Reducing mapping reference and lineage bias in Mycobacterium tuberculosis.

Area of Science:

• Genomics
• Bioinformatics
• Microbiology

Background:

• Whole-genome sequencing (WGS) generates extensive genetic data.
• Accurate genomic variant inference is crucial for clinical and epidemiological studies.
• Comparative genomics of Mycobacterium tuberculosis often uses short-read mapping to a single reference, risking mapping bias.

Purpose of the Study:

• To analyze the impact of mapping bias in Mycobacterium tuberculosis comparative genomics.
• To propose and validate a novel hybrid workflow to overcome mapping bias.
• To improve the characterization of genomic variations and lineages in M. tuberculosis.

Main Methods:

• Analysis of mapping reads from diverse M. tuberculosis lineages to the H37Rv reference.
• Development of a three-step hybrid workflow: de novo assembly, alignment to reference, and read mapping to the aligned assembly.
• Evaluation of bias correction and improved characterization of lineages and hypervariable regions.

Main Results:

• Mapping bias was observed to vary by M. tuberculosis lineage and gene.
• The proposed hybrid workflow significantly corrected lineage and gene-specific mapping biases.
• The new approach enhances the characterization of bacterial genomic diversity.

Conclusions:

• Standard short-read mapping can introduce significant bias in M. tuberculosis comparative genomics.
• The novel hybrid workflow effectively mitigates mapping bias, leading to more accurate variant detection.
• This method facilitates deeper understanding of genetic variations in M. tuberculosis and other bacterial pathogens.