.. I would also expect this to be a low-research area, but not due to sounding silly, but rather due to high-noise datasets. People lie about their sexual desires a lot. This particular desire is even more likely to be denied or concealed from researchers than most, so I would expect most people setting out to look into this to get to the "Design data-collection protocol" stage, acquire a monumental headache, and then go research which kind of diet is easiest to stick with instead.