STRAIN: an R package for multi-locus sequence typing from whole genome sequencing data.

Area of Science:

• Microbiology
• Bioinformatics
• Genomics

Background:

• Multi-locus sequence typing (MLST) is a standard method for bacterial isolate characterization.
• Whole genome sequencing (WGS) provides rich data for direct ST assignment.
• Existing WGS-based ST assignment methods include read mapping, k-mer distribution, selective assembly, and whole genome assembly.

Purpose of the Study:

• To introduce STRAIN, an R package for bacterial sequence type (ST) assignment from WGS read sets.
• To present a novel hybrid strategy combining assembly and mapping for enhanced ST identification.
• To evaluate STRAIN's performance against established ST typing tools.

Main Methods:

• STRAIN implements a hybrid approach integrating sequence assembly and read mapping.
• The method processes raw read sets from whole genome sequencing (WGS).
• Performance was benchmarked against SRST2, stringMLST, and GRAbB using 540 Illumina read sets.

Main Results:

• STRAIN demonstrated superior accuracy in allele assignment (97%) and novel allele identification (98.7%) compared to other tools.
• Existing methods showed lower capabilities, with correct allele identification ranging from 28.5% to 96.9%.
• Novel allele identification by other tools varied significantly, from 1.1% to 97.1%.

Conclusions:

• STRAIN is a highly accurate and efficient method for assigning sequence types (STs) and alleles from WGS data.
• The R package effectively identifies both known and novel allele sequences.
• STRAIN offers improved performance over existing tools and facilitates integration into downstream bioinformatics pipelines.