Retreived PairHMM benchmarks from archive and made improvements
PairHMMSyntheticBenchmark and PairHMMEmpirical benchmark were written to test the banded pairHMM, and were archived along with it. I returned them to the test directory for use in benchmarking the ArrayLoglessPairHMM. I commented out references to the banded pairHMM (which was left in archive), rather than removing those references entirely. Renamed PairHMMEmpiricalBenchmark to PairHMMBandedEmpiricalBenchmark and returned it to the archive. It has a few problems for use as a general benchmark, including initializing the HMM too frequently and doing too much setup work in the 'time' method. However, since the size selection and debug printing are useful for testing the banded implementation, I decided to keep it as-is and archive it alongside with the other banded pairHMM classes. I did fix one bug that was causing the selectWorkingData function to return prematurely. As a result, the benchmark was only evaluating 4-40 pairHMM calls instead of the desired "maxRecords". I wrote a new PairHMMEmpiricalBenchmark that simply works through a list of data, with setup work and hmm-initialization moved to its own function. This involved writing a new data read-in function in PairHMMTestData. The original was not maintaining the input data in order, the end result of which would be an over-estimate of how much caching we are able to do. The new benchmark class more closely mirrors real-world operation over large data. It might be cleaner to fix some of the issues with the BandedEmpiricalBenchmark and use one read-in function. However, this would involve more extensive changes to: PairHMMBandedEmpiricalBenchmark PairHMMTestData BandedLoglessPairHMMUnitTest I decided against this as the banded benchmark and unit test are archived.
This commit is contained in:
bradtaylor
parent
86fe9fae76
commit
0435bbe38f
|
|
@ -0,0 +1,115 @@
|
|||
/*
|
||||
* 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.utils.pairhmm;
|
||||
|
||||
import com.google.caliper.Param;
|
||||
import com.google.caliper.SimpleBenchmark;
|
||||
|
||||
import java.io.File;
|
||||
import java.util.LinkedList;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.lang.Math;
|
||||
|
||||
/**
|
||||
* Caliper microbenchmark for empirical test data for PairHMM
|
||||
*/
|
||||
public class PairHMMEmpiricalBenchmark extends SimpleBenchmark {
|
||||
@Param ({"array_logless", "logless"})
|
||||
String algorithm;
|
||||
|
||||
@Param({"likelihoods_NA12878_HiSeqWGS_chr20_1mb.txt"})
|
||||
String likelihoodsFile;
|
||||
|
||||
@Param({"1000","10000","70000"})
|
||||
int records;
|
||||
|
||||
PairHMM hmm =null;
|
||||
|
||||
List<PairHMMTestData> empiricalData = new LinkedList<>();
|
||||
List<PairHMMTestData> workingData = new LinkedList<>();
|
||||
|
||||
|
||||
@Override
|
||||
protected void setUp() throws Exception {
|
||||
empiricalData = PairHMMTestData.readLikelihoodsInOrder(new File(likelihoodsFile));
|
||||
records = Math.min(records, empiricalData.size());
|
||||
workingData = empiricalData.subList(0,records);
|
||||
|
||||
int maxReadLength = PairHMMTestData.calcMaxReadLen(workingData);
|
||||
int maxHaplotypeLength = PairHMMTestData.calcMaxHaplotypeLen(workingData);
|
||||
|
||||
hmm = getHmm();
|
||||
hmm.initialize(maxReadLength,maxHaplotypeLength);
|
||||
}
|
||||
|
||||
private PairHMM getHmm() {
|
||||
switch (algorithm) {
|
||||
case "logless": return new LoglessPairHMM();
|
||||
case "array_logless": return new ArrayLoglessPairHMM();
|
||||
default: throw new IllegalStateException("Unexpected algorithm " + algorithm);
|
||||
}
|
||||
}
|
||||
|
||||
public double timeHMM(int rep){
|
||||
double result = 0;
|
||||
for (int i = 0; i < rep; i++) {
|
||||
for (final PairHMMTestData datum : workingData){
|
||||
result += hmm.computeReadLikelihoodGivenHaplotypeLog10(datum.ref.getBytes(),
|
||||
datum.getRead().getBytes(),
|
||||
datum.baseQuals,
|
||||
datum.insQuals,
|
||||
datum.delQuals,
|
||||
datum.gcp,
|
||||
datum.newRead,
|
||||
datum.nextRef.getBytes());
|
||||
|
||||
}
|
||||
}
|
||||
return result;
|
||||
}
|
||||
}
|
||||
|
|
@ -0,0 +1,127 @@
|
|||
/*
|
||||
* 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.utils.pairhmm;
|
||||
|
||||
import com.google.caliper.Param;
|
||||
import com.google.caliper.SimpleBenchmark;
|
||||
import net.sf.samtools.Cigar;
|
||||
import net.sf.samtools.CigarElement;
|
||||
import net.sf.samtools.TextCigarCodec;
|
||||
import org.broadinstitute.sting.utils.Utils;
|
||||
import org.broadinstitute.sting.utils.sam.AlignmentUtils;
|
||||
|
||||
import java.util.*;
|
||||
|
||||
/**
|
||||
* Caliper microbenchmark for synthetic test data for PairHMM
|
||||
*/
|
||||
public class PairHMMSyntheticBenchmark extends SimpleBenchmark {
|
||||
@Param ({"array_logless", "logless"})
|
||||
// @Param({"logless", "array_logless"})
|
||||
// @Param({"logless", "banded_w5_mle10", "banded_w5_mle20"})
|
||||
// @Param({"logless", "banded_w10_mle20", "banded_w5_mle20", "banded_w5_mle10"})
|
||||
String algorithm;
|
||||
|
||||
// @Param({"40", "100", "200", "300", "500"})
|
||||
@Param({"40", "300"})
|
||||
// @Param({"300"})
|
||||
int refLength;
|
||||
|
||||
// @Param({"200"})
|
||||
@Param({"40", "101", "200"})
|
||||
// @Param({"40", "100", "200", "300", "500"})
|
||||
int readLength;
|
||||
|
||||
private PairHMM hmm;
|
||||
private PairHMMTestData testData;
|
||||
|
||||
|
||||
@Override
|
||||
protected void setUp() throws Exception {
|
||||
hmm = getHmm();
|
||||
final String ref = generateSeq(refLength);
|
||||
final String nextRef = generateSeq(refLength);
|
||||
final String read = generateSeq(readLength);
|
||||
testData = new PairHMMTestData(ref, nextRef, read, (byte)30);
|
||||
System.out.println(testData.toString());
|
||||
}
|
||||
|
||||
private PairHMM getHmm() {
|
||||
switch (algorithm) {
|
||||
case "logless": return new LoglessPairHMM();
|
||||
case "array_logless": return new ArrayLoglessPairHMM();
|
||||
// case "banded_w10_mle20": return new BandedLoglessPairHMM(10, 1e-20);
|
||||
// case "banded_w5_mle20": return new BandedLoglessPairHMM(5, 1e-20);
|
||||
// case "banded_w5_mle10": return new BandedLoglessPairHMM(5, 1e-10);
|
||||
default: throw new IllegalStateException("Unexpected algorithm " + algorithm);
|
||||
}
|
||||
}
|
||||
|
||||
private String generateSeq(final int len) {
|
||||
final List<String> root = Arrays.asList("A", "C", "G", "T");
|
||||
|
||||
String seq = "";
|
||||
for ( int i = 0; true; i++ ) {
|
||||
final String base = root.get(i % root.size());
|
||||
final int copies = i / root.size() + 1;
|
||||
seq += Utils.dupString(base, copies);
|
||||
if ( seq.length() >= len )
|
||||
return seq.substring(0, len);
|
||||
}
|
||||
}
|
||||
|
||||
public void timePairHMM(int rep) {
|
||||
for ( int i = 0; i < rep; i++ ) {
|
||||
testData.runHMM(hmm);
|
||||
}
|
||||
// if ( hmm instanceof BandedLoglessPairHMM ) {
|
||||
// final BandedLoglessPairHMM banded = (BandedLoglessPairHMM)hmm;
|
||||
// System.out.printf("Banded n cells possible : %d%n", banded.nCellsOverall);
|
||||
// System.out.printf("Banded n cells evaluated : %d%n", banded.nCellsEvaluated);
|
||||
// }
|
||||
}
|
||||
}
|
||||
|
|
@ -71,8 +71,9 @@ public class PairHMMTestData {
|
|||
private final String read;
|
||||
public final byte[] baseQuals, insQuals, delQuals, gcp;
|
||||
public final double log10l;
|
||||
public final boolean newRead;
|
||||
|
||||
PairHMMTestData(String ref, String nextRef, String read, byte[] baseQuals, byte[] insQuals, byte[] delQuals, byte[] gcp, double log10l) {
|
||||
PairHMMTestData(String ref, String nextRef, String read, byte[] baseQuals, byte[] insQuals, byte[] delQuals, byte[] gcp, double log10l, boolean newRead) {
|
||||
this.ref = ref;
|
||||
this.nextRef = nextRef;
|
||||
this.read = read;
|
||||
|
|
@ -81,6 +82,7 @@ public class PairHMMTestData {
|
|||
this.delQuals = delQuals;
|
||||
this.gcp = gcp;
|
||||
this.log10l = log10l;
|
||||
this.newRead = newRead;
|
||||
}
|
||||
|
||||
PairHMMTestData(String ref, String nextRef, String read, final byte qual) {
|
||||
|
|
@ -90,6 +92,7 @@ public class PairHMMTestData {
|
|||
this.baseQuals = this.insQuals = this.delQuals = Utils.dupBytes(qual, read.length());
|
||||
this.gcp = Utils.dupBytes((byte)10, read.length());
|
||||
this.log10l = -1;
|
||||
this.newRead = true;
|
||||
}
|
||||
|
||||
public double runHMM(final PairHMM hmm) {
|
||||
|
|
@ -104,25 +107,28 @@ public class PairHMMTestData {
|
|||
"ref='" + ref + '\'' +
|
||||
", nextRef=" + nextRef + '\'' +
|
||||
", read='" + getRead() + '\'' +
|
||||
", log10l=" + log10l +
|
||||
", log10l=" + log10l + '\'' +
|
||||
", newRead=" + newRead +
|
||||
'}';
|
||||
}
|
||||
|
||||
public static void runHMMs(final PairHMM hmm, final List<PairHMMTestData> data, final boolean runSingly) {
|
||||
public static double runHMMs(final PairHMM hmm, final List<PairHMMTestData> data, final boolean runSingly) {
|
||||
double result = 0;
|
||||
if ( runSingly ) {
|
||||
for ( final PairHMMTestData datum : data )
|
||||
datum.runHMM(hmm);
|
||||
result += datum.runHMM(hmm);
|
||||
} else {
|
||||
// running in batch mode
|
||||
final PairHMMTestData first = data.get(0);
|
||||
int maxHaplotypeLen = calcMaxHaplotypeLen(data);
|
||||
hmm.initialize(first.getRead().length(), maxHaplotypeLen);
|
||||
for ( final PairHMMTestData datum : data ) {
|
||||
hmm.computeReadLikelihoodGivenHaplotypeLog10(datum.ref.getBytes(), datum.getRead().getBytes(),
|
||||
datum.baseQuals, datum.insQuals, datum.delQuals, datum.gcp, false, datum.nextRef.getBytes());
|
||||
result += hmm.computeReadLikelihoodGivenHaplotypeLog10(datum.ref.getBytes(), datum.getRead().getBytes(),
|
||||
datum.baseQuals, datum.insQuals, datum.delQuals, datum.gcp, datum.newRead, datum.nextRef.getBytes());
|
||||
|
||||
}
|
||||
}
|
||||
return result;
|
||||
}
|
||||
|
||||
public static int calcMaxHaplotypeLen(final List<PairHMMTestData> data) {
|
||||
|
|
@ -132,6 +138,13 @@ public class PairHMMTestData {
|
|||
return maxHaplotypeLen;
|
||||
}
|
||||
|
||||
public static int calcMaxReadLen(final List<PairHMMTestData> data) {
|
||||
int maxReadLen = 0;
|
||||
for ( final PairHMMTestData datum : data )
|
||||
maxReadLen = Math.max(maxReadLen, datum.getRead().length());
|
||||
return maxReadLen;
|
||||
}
|
||||
|
||||
public static Map<String, List<PairHMMTestData>> readLikelihoods(final File file) throws IOException {
|
||||
final Map<String, List<PairHMMTestData>> results = new LinkedHashMap<>();
|
||||
|
||||
|
|
@ -153,7 +166,8 @@ public class PairHMMTestData {
|
|||
SAMUtils.fastqToPhred(thisEntry[3]),
|
||||
SAMUtils.fastqToPhred(thisEntry[4]),
|
||||
SAMUtils.fastqToPhred(thisEntry[5]),
|
||||
Double.parseDouble(thisEntry[6]));
|
||||
Double.parseDouble(thisEntry[6]),
|
||||
! results.containsKey(thisEntry[1]));
|
||||
|
||||
if ( ! results.containsKey(info.read) ) {
|
||||
results.put(info.read, new LinkedList<PairHMMTestData>());
|
||||
|
|
@ -171,7 +185,8 @@ public class PairHMMTestData {
|
|||
SAMUtils.fastqToPhred(thisEntry[3]),
|
||||
SAMUtils.fastqToPhred(thisEntry[4]),
|
||||
SAMUtils.fastqToPhred(thisEntry[5]),
|
||||
Double.parseDouble(thisEntry[6]));
|
||||
Double.parseDouble(thisEntry[6]),
|
||||
! results.containsKey(thisEntry[1]));
|
||||
|
||||
if ( ! results.containsKey(info.read) ) {
|
||||
results.put(info.read, new LinkedList<PairHMMTestData>());
|
||||
|
|
@ -182,6 +197,59 @@ public class PairHMMTestData {
|
|||
return results;
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
* simplified likelihoods file reader that returns a list instead of a map
|
||||
*
|
||||
* readLikelihoods() method was reordering inputs, with the result that caching would be more efficient
|
||||
* This method simply returns a list of read/haplotype pairs in their original order, providing a more realistic caching scenario
|
||||
*/
|
||||
public static List<PairHMMTestData> readLikelihoodsInOrder(final File file) throws IOException {
|
||||
final List<PairHMMTestData> results = new LinkedList<>();
|
||||
|
||||
InputStream in = new FileInputStream(file);
|
||||
if ( file.getName().endsWith(".gz") ) {
|
||||
in = new GZIPInputStream(in);
|
||||
}
|
||||
|
||||
String previousRead = null;
|
||||
String[] nextEntry;
|
||||
String[] thisEntry = null;
|
||||
for ( final String line : new XReadLines(in) ) {
|
||||
// peak at the next entry (to get the haplotype bases)
|
||||
nextEntry = line.split(" ");
|
||||
// process the current entry
|
||||
if (thisEntry != null) {
|
||||
final PairHMMTestData info = new PairHMMTestData(
|
||||
thisEntry[0], nextEntry[0], thisEntry[1],
|
||||
SAMUtils.fastqToPhred(thisEntry[2]),
|
||||
SAMUtils.fastqToPhred(thisEntry[3]),
|
||||
SAMUtils.fastqToPhred(thisEntry[4]),
|
||||
SAMUtils.fastqToPhred(thisEntry[5]),
|
||||
Double.parseDouble(thisEntry[6]),
|
||||
!(thisEntry[1].equals(previousRead)));
|
||||
|
||||
results.add(info);
|
||||
previousRead = info.getRead();
|
||||
}
|
||||
// update the current entry
|
||||
thisEntry = nextEntry;
|
||||
}
|
||||
// process the final entry
|
||||
final PairHMMTestData info = new PairHMMTestData(
|
||||
thisEntry[0], null, thisEntry[1],
|
||||
SAMUtils.fastqToPhred(thisEntry[2]),
|
||||
SAMUtils.fastqToPhred(thisEntry[3]),
|
||||
SAMUtils.fastqToPhred(thisEntry[4]),
|
||||
SAMUtils.fastqToPhred(thisEntry[5]),
|
||||
Double.parseDouble(thisEntry[6]),
|
||||
!(thisEntry[1].equals(previousRead)));
|
||||
|
||||
results.add(info);
|
||||
|
||||
return results;
|
||||
}
|
||||
|
||||
public String getRead() {
|
||||
return read;
|
||||
}
|
||||
Loading…
Reference in New Issue