The idea that it has at least a fair amount to do with epigenetic issues in the womb makes HUGE sense to me, as statistical studies show consecutive male births (no intervening females) increase the likelihood of each successive male being gay:
Writing in The Quarterly Review of Biology, researchers William Rice, a professor at the University of California, Santa Barbara, and Urban Friberg, a professor at Uppsala University in Sweden, believe that homosexuality can be explained by the presence of epi-marks — temporary switches that control how our genes are expressed during gestation and after we're born.Specifically, the researchers discovered sex-specific epi-marks which, unlike most genetic switches, get passed down from father to daughter or mother to son. Most epi-marks don't normally pass between generations and are essentially "erased." Rice and Friberg say this explains why homosexuality appears to run in families, yet has no real genetic underpinning.Epigenetic mechanisms can be seen as an added layer of information that clings to our DNA. Epi-marks regulate the expression of genes according to the strength of external cues. Genes are basically the instruction book, while epi-marks direct how those instructions get carried out. For example, they can determine when, where, and how much of a gene gets expressed.Moreover, epi-marks are usually produced from scratch with each generation — but new evidence is showing that they can sometimes carryover from parent to child. It's this phenomenon that gives the impression of having shared genes with relatives.