MGmapper: Reference based mapping and taxonomy annotation of metagenomics sequence reads.

Area of Science:

• Genomics
• Bioinformatics
• Microbial Ecology

Background:

• Next-generation sequencing (NGS) is crucial for species and gene identification in isolate and metagenomics samples.
• Existing taxonomic annotation methods often produce false positives, necessitating post-processing for accuracy.
• Reliable species and strain-level resolution is essential for accurate biological interpretation.

Purpose of the Study:

• To introduce MGmapper, a bioinformatics package for accurate taxonomic annotation of NGS data.
• To evaluate MGmapper's performance in species and strain-level classification using a mock bacterial community.
• To compare MGmapper with Kraken for taxonomic assignment efficiency and accuracy.

Main Methods:

• MGmapper processes raw NGS data for reference-based sequence assignment.
• A post-processing analysis filter is applied to refine taxonomic annotations.
• Performance was benchmarked using an in-vitro bacterial mock community (8 genera, 11 species, 12 strains).

Main Results:

• MGmapper achieved 100% correct taxonomy assignments at species and genus levels after post-processing.
• Strain-level annotations showed 75% sensitivity and precision.
• MGmapper utilized 84.8% of reads for species-level assignment, outperforming Kraken (70.5%), with no false positives for either method.

Conclusions:

• MGmapper provides a robust solution for reliable taxonomic annotation from NGS data.
• The package effectively minimizes false positives, ensuring high accuracy at species and genus levels.
• MGmapper offers a valuable tool for microbial community analysis and strain identification.