AAI using MAGs

[blancaverag]

I have a question regarding the use of AAI for assessing genomic relatedness of metagenome-aseembled genomes to isolates' ones. I thought that I would rather post it here even if it is not purely software-related, just in case it is useful for someone in the future.
I haven't been able to find a value for the minimum completion of a MAG required in order to AAI results to be reliable. For ANI, I have read that having a 20 % of genome coverage should be enough to confidently use it (Richter & Rosselló-Móra, PNAS 2009). Since these indices are especially relevant to evaluate the relatedness of environmental genomes sequences to already known and formally described ones, it would be extremely useful to know how they perform on incomplete genomes.

I wonder whether someone has experience on it and can propose some kind of quality threshold for environmental genomes to be amenable for reliable AAI calculation.