There might be an easier way, but one suggestion is:
Do a Pawley fit to your raw data. This will produce a list with h, k, l, intensity in the INP file. You could then change the intensities to fixed values you calculate from your mem F's. Then do a single calculation to get the ycalc plot.
You'll need to be careful about what intensity corrections need to be applied. One way to check would be to get topas to output F's from the fixed intensity refinement to make sure they're what you'd expect.