H37Rv Genome: Empirical Base Pair Recall and Mappability Scores for 4.4 Million Positions

A table of 4,411,532 rows provides metrics for each base pair position in the H37Rv reference genome of Mycobacterium tuberculosis. The data includes Empirical Base Pair Recall scores from 36 clinical isolates, Pileup Mappability scores under three parameter sets, GC content in 100 bp windows, and putative low confidence position tags. Maximillian G. Marin of Harvard University created this dataset, which is associated with a paper on the paperswithcode platform.

Use Cases

• Identify putative low confidence genomic regions based on the PLC_Tag column.
• Assess sequencing coverage and recall variability using the EBR_36CI scores from 36 isolates.
• Compare sequence mappability under different k-mer lengths using the Pmap_K50_E4, Pmap_K100_E4, and Pmap_K150_E4 columns.
• Analyze local GC content patterns using the GC%_100bpWindowSize metric.

Strengths

• Contains one row for each of the 4,411,532 base pair positions in the H37Rv reference genome, providing complete positional coverage.
• Includes multiple calculated metrics per position: EBR, three mappability scores, GC%, and a confidence tag.
• Empirical Base Pair Recall scores are derived from 36 clinical Mycobacterium tuberculosis isolates.

Limitations

• Pileup Mappability scores were not calculated for the first k base positions of each parameter set due to sequence edge effects.
• Column-level documentation is absent; field semantics must be inferred from the description after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Harvard University

Collection Method

Calculated metrics from the H37Rv reference genome (NC_000962.3) and analysis of 36 clinical isolates.