Charles Darwin once said, in his 1859 publication, 'On The Origin Of Species', "Can we doubt […] that individuals having any advantage, however slight, over others, would have the best chance of surviving and of procreating their kind?" The depth of his brilliant foresight into the of the inner-workings of the natural world, the means by which organisms propagate themselves, and that they can even change over time, was perhaps among the most grand of all philosophical thought, and certainly more scientific - it was, after all, the ultimate conclusion of natural philosophy. It paved the road ahead for future scientific inquiries into the natural realm, and firmly demarcated the boundary of Biology, and natural philosophy. However, Darwin was not the only person to have such thought; indeed, Alfred Russel Wallace, a contemporary of Darwin, "converged" upon the same solution independently(I do intend the pun), and around the same time. They had both been influenced by a book, "An Essay on the Principle of Population", authored by Thomas Robert Malthus, which detailed the discrepancy between increasing food supply, and exponential population growth, and in a mathematical fashion(succinctly put: the supply versus the demand). Now we're dealing in numbers, not just eruditic, well-educated guesses - this is a hard science; it is now daring to lend itself to a subject wherein the closest comparable definition of an actual hard science was, perhaps, Taxonomy: the hierarchical categorization of living things within an ascribed taxonomic ranking, ultimately dealing with an individual unit of Taxonomy: the species. What about the individual unit, the instrument of 'selection', in Darwin's thought? That would be the species - but yet, crucially, he realized that it was not just the species; it was the individual trait that was at play.
About traits, there is one in due time I should mention: sexuality. We all know what sex is - it is where all the gene-swapping takes place, and it is necessary. It is crucial because, genes are subject to damage, and genetic recombination, that very gene swapping, is the only thing that prevents - nay, ameliorates - the super gradual accumulation of deleterious copying errors that are the by-product of repetitious cell division. I say 'super gradual', because the accumulation is very, very slow. In Eukaryotes - in multicellular life - such errors are nevertheless many; they are legion. Now, a happy(sometimes tragic) quirk of gene-swapping is that the fitness of other traits may be preserved, even if these are not directly expressed. Such traits are genotypic; one gene possesses, or directs, the expression of those given traits.
A gene will have alleles, phenotypes of that genotype. A phenotype is the observable, or measurable expression of an individual trait in question, where both are ultimately the consequence of alleles from that same gene - this defines a given genotype. Indeed, the genotype of an individual may be Trait1/trait2, but you only typically observe the character of one, the dominant trait(yes, it helps to only capitalize the dominant trait - it tends to take charge, after all!).
Hence, the distinction between phenotype and genotype. Both phenotypes are not usually expressed simultaneously, unless they are co-dominant(this is admittedly weird in more ways than one, of course). Gene swapping - genetic recombination - occurs with only two pairs of alleles; yet there are typically many more than two alleles, or two traits, for each given gene among a species population. An individual possessing two traits, that is, one copy of each allele, is said to be heterozygous.
Suppose that something should favor one phenotype over the other(red coloration, say), but might still favor the other phenotype(say, orange coloration). On average, red will come to dominate, but never entirely, because orange still does well at times. Where am I going with this? There is no one gene typifying, or directly influencing the outcome, or prevalence of human sexuality. I should highly doubt, and find skepticism of any such reported finding. Not because it isn't realistic, no, it just isn't necessary. We are social beings; we have civilization, and thus sex itself is not as necessary to our survival, and this is by virtue of our unique position; we don't have to propagate all the time. Pretty nifty, actually. It is because, rather crucially, behaviors, and thought patterns by means of extension as a downstream result of a gene, or genes, ultimately direct our behavior; imparting incidental fitness to those behaviors, and so also to the genes that 'bear' those behaviors - us.
That is not that we choose to be gay, or that we choose to be bi; it is that these permutations yield to a position in society - a position analogous to a niche - that has, ultimately, incidental reproductive fitness, even if that fitness is twofold: One, the genes that permit a trait, or traits conducive to any bisexual/gay proclivities, and any fitness thereof, will be latent in some, expressed in others, and more-or-less fractionally distributed throughout populations in what I may term a pseudophenotypic distribution. Two, these 'phenotypes' are pseudo because they are elegantly interdependent upon both their innately latent nature - their genetic aptitude - and environmental cues that provide, or not, for them to flourish - their nurture. Three, these behaviors are nonetheless replicators - like genes, but symbiotic, and separate. It - (bi-, etc.)sexuality - is the result of a cultural replicator, wherein behavior(s), and a biological replicator are acting in unison, which incidentally promulgates that behavior. I endeavor to note, however, that a pseudophenotype is distinct from Richard Dawkins' extended phenotype; the concept of an exemplary, contemporaneous evolutionary thinker, from whom I derive considerable influence. It is distinct in this regard: The organism, and it's behavior benefit in a manner such that the benefit to the behavior is greater, and disproportional to that of the organism. These three supposed rules are complementary, and complete a mutually-reinforcing cycle. This is the basic foundation of my hypothesis, which is by no means complete.