MitoHiFi

------ This is v3.2.2 ------

MitoHiFi v3.2.2 is a python pipeline distributed under MIT License !

MitoHiFi was first developed to assemble the mitogenomes for a wide range of species in the Darwin Tree of Life Project (DToL)

Find out more Darwin Tree of Life data portal

1. Background

MitoHiFi is a python workflow that assembles mitogenomes from Pacbio HiFi reads.

With MitoHiFi v3.2 you can start from the raw Pacbio HiFi reads (flag -r) or from the assembled contigs (flag -c). You also need a reference mitochondria sequence in FASTA and GenBank format. We provide an internal script (findMitoReference.py) that can be used to find and download the most closely-related reference genome for your species from NCBI..

The availability of high-quality long reads, such as PacBio HiFi, greatly simplifies the assembly of accurate mitogenomes. However, due to the circular nature of the molecule, mitogenomes are typically assembled redundantly, resulting in multicopy mitogenome contigs. To address this, we have developed a pipeline specifically designed to finalize the assembly and the annotation of the mitogenome. Our pipeline takes into account heteroplasmy, aiming to assemble and annotate all mtDNA variants present in your sample. Among these variants, MitoHiFi selects a representative as the final genome assembly based on several criteria including circularization and gene completeness. Additionally, MitoHiFi provides multiple intermediate outputs such as coverage and annotation plots and a multiple sequence alignment (MSA) of all the variants, which facilitates the analysis of mitochondrial heteroplasmy.

MitoHiFi v3.2 will:

(i) Extract mitochondrial reads and assemble them with hifiasm (using the -r flag) or identify mitochondrial contigs among assembled contigs (using the -c flag).

(ii) Identify and separate NUMTS (Nuclear Mitochondrial DNA Sequences) from genuine mitogenome contigs.

(iii) Generate a circularized, non-redundant, and annotated version of all mitochondrial variants present in your sample.

(iv) Select a representative as the final mitochondrial genome assembly.

(v) Plot an annotation image and a coverage plot of the reads (if initiated with the -r flag).

You can view a diagram illustrating the general MitoHiFi workflow here. For documentation on the behaviour of each script visit here.

2. Installation

Below, we describe the three different ways to install MitoHiFi.

2.1 Using Docker and Singularity

We provide a Docker container for MitoHiFi. The container is built using GitHub Actions and can be obtained by running the following command:

docker pull ghcr.io/marcelauliano/mitohifi:master

Once the container is pulled, you can execute MitoHiFi within Docker. If Docker is not available on your cluster computer, you can use Singularity instead. Different environments may require slightly different Singularity parameters, but a generic Singularity command would be as follows:

singularity exec --bind /path/to/container_directory:/path/to/container_directory docker://ghcr.io/marcelauliano/mitohifi:master mitohifi.py -h

2.2 Conda: Partially Installing Dependencies with Conda

We are unable to create a complete Conda recipe for MitoHiFi due to conflicting Python versions required by different dependencies, making it impossible to coexist within the same Conda environment. However, we provide a partial Conda recipe that installs most of the dependencies. To complete the installation, follow these steps:

1. Install MitoFinder and/or MITOS outside of Conda.
2. Ensure MitoFinder and/or MITOS are added to the PATH before starting the run. Please note that MitoFinder and/or MITOS should be installed separately and made accessible via the PATH environment variable to ensure their proper integration with MitoHiFi. Once those are installed, do:

   
   #Clone MitoHiFi git repo
   git clone https://github.com/marcelauliano/MitoHiFi.git

create a conda environment with our yml file that is inside MitoHiFi/environment

conda env create -n mitohifi_env -f MitoHiFi/environment/mitohifi_env.yml

Add MitoFinder and/or MITOS to the PATH and then activate your mitohifi_env conda environment.

To run MitoHiFi, do:

(mitohifi_env) python MitoHiFi/src/mitohifi.py -h

### 2.3 Manually install all dependencies

This is the least recommended way to install MitoHiFi, but below you will find a list of software that needs to be installed and added to your PATH before installing MitoHiFi. The software versions provided are the latest ones we have tested and confirmed to be working:

  - python=3.7
  - samtools=1.11
  - cd-hit=4.8.1
  - minimap2=2.19
  - hifiasm=0.19.5
  - mafft=7.520
  - biopython=1.79
  - matplotlib=3.2.2
  - dna_features_viewer=3.1.2
  - pandas=1.3.5
  - bedtools=2.31.0
  - pillow=6.2.1
  - bcbio-gff=0.7.0
  - MitoFinder=v1.4.0
  - MITOS=2.1.0
  - ncbi-blast+

Once all dependencies are installed and in your PATH, git clone MitoHiFi and execute it:

git clone https://github.com/marcelauliano/MitoHiFi.git python MitoHiFi/src/mitohifi.py -h

## 3. Parameter list

usage: MitoHiFi [-h] (-r .fasta | -c .fasta) -f