We would like to analyse thousands of E. coli genomes and identify orthologous genes. Before this we would like to filter the bad quality genomes by BUSCO (https://busco.ezlab.org/). What % of BUSCO genes should we accept to consider the genome as "good quality"?