If at least one version of the gene is present in the chromosome [...]

This is so wrong, that I had to go and check the news from University of Bialystok site, too. And it's even there. Must be bad translation, or something. Because everyone has at least one version of the gene, most people have two.

Elsewhere they did say they identified a variant that doubles the risk of severe covid infection. As all this was announced in a press conference, not in a peer reviewed publication, we don't really know what methodology they used. If was "merely" an association analysis, they can well be way off, since it's basically just regression analysis done the wrong way. There's always a big chance that the "association" is just a statistical fluke due to the method doing it's best to find something.

The best would be for somebody to come up with a testable biochemical explanation why a certain variation of a certain gene raises the risk of severe infection. The we would, perhaps, learn how to mitigate it in all patients.