hifisr

HiFi-SR is a Python-based pipeline for the detection of plant mitochondrial structural rearrangements based on the mapping of PacBio high-fidelity (HiFi) reads or Circular Consensus Sequencing (ccs) data, to a reference genome (i.e., the hypothetical master cycle DNA).

Deps

The pipeline has been tested in WSL2 distribution Ubuntu-20.04. It shall work in other Linux operating system, such as CentOS.

Download the hifisr repository

git clone https://github.com/zouyinstein/hifisr

Create and activate a conda environment.

conda create -n hifisr python=3.9
conda activate hifisr

Use Anaconda3 to install required packages.

conda install pigz -c conda-forge
conda install samtools bamtools blast seqkit -c bioconda
# create a soft link to ensure samtools can work
ln -sf ${HOME}/.conda/envs/hifisr/lib/libcrypto.so.1.1 ${HOME}/.conda/envs/hifisr/lib/libcrypto.so.1.0.0

Install minimap2

cd hifisr/deps
curl -L https://github.com/lh3/minimap2/releases/download/v2.24/minimap2-2.24_x64-linux.tar.bz2 | tar -jxvf -
./minimap2-2.24_x64-linux/minimap2

Install required Python packages

pip install biopython pandas openpyxl

Test

# Make sure the HiFi-SR repository has been downloaded (git clone https://github.com/zouyinstein/hifisr).
# Make sure the hifisr environment has been activated (conda activate hifisr).
# Make sure the dependent third-party softwares/packages has been installed.
# Change working directory to hifisr
cd hifisr
# Add minimap2 executable to the system PATH
export PATH="$PWD/deps/minimap2-2.24_x64-linux":$PATH
# Change working directory to test
cd test
# Copy the Arabidopsis thaliana (Col-0) reference and demo HiFi reads to a new directory Col
cp -R start Col
# Change working directory to the new directory Col
cd Col
# unzip the files
pigz -d -p 8 *.gz
# Run the HiFi-SR pipeline for Col
python ../../hifisr.py -s Col single &
# OR python ../../hifisr.py -s Col -t 16 -i fastq single &
# clean results if you want to rerun the test
rm -rf Col

Example 1

Analyze of an example wild-type Arabidopsis thaliana dataset Col-CEN (ERR6210723, 14.6 Gb, Naish et al., 2021, Science):

cd hifisr/pre
mkdir CEN  # CEN is the sample name
cd CEN
pigz -d -k -p 16 ERR6210723.fastq.gz
ln -sf ERR6210723.fastq CEN.fastq
# manually change the FASTA headers of mitochondrial and plastid genome refercences into mito and plastid for easily manipulation
cat Athaliana_447_TAIR10.id_Chr{1,2,3,4,5}.fa refs_cp28673mod.id_mito.fas refs_cp28673mod.id_plastid.fas > CEN_ref.fa
ln -sf refs_cp28673mod.id_mito.fas CEN_mito.fa
ln -sf refs_cp28673mod.id_plastid.fas CEN_plastid.fa
# check conda activate hifisr
# check export PATH=/path/to/hifisr/deps/minimap2-2.24_x64-linux:$PATH
python ../../hifisr.py -s CEN -t 10 -i fastq single &

Example 2

Analyze of an example wild-type Arabidopsis thaliana dataset Col-XJTU (CRR302668, 22.9 Gb, Wang et al., 2021, GPB):

cd hifisr/pre
mkdir XJTU  # XJTU is the sample name
cd XJTU
pigz -d -k -p 16 CRR302668.fastq.gz
ln -sf CRR302668.fastq XJTU.fastq
# manually change the FASTA headers of mitochondrial and plastid genome refercences into mito and plastid for easily manipulation
cat Athaliana_447_TAIR10.id_Chr{1,2,3,4,5}.fa refs_cp28673mod.id_mito.fas refs_cp28673mod.id_plastid.fas > XJTU_ref.fa
ln -sf refs_cp28673mod.id_mito.fas XJTU_mito.fa
ln -sf refs_cp28673mod.id_plastid.fas XJTU_plastid.fa
# check conda activate hifisr
# check export PATH=/path/to/hifisr/deps/minimap2-2.24_x64-linux:$PATH
python ../../hifisr.py -s XJTU -t 10 -i fastq single &

Merge reports of multiple samples

Merge reports of Col-CEN and Col-XJTU:

cd hifisr/pre
echo CEN >> merge_1.txt
echo XJTU >> merge_1.txt
# check conda activate hifisr
# check export PATH=/path/to/hifisr/deps/minimap2-2.24_x64-linux:$PATH
python ../../hifisr.py -m merge1 merge &

Genesislearn/hifisr

hifisr