Genome Completeness Evaluation

BUSCO is a computational tool used to assess the completeness and quality of genome assemblies(23). ASFV is not included in the BUSCO’s database (OrthoDB). Therefore, we developed a completeness evaluation system to generate a BUSCO-like notation.

completeness.py

Description

We generated consensus sequences from genotype I and II isolates. The CDS predicted from input ASFV genome using Prodigal were compared to consensus sequences using BLASTN (e-value 1e-5). This comparison yielded a BUSCO like genome completeness evaluation.

Input

An ASFV genome file.

Arguments

Argument name	Required	Description
-c	No	consensus sequences to be use: I or II (default='II')

Example

completeness.py single_fasta/OM966717.1.fasta -c II > OM966717.1_completeness.tsv

Output

A 5-column tsv table, the 5 fields are file name, genome size, gene number by prodigal, completeness evaluation with MGF genes and completeness evaluation without MGF genes. The following table is a partial example of the output.

file_name	size	prodigal_gene_num	with_MGF	without_MGF	duplicate_genes	fragmented_genes	missing_genes
OM966717.1.fasta	189125	168	C:99.32%[D:0.0%],F:0.68%,M:0.0%,n:148	C:99.13%[D:0.0%],F:0.87%,M:0.0%,n:115		C122R

The completeness evaluation is a BUSCO like notation, with C:complete [D:duplicated], F:fragmented, M:missing, n:number of genes used. If a consensus CDS term can find a mapping from the predicted gene sequences with an identity larger than 90%, together with a unique mapping length longer than 90%, it can be considered "complete." A consensus CDS term that cannot find a valid mapping (with an identity greater than 30% and a mapping length greater than 30%) in the predicted genes is considered "missing". The other consensus CDS with partial hit is termed "fragmented". The "duplicated" term means that there is more than one 'complete' hit in the predicted gene sequences.