Skip to content

BUSCO stat precision request #265

Description

@jbh-cas

With long read quality and algorithmic improvements, assemblies often are receiving high BUSCO scores (especially when using compleasm or BUSCO with miniprot as the engine). For example, several avian genomes we have internally or that are GenBank RefSeqs have only 1, 2 or 3 Missing BUSCOs of the 8338 aves_odb10 lineage (and small numbers of Fragmented BUSCOs as well).

With single digit float precision for these, the snail plot displays 0.0% instead 0.01% or 0.04%, as the case would be using 2 digit precision. Looks like compleasm reports using 2 digit precision btw.

Also a Complete score of 99.95%, 8334 Complete out of 8338 in our case, reports as 100.0% Complete in the snail plot (blob). So rounding might still be a problem. We would not want 99.95% to be displayed as 100.00%.

Is it possible to increase the precision to 2 or show actual counts for Missing at least. The Missing count is our quick eval number and we miss not seeing that in the plot.

Removing trailing 0s could reduce spacing pressure if that is a consideration on modifying the display. (though I see a font_large var in the code but do not know if that is user-settable or would make any difference).

Thanks for any consideration and best,

Jim Henderson

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions