Table 1_In silico performance of a targeted enriched metagenomics approach to infer Mycopl

A synthetic benchmark dataset of 620 Mycoplasma bovis whole-genome sequences, of which 162 (26.1%) originated from milk. The dataset, authored by Marit M. Biesheuvel and last updated in March 2026, contains results from an in silico evaluation of two classification tools (Kraken2 and Themisto/mSWEEP) on simulated targeted enrichment sequencing data for strain-level identification. Performance metrics include read classification accuracy, sensitivity, and positive predictive value across varying numbers of genomically clustered sequence variants (GSVs) and enrichment proportions.

Use Cases

• Benchmarking metagenomic classification tools based on simulated targeted enrichment sequencing data.
• Evaluating strain-level identification accuracy for Mycoplasma bovis based on performance metrics like sensitivity and PPV.
• Assessing the impact of mixed-strain samples on classification performance based on the simulation of 1 to 9 GSVs per sample.
• Studying the effect of sequencing enrichment proportion (0.3 to 0.9) on the recovery of strain-level genomic data.

Strengths

• Includes performance metrics for two classification tools (Themisto/mSWEEP achieved 84.9% average read classification accuracy).
• Based on a foundation of 620 M. bovis whole-genome sequences from NCBI.
• Simulation covers multiple experimental conditions, including 4 enrichment proportions and 4 levels of strain complexity.

Limitations

• Data is entirely synthetic, generated via in silico simulation of reads from selected genomes.
• Row count and column-level documentation are unknown, limiting suitability assessment.
• The biological relevance and practical application of the genomic clusters require further validation, as noted in the description.

Provenance

Source

Genome sequences sourced from NCBI; analysis and simulation by the author.

Collection Method

In silico simulation of targeted enrichment shotgun sequencing from selected M. bovis genomes, fragmented into 150-bp reads.

Freshness

Last updated 2026-03-18 15:43:10.