bifrost_assemblatron

This component is used to perform a de novo assembly from trimmed paired-end sequence reads obtained from another component. Following the de novo assembly, basic filtering is performed and metrics is calculated, to ensure high quality contigs.

Requirements

• The component creates the de novo assembly using the tool spades.
• The versions are described in the environment.yaml
• The alignment uses the trimmed reads from another component.

Download

git clone https://github.com/ssi-dk/bifrost_assemblatron.git
cd bifrost_assemblatron
git submodule init
git submodule update
bash install.sh -i LOCAL
conda activate bifrost_assemblatron_vx.x.x
export BIFROST_INSTALL_DIR='/your/path/'
BIFROST_DB_KEY="/your/key/here/" python -m bifrost_assemblatron -h

Run the snakemake analysis

Each component can be run on each sample individually using one snakemake command, replacing the string passed to the --config sample_name=" " with the appropriate dataset name. The provided component_name= takes as an argument <component_name>__<version_number>. The component name aligns with the GitHub repo name, which is structured like bifrost_<component_name> (e.g. bifrost_assemblatron -> component name assemblatron), and the version number aligns with the current GitHub tag / or conda environment version (e.g. v.0.0.2) defined during the bifrost component setup.

snakemake -p --nolock --cores 5 -s <github_path>/pipeline.smk --config sample_name="insert sample name" component_name=assemblatron__v2.3.3 --rerun-incomplete

Analysis

do novo assembly

The spades command is defined in the pipeline, using the assembly running mode "--isolate" (spades documentation) is recommended for high-coverage isolate and multi-cell Illumina data, and this improves the assembly quality and decreases the running time.

One example of the assembled contig headers

>24000006_MW-ESCEC_1_length_505018_cov_24.431370
AACCTGCGACCAATTGATTAAAAGTCAACTGCTCTACCAACTGAGCTAACGACCCACTTT
TTCGTTGCTTTCGGTTTGTTTGATATCCCGTGGCAACGGCGGCATATATTACTGATTTCA
GACTTGAGCGCAACAAAAATTTCGATGTAGATCACTCAACTGCTTATGATTCGCACGACA

From the contig header, the depth of coverage can be extracted and used for QC on the assembled contigs.

rule assembly_qc

The following filtering and metric calculations are performed in the rule__assembly_qc using the following steps:

1. Removes all contigs below the length of 500 bp (option --min-contig-len & --failed-length)
2. Removes all the contigs with a coverage below 10x (extracted from the spades header, option --cov-threshold & --failed-coverage)
3. Stores the passed files as the final filtered assembly files used for future components (--passed)

The command takes prefixes for the filtered files to generate both the filtered fasta files and statistical measurements

python rule__assembly_qc.py --assembly generated_de_novo_assembly.fasta --cov-threshold 10 --min-contig-len 500 --passed passed_file_prefix --failed-length filtered_on_length_prefix --failed-coverage filtered_on_coverage_prefix

The calculated metrics include:

• shortest contig length
• total number of filtered contigs
• total length of contigs
• N50 : which is a measurement representing the shortest contig length required to cover 50% of the total assembly length
• Average depth of coverage from the filtered contigs

With one example of an output statistical file shown below.

group   cov_threshold   min_contig_len  no_contigs      sum_len n50     mean_cov
cov_ge_10       10      500     95      5018390 186344  24.88