One of the great questions for humanity, one which reaches to the center of what makes humans the sort of creatures they are (whether one takes them to be rational animals, spiritual animals, or determinalistic animals) is why we contemplate religious questions.
Anselm's Ontological Argument is perhaps the classic Christian attempt to address this question, stating essentially that God is that thing which is the greatest and most perfect thing imaginable, and since existing is more perfect than being imaginary, God must therefor by definition exist.
In apologetics terms, many authors talk about how the "God-shaped hole" in the human heart is a key piece of evidence that we are creatures who yearn for our creator.
Nor is this question strictly of interest to theists. One of the great questions for strict materialists is: If there is no such thing as a God or a spiritual realm, why does nearly everyone think there is? Those with an interest in finding evolutinary explanations for such things, such as Daniel Dennett and Richard Dawkins, have put a fair amount of work into coming up with explanation as to how developing a religious sense would have been biologically advantageous (even if philosophically erronius) for early humans. In this materialist conception, there must be some sort of 'religion gene' which appeared at some point in early human history and resulted in better survival among humans with a religious sense than without.
Now, I'll admit to having little truck with anything that smacks of evolutionary psychology, but there's something rather deeper that strikes me as odd about the idea of a 'religion gene'. It is, I think, the implicit assumption that if one could somehow come up with some sort of apparently biological explanation for why people are able to think about religiousmatters, that this would somehow rule out the possibility that religion is false. Religions are systems of truth claims. To be Catholic is to endorse the claim that certain things are true. To be Bhuddist is to endorse the claim that certain other, contradictory, things are true. However, the claims themselves must have a truth or falsehood separate from anyone's ability to evaluate them or be interested in them.
Once upon a time (when the ancestors of human beings were not yet imbued with reason) there was no creature on the Earth capable of understanding mathematics. However, the fact that no one was capable of grasping what a sphere was or how addition worked in no way detracted from the truth of mathematics. Nor would it disprove the truth of mathematics if a "math gene" or "abstract thought gene" were responsible for making humans capable of grasping mathematical realities. In this sense, the human discovery of mathematics (whether one chooses to think in terms of a 'math gene' which benefited humans' survival or in terms of humanity being imbued with reason and thus able to contemplate the world) thus fits in a very Platonic mode, in that it involves the discovery by the mind of a reality deeper than the physical instantiations of mathematical concepts.
Now, if one has already concluded that there is no God, the 'religion gene' idea perhaps provides some insight into why most humans have always believed in some sort of supernatural level of existence. But I can't see that the idea in any sense disproves the existence of the supernatural -- which is what many of its supporters seem to believe it does. No more, at any rate, than the exitence of a 'math gene' would prove that numbers don't exist.
If there is such a thing, and I must say that I find the idea of looking for genetic origins for thought rather odd and unhelpful, its existence doesn't speak to whether there is or is not a God, merely how it is that humans came to ask the question. The question of God's existence is a question about reality and thus independant of whether we ask ourselves the question or not.