Mitogenome_assembler

This assembler is developped to de novo assemble plant mitochondrial genomes

Our mitogenome assembly pipeline uses whole-genome HiFi sequencing data as input and outputs a complete mitogenome sequence.

The following figure shows the workflow of our assembly pipeline:

Step 1. Read correction using Canu correct module if the input data was not generated from PacBio HiFi sequencing

    canu -correct -p Correct -d PacBio genomeSize=500m rawErrorRate=0.3 correctedErrorRate=0.045 batThreads=20 "batOptions=-dg 3 -db 3 -dr 1 -ca 500 -cp 50" useGrid=false -pacbio PacBio.fastq.gz
    # -p <assembly-prefix>
    # -d <assembly-directory>
    # genomeSize=<number>[g|m|k]
    # rawErrorRate:The allowed difference in an overlap between two raw uncorrected reads. The defaults are 0.300 for PacBio reads
    # correctedErrorRate:The allowed difference in an overlap between two corrected reads. Defaults are 0.045 for PacBio reads

Step 2. Convert long reads into a new file for Newbler assembly (<30k)

    perl PacBio_to_Newbler.pl -i Hifi.fa -o Hifi.cut20k.fa -b 20000
    # -i <corrected or HiFi fasta file>
    # -o <coverted fasta file>
    # -b Braek the reads (>30k) into more short reads (<20k); default: 20000 bp

Step 3. De novo assembly using Newbler v.3.0

You can get the packaged Newbler runAssembly.sif here

singularity exec runAssembly.sif runAssembly -cpu 10 -het -sio -m -urt -large -s 100 -nobig -o $output_dir Hifi.cut20k.fa
    # -het: Flag to enable Heterozygotic mode, which causes the assembler to choose a path between 2 contigs when there is ambiguity regarding the path that should be taken.
    #       By default, the assembler will not choose a path when ambiguity exists
    # -sio: Flag to invoke serial I/O. Use for projects with > 4 million reads.
    # -m: Keeps sequence data in memory Increases speed
    # -urt: Extend contigs using the ends (tips) of single reads
    # -large: Speeds up assembly but reduces accuracy
    # -nobig: Skip output of large files (.ace, 454AlignmentInfo.tsv) Default: no
    # -o <output-directory>

Step 4. Selecting extended seed contigs (longest 3~5 contigs with medium depth)

    awk '{if($2 ~ /^contig/)print}' 454ContigGraph.txt > Species.csv
    # draw coverage distribution graph with Scripts/Plot_coverage.R

The following figure shows the length and coverage distribution for selecting seeds:

Step 5. Extending seed contigs and constructing initial assembly graph

gunzip Example/Organism/Assembly_result/454AllContigs.fna.gz
    python3 Get_mitogenome_from_assembly.py -c 454ContigGraph.txt -a 454AllContigs.fna -o Assembly_graph.gfa -s 406 501 566 831 1029
    # -c <454ContigGraph.txt> Assembly graph file generated from Newbler
    # -a <454AllContigs.fna> All generated contig sequences
    # -s <seeds> Selected from step4
    # -o <Output graph file> Output file for Step6

Step 6. Visualizing and simplifying assembly graph using Bandage

    # Open Assembly.gfa with Bandage
    # Remove full-path cp contigs and tip contigs
    # Decode the revised assembly graph based on the copy number of each contig
    # Merged all possible nodes and exported the complete mitogenome sequence to a FASTA format file

Step 7. Exporting the final complete mitogenome sequence using Bandage

The following figure shows the raw and simplified assembly graph: