Gene Model Translation / Correspondence

  • Background and methods

    The set of allelic genes found in multiple individuals in a species or closely related species may be called a "pangene set," with the gene models that correspond by homology and position being called a pangene. The pangene set calculated for Arachis accessions at PeanutBase can be used to find corresponding genes across assemblies and annotations.

    If you have one or several (fewer than 100) genes to look up, use the Pangene Lookup tool below. This page accepts a list of genes (separated by spaces or line returns).

    If you have hundreds or thousands of genes to look up, you can download a correspondence table for for all pangene accessions.

    The method for generating the pangene correspondences is described briefly here:

    The Pandagma software package (Cannon, Lee, Weeks, Berendzen) was used to identify pangene and gene family sets. The main steps in Pandagma's pangene process are:

    • Make pairwise homology comparisons between each annotation set;
    • Filter by provided percent identity and coverage parameters;
    • Identify synteny blocks among all annotation sets;
    • Cluster genes in synteny blocks;
    • Add back remaining genes based on homology, constraining by chromosome (e.g., chr1 genes to chr1 clusters)
    • Add "extra" annotation sets (those with more fragmentary assemblies or questionable annotation quality) to clusters identified above.

    The Pandagma package is available at https://github.com/legumeinfo/pandagma, including the configuration used to calculate the pangene data above.

    If you have extensive programmatic work and need to translate among arbitrary accessions, the gene_translate.pl utility in pandagma may be helpful.