HAPDeNovo: a haplotype based approach for filtering and phasing de novo mutations on barcoded linked read sequencing
HAPDeNovo is a software package to detect de novo mutations on barcoded linked read sequencing. We identified the major source of de novo mutation from NGS data was caused by the biased haplotype coverage. HAPDeNovo can eliminate more than 99% false positive de novo calls with the help of 10X linked read sequencing.
git clone https://github.com/maiziex/HAPDeNovo.git
To download human reference genome, go to 10X GENOMICS Website, or
wget http://cf.10xgenomics.com/supp/genome/refdata-hg19-2.1.0.tar.gz
tar -xzvf refdata-hg19-2.1.0.tar.gz
cd HAPDeNovo
chmod +x install.sh
./install.sh
use 23 for X chromosome
1. X in female offspring:
--child_sex = female
2. X in male offspring:
--child_sex = male
For X chromosome, make sure parent1 is father, and parent2 is mother
Inputs are 10X fastqs files, to get bam files:
../lib/longranger/longranger wgs --id=12878 --fastqs=NA12878.fastqs --reference=refdata-hg19-2.1.0/fasta/genome.fa
../lib/longranger/longranger wgs --id=12891 --fastqs=NA12891.fastqs --reference=refdata-hg19-2.1.0/fasta/genome.fa
../lib/longranger/longranger wgs --id=12892 --fastqs=NA12892.fastqs --reference=refdata-hg19-2.1.0/fasta/genome.fa
Performs a multiple-sample variants call by FreeBayes or GATK.
To generate trio_merge.vcf by using FreeBayes:
../lib/freebayes/bin/freebayes -f refdata-hg19-2.0.0/fasta/genome.fa ../example/NA12878_GRCh37.bam ../example/NA12891_GRCh37.bam ../example/NA12892_GRCh37.bam > trio_merge.vcf
Or to generate trio_merge.vcf by using GATK. For GATK, more information can be found from here.
java -jar ../lib/GenomeAnalysisTK/GenomeAnalysisTK.jar -T HaplotypeCaller -R refdata-hg19-2.1.0/fasta/genome.fa -I ../example/NA12878_GRCh37.bam -o child_gvcf -ERC GVCF
java -jar ../lib/GenomeAnalysisTK/GenomeAnalysisTK.jar -T HaplotypeCaller -R refdata-hg19-2.1.0/fasta/genome.fa -I ../example/NA12891_GRCh37.bam -o parent1_gvcf -ERC GVCF
java -jar ../lib/GenomeAnalysisTK/GenomeAnalysisTK.jar -T HaplotypeCaller -R refdata-hg19-2.1.0/fasta/genome.fa -I ../example/NA12892_GRCh37.bam -o parent2_gvcf -ERC GVCF
java -jar ../lib/GenomeAnalysisTK/GenomeAnalysisTK.jar -T GenotypeGVCFs -R refdata-hg19-2.1.0/fasta/genome.fa -V child_gvcf -V parent1_gvcf -V parent2_gvcf -o trio_merge.vcf
To split the multiple-sample vcf file:
../lib/vcftools/src/perl/vcf-subset -c trio_merge.vcf
To run phased variants call, all the precalled vcf files are generated from the previous step:
../lib/longranger wgs --id=NA12878 --sex=female --fastqs=NA12878.fastqs --reference=refdata-hg19-2.1.0/fasta/genome.fa --precalled=20976.vcf
../lib/longranger wgs --id=NA12891 --sex=male --fastqs=NA12891.fastqs --reference=refdata-hg19-2.1.0/fasta/genome.fa --precalled=20971.vcf
../lib/longranger wgs --id=NA12892 --sex=female --fastqs=NA12892.fastqs --reference=refdata-hg19-2.1.0/fasta/genome.fa --precalled=20972.vcf
python3 HAPDeNovo_step2.py --child_vcf ../example/NA12878_phased_variants.vcf.gz --parent1_vcf ../example/NA12891_phased_variants.vcf.gz --parent2_vcf ../example/NA12892_phased_variants.vcf.gz --chr_start 1 --chr_end 22 --out_dir ../output/
--child_vcf: "NA12878_phased_variants.vcf.gz" is the phased gzipped vcf file of the child generated by Long Ranger from step 1.
--parent1_vcf: "NA12891_phased_variants.vcf.gz" is the phased gzipped vcf file of the first parent generated by Long Ranger from step 1.
--parent2_vcf: "NA12892_phased_variants.vcf.gz" is the phased gzipped vcf file of the second parent generated by Lon gRanger from step 1.
--chr_start: "1" specifies chromosome starting from 1.
--chr_end: "23" specifies chromosome ending by 23 (chrX).
--out_dir: "../output/" specifies the customized folder name for output results.
For large memory server, chromosomes can be processed by multi-threading, and for small memory server, chromosome is suggested to be processed one by one. This step is time consuming.
python3 HAPDeNovo_step3.py --child NA12878 --parent1 NA12891 --parent2 NA12892 --child_id 20976 --parent1_id 20971 --parent2_id 20972 --child_bam ../example/NA12878_GRCh37.bam --parent1_bam ../example/NA12891_GRCh37.bam --parent2_bam ../example/NA12892_GRCh37.bam --reference refdata-hg19-2.1.0/fasta/genome.fa --chr_start 1 --chr_end 1 --out_dir ../output/ --child_sex female
--child: "NA12878" is the origin sample id of the child.
--parent1: "NA12891" is the origin sample id of the first parent.
--parent2: "NA12892" is the origin sample id of the second parent.
--child_id: "20976" is the sample id from the header of the bam file for child. To check this sample id, you can use /src/Check_sampleID.py
--parent1_id: "20971" is the sample id from the header of the bam file for the first parent. To check this sample id, you can use /src/Check_sampleID.py
--parent2_id: "20972" is the sample id from the header of the bam file for second parent. To check this sample id, you can use /src/Check_sampleID.py
--child_bam: "NA12878_GRCh37.bam" is the bam file of the child.
--parent1_bam: "NA12891_GRCh37.bam" is the bam file of the first parent.
--parent2_bam: "NA12892_GRCh37.bam" is the bam file of the second parent.
--reference: "refdata-hg19-2.1.0/fasta/genome.fa" is the reference as before.
--chr_start: "1" specifies chromosome starting from 1.
--chr_end: "1" specifies chromosome ending by 1.
--out_dir: "../output/" specifies the customized folder name for output results.
--child_sex: "female" used by daughter.
python3 HAPDeNovo_final.py --chr_start 1 --chr_end 23 --child_id 20976 --parent1_id 20971 --parent2_id 20972 --out_dir ../output/ --output_file_prefix HAPDeNovo_filer1_result --depth 1 --input_file ../example/trio_merge.vcf --child_sex female
--chr_start: "1" specifies chromosome starting from 1.
--chr_end: "23" specifies chromosome ending by 23 (chrX).
--child_id: "20976" is the sample id from the header of the bam file for child.
--parent1_id: "20971" is the sample id from the header of the bam file for first parent.
--parent2_id: "20972" is the sample id from the header of the bam file for second parent.
--out_dir: "../output/" specifies the customized folder name for output results.
--output_file_prefix: "denovo_triodenovo_depth15_HAPDeNovo_filer1" is the user customized file name for final de novo mutation output.
--depth: "1" specifies the depth filter by 1.
--input_file: "trio_merge.vcf" is the variant call file from FreeBayes (default). This input can also be the de novo results from other tools like GATK/TrioDenovo/DeNovoGear (check doc
), or any user customized txt file for de novo muations. This text file only needs to contain chromose name and the corresponding locus per row for each de novo mutation.
--child_sex: "female" used by daughter.
Chr Locus REF ALT child_GT parents1_GT parents2_GT Confidence
chr15 21173525 C T 0|1 0|0 0|0 H
chr15 22659065 G A 0|1 0|0 0|0 L
chr15 32590720 C G 1|0 0|0 0|0 L
chr15 33183990 C T 0|1 0|0 0|0 H
chr15 50953965 A C 1|0 0|0 0|0 H
chr15 51248561 C T 0|1 0|0 0|0 H
chr15 52560491 G A 0/1 0|0 0|0 L
chr15 58149075 T A 1|0 0|0 0|0 H
chr15 58669774 T C 0|1 0|0 0|0 H
chr15 63727259 C A 0/1 0|0 0|0 L
chr15 64217272 T G 0/1 0|0 0|0 L
chr15 74880589 A G 1|0 0|0 0|0 H
chr15 83299364 A C 0|1 0|0 0|0 L
chr15 83299956 T A 0|1 0|0 0|0 L
chr15 83305550 A G 0|1 0|0 0|0 L
chr15 83307063 T A 0|1 0|0 0|0 L
chr15 83319622 G C 0|1 0|0 0|0 L
chr15 83323255 C G 0|1 0|0 0|0 L
chr15 83336872 C T 0|1 0|0 0|0 L
chr15 83358339 C T 0|1 0|0 0|0 L
chr15 83361080 C A 0|1 0|0 0|0 L
chr15 83386567 C T 0|1 0|0 0|0 L
chr15 83389209 T C 0|1 0|0 0|0 H
chr15 83450721 C T 0|1 0|0 0|0 L
chr15 83457052 G A 0|1 0|0 0|0 L
chr15 85056658 C T 1|0 0|0 0|0 L
chr15 85715561 C T 0|1 0|0 0|0 H
chr15 87791919 T C 1|0 0|0 0|0 H
chr15 99175976 A G 0|1 0|0 0|0 H
chr15 100884734 A G 0/1 0|0 0|0 L