTest extends RankSumTest implements StandardAnnotation { - public List getKeyNames() { - return Arrays.asList("ReadPosRankSum"); - } + @Override + public List getKeyNames() { return Arrays.asList("ReadPosRankSum"); } + @Override public List getDescriptions() { return Arrays.asList(new VCFInfoHeaderLine("ReadPosRankSum", 1, VCFHeaderLineType.Float, "Z-score from Wilcoxon rank sum test of Alt vs. Ref read position bias")); } + @Override protected Double getElementForRead(final GATKSAMRecord read, final int refLoc) { final int offset = ReadUtils.getReadCoordinateForReferenceCoordinate( read.getSoftStart(), read.getCigar(), refLoc, ReadUtils.ClippingTail.RIGHT_TAIL, true ); if ( offset == ReadUtils.CLIPPING_GOAL_NOT_REACHED ) @@ -91,6 +92,7 @@ public class ReadPosRankSumTest extends RankSumTest implements StandardAnnotatio return (double)readPos; } + @Override protected Double getElementForPileupElement(final PileupElement p) { final int offset = AlignmentUtils.calcAlignmentByteArrayOffset(p.getRead().getCigar(), p, 0, 0); return (double)getFinalReadPosition(p.getRead(), offset);