I did a linear conversion by taking Ilford's full scale values and dividing, worked fine.
However, no matter what method you'd use, you will need to do some testing. If you want to get to the exact grades, it's helpful if you have a set of Ilford filters, you can make a set of exposures with the gel filters, and another with your calculated settings and compare.

A much easier method is to treat it arbritarily which is what it is really anyway. Your end points are fixed, that is, you get the minimum possible contrast with full yellow, and max with full magenta. Everything in between is more of one and less of the other. So if your print needs less contrast take away some magenta and add yellow, or the other way around if you need more, what grade you're at doesn't really matter.

Make some tests, keep track of the filtration settings so you can use them as reference points, then forget about grades and just print, this is the beauty of variable contrast papers.