MitoHPC2

Mitochondrial High Performance Caller v2

A pipeline for detecting mitochondrial homoplasmies and heteroplasmies from sequencing data.

Capabilities

Detects homoplasmies and heteroplasmies in mitochondria
Supports short paired and long reads
Multiple SNV callers
Runs 2 iterations: the first uses a standard reference, the second uses the sample consensus sequence
Handles NUMTs and origin regions carefully
Supports human and mouse genomes, with multiple reference versions
Allows multiple heteroplasmy thresholds
Calls maternal haplogroup , consensus sequence
Detects sample contamination
Processes multiple samples in parallel and combines results
Provides detailed output summaries
Optimized for low CPU, memory, and runtime usage

Citing

If you use this pipeline, please cite:

Battle et al., NAR 2022: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9112767/

Reference

For additinal information please check (MitoHPC v1)[https://github.com/dpuiu/MitoHPC]

Prerequisites

The pipeline requires the following tools:

Reference Genomes:

Default: hg38
Setting available in scripts/init.sh

Input

The pipeline expects pre-aligned, coordinate sorted reads:

File formats: .bam or .cram
Sequencing platforms: Illumina paired-end, PacBio HiFi, Oxford Nanopore (ONT) Dorado R9+
Index files: .bai or .crai
Indexstats files: .idxstats: contain the number of reads aligned to each chromosome

Output

count.tab – total reads & mtDNA-CN
cvg.tab – coverage stats

Iteration 1

snvcaller.*.vcf – SNVs at multiple heteroplasmy level
snvcaller.*.tab / .summary / .pos – SNV counts & summaries
snvcaller.fa – consensus sequence
snvcaller.haplogroup.tab – maternal haplogroup
snvcaller.haplocheck.tab – contamination check

Iteration 2

snvcaller.snvcaller.* – final results

snvcaller is usuall mutect2(Illumina), deepsomatioc (PacBio hifi), clairs-to (ONT)

Installation

1. Download the Pipeline

git clone https://github.com/dpuiu/MitoHPC2.git

2. Update the Pipeline (optional)

cd MitoHPC2/
git pull
# or
git checkout .

3. Set up the Environment (important)

cd MitoHPC2/scripts
export HP_SDIR=`pwd`                           # set script directory
echo "export HP_SDIR=`pwd`" >> ~/.bashrc       # add variable to .bashrc
. ./init.sh                                    # or init.{hs38DH,hg19,mm39}.sh for different references

4. Install System Prerequisites (optional)

sudo $HP_SDIR/install_sysprerequisites.sh      # installs perl, python, java, wget, etc.

5. Install Pipeline Prerequisites

$HP_SDIR/install_prerequisites.sh             # installs bwa, samtools, bedtools, etc.
# or force latest versions
$HP_SDIR/install_prerequisites.sh -f

6. Install Long-Read SNV Callers (optional)

List and download pre-built Singularity images to $HP_BDIR:

curl -s ftp://ftp.ccb.jhu.edu/pub/dpuiu/Homo_sapiens_mito/MitoHPC2/bin/
# example files:
# clair3_v1.2.0.sif
# clairs-to_v0.4.2.sif
# deepsomatic_1.9.0.sif
# deepvariant_1.10.0-beta.sif

curl -LO ftp://ftp.ccb.jhu.edu/pub/dpuiu/Homo_sapiens_mito/MitoHPC2/bin/clairs-to_v0.4.2.sif
curl -LO ftp://ftp.ccb.jhu.edu/pub/dpuiu/Homo_sapiens_mito/MitoHPC2/bin/deepsomatic_1.9.0.sif

Convert to sandbox for faster execution:

singularity build --sandbox ~/clairs-to_sandbox/   $HP_BDIR/clairs-to.sif
singularity build --sandbox ~/deepsomatic_sandbox/ $HP_BDIR/deepsomatic.sif
du -hs ~/clairs-to_sandbox/ ~/deepsomatic_sandbox/

7. Custom Annotation

The pipeline uses MitoHPC2/RefSeq/*.{vcf,bed}.gz for annotation.
Copy any custom VCF/BED files to MitoHPC2/RefSeq/.
Ensure files are bgzipped and indexed.

8. Check Installation

$HP_SDIR/checkInstall.sh
cat checkInstall.log

"Success message!" expected

Setup environment

1. Illumina Data

nano scripts/init.sh                          # check/edit local init file
    ...
    export HP_O=Human                         # organism
    export HP_MT=chrM
    export HP_MTLEN=16569
    export HP_RNAME=hs38DH                    # reference assembly used for alignment (Ex: hs38DH, hg19)   
    export HP_RMT=chrM                        # MT sequence name
    export HP_RNUMT="chr1:629084-634672 ..."  # NUMT regions to be considered     
    export HP_RURL=https://hgdownload..."     # URL location (if needed to be downloaded)
    export HP_IN=$PWD/in.txt                  # input TSV file, contains sample names, file paths (automatically generated by runnng ". ./init.sh")
    export HP_ADIR=$PWD/bams/                 # pre-existing alignment dir; contains .bam, .bai, [.idxstats] files, or
    export HP_ADIR=$PWD/crams/                #                                       .cram, .crai, [.idxstats] files
    export HP_ODIR=$PWD/out/                  # output dir  
    export HP_L=222000                        # Use at most 222K reads (random subsampling; ~2K cvg of MT for 150bp reads); if empty, use all the reads
    export HP_M=mutect2                       # SNV caller: mutect2, mutserve, or freebayes
    export HP_I=2                             # number of SNV calling iterations
                                              #  1: one, uses rCRS or RSRS as reference 
                                              #  2: two, the 2nd one used the sample consensus 
                                              #     computed in the 1st iteration; higher accuracy 
    export HP_T1=03                           # heteroplasmy thold 1: 3%
    export HP_T2=05                           # heteroplasmy thold 2: 5%
    export HP_T3=10                           # heteroplasmy thold 3: 10%
    export HP_SH=bash                         # job scheduling: bash, qsub,sbatch, ...
    ...

2. Illumina Data: Multiple SNV Callers

# difference compared to scripts/init.sh 
nano scripts/init3.sh            
    ...
    export HP_M=
    export HP_M1=mutect2
    export HP_M2=varscan
    export HP_M3=freebayes
    export HP_M=merge3
    ...

3. PacBio HiFi Data

# difference compared to scripts/init.sh 
nano scripts/init.hifi.sh
    ...
    export HP_L=4000                              # maximum number of alignments sampled
    export HP_M=deepsomatic                       # SNV caller: deepsomatic, clairs-to ... varscan,bcftools,clair3,deepvariant
    export HP_MODELTYPE="PACBIO_TUMOR_ONLY"       # deepsomatic:WGS,WES,PACBIO,ONT,FFPE_WGS,FFPE_WES,WGS_TUMOR_ONLY,PACBIO_TUMOR_ONLY,ONT_TUMOR_ONLY
    export HP_MINLEN=6000                         # min total alignment (> longest NUMT)
    export HP_MINPC="0.95"                        # min alignment coverage
    export HP_MAXDP=2000                          # max depth
    export HP_MINAF=0.05                          # min AF
    ...

4. ONT Data

# difference compared to scripts/init.sh , scripts/init.hifi.sh
nano scripts/init.ont.sh
    ...
    export HP_M=clairs-to                         # SNV caller: varscan,...
    export HP_PLATFORM="ont_r10_dorado_sup_5khz"  # clairs-to  {ont_r10_dorado_sup_4khz, ont_r10_dorado_hac_4khz, ont_r10_dorado_sup_5khz, ont_r10_dorado_sup_5khz_ss, ont_r10_dorado_sup_5khz_ssrs
    ...

Running the Pipeline

1. Illumina Data

cp $HP_SDIR/init.sh .
nano ./init.sh            
. ./init.sh
$HP_SDIR/run.sh | tee run.all.sh | bash

2. Illumina Data: Multiple SNV Callers

cp $HP_SDIR/init3.sh .
nano ./init3.sh       
. ./init3.sh
$HP_SDIR/run3.sh | tee run.all.sh | bash

3. PacBio HiFi Data

cp $HP_SDIR/init.hifi.sh .
nano ./init.hifi.sh        
. ./init.hifi.sh
$HP_SDIR/run.lr.sh | tee run.all.sh | bash

4. ONT Data

cp $HP_SDIR/init.ont.sh .
nano ./init.ont.sh         
. ./init.ont.sh
$HP_SDIR/run.lr.sh | tee run.all.sh | bash

Output & Evaluation

ls $HP_ODIR/.*

Compare results to Illumina mutect2 (T=10):

$HP_SDIR/eval.sh Illumina/out/mutect2.10.concat.vcf $HP_M.10.concat.vcf | uniq.pl | column -t > $HP_M.10.eval

Examples

1. 39 HPRC Samples

Download example files via FTP

HiFi10 clairs-to: 10.10, 05.10, 10.05, 05.05, 10.00, 05.00
HiFi10 deepsomatic: 10.10, 05.10, 10.05, 05.05, 10.00, 05.00
HiFi10 varscan: 10.10, 05.10, 10.05, 05.05, 10.00, 05.00
ONT10 clairs-to: 10.10, 05.10, 10.05, 05.05, 10.00, 05.00
ONT10 deepsomatic: 10.10, 05.10, 10.05, 05.05, 10.00, 05.00
ONT10 varscan: 10.10, 05.10, 10.05, 05.05, 10.00, 05.00

2. 39 HPRC bases Simulated samples

HiFi10Sim

500x cvg of Illumina paired end reads (2*150bp) were simulated using the best HiFi10 raeds (avg Q30)
commands: wgsim -N 22200 -1 150 -2 150 -d 300 -e 0 -r 0 -R 0
There reads were processed using the MitoHPC2 pipeline ("mutect2" SNV caller) 10.10

ONT10Sim

500x cvg of Illumina paired end reads (2*150bp) were simulated using the best ONT10 raeds (avg Q20)
commands: wgsim -N 22200 -1 150 -2 150 -d 300 -e 0 -r 0 -R 0
There reads were processed using the MitoHPC2 pipeline ("mutect2" SNV caller) 10.10

Note: 1st number: short read heteroplasmy thold; 2nd: long read heteroplasmy thold

Name		Name	Last commit message	Last commit date
Latest commit History 132 Commits
RefSeq		RefSeq
examples		examples
scripts		scripts
.gitattributes		.gitattributes
CHANGELOG.md		CHANGELOG.md
Dockerfile		Dockerfile
LICENSE		LICENSE
README.md		README.md
VERSION.md		VERSION.md

Folders and files

Latest commit

History

Repository files navigation

MitoHPC2

Mitochondrial High Performance Caller v2

Capabilities

Citing

Reference

Table of Contents

Prerequisites

Input

Output

Iteration 1

Iteration 2

Installation

1. Download the Pipeline

2. Update the Pipeline (optional)

3. Set up the Environment (important)

4. Install System Prerequisites (optional)

5. Install Pipeline Prerequisites

6. Install Long-Read SNV Callers (optional)

7. Custom Annotation

8. Check Installation

Setup environment

1. Illumina Data

2. Illumina Data: Multiple SNV Callers

3. PacBio HiFi Data

4. ONT Data

Running the Pipeline

1. Illumina Data

2. Illumina Data: Multiple SNV Callers

3. PacBio HiFi Data

4. ONT Data

Output & Evaluation

Examples

1. 39 HPRC Samples

2. 39 HPRC bases Simulated samples

HiFi10Sim

ONT10Sim

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages