I just pulled this from the Wikipedia entry for "Causes of Transsexualism." I haven't read every line of it, but it is certainly suggestive that there may be biological-based factors that lead to it. It is more complicated than whether one has an X or Y chromosome.
The androgen receptor (AR), also known as NR3C4, is activated by the binding of testosterone or dihydrotestosterone, where it plays a critical role in the forming of primary and secondary male sex characteristics. Hare et al. found that male-to-female transsexuals were found to have longer repetitions of the gene, which reduced its effectiveness at binding testosterone.
A variant genotype for a gene called CYP17, which acts on the sex hormones pregnenolone and progesterone, has been found to be linked to female-to-male transsexualism but not MTF transsexualism. Most notably, the FTM subjects not only had the variant genotype more frequently, but had an allele distribution equivalent to male controls, unlike the female controls. The paper concluded that the loss of a female-specific CYP17 T -34C allele distribution pattern is associated with FtM transsexualism.
In a first-of-its-kind study, Zhou et al. (1995) found that in a region of the brain called the bed nucleus of the stria terminalis (BSTc), a region known for sex and anxiety responses, MTF transsexuals have a female-normal size while FTM transsexuals have a male-normal size. While the transsexuals studied had taken hormones, this was accounted for by including non-transsexual male and female controls who, for a variety of medical reasons, had experienced hormone reversal. The controls still retained sizes typical for their gender. No relationship to sexual orientation was found.
In a follow-up study, Kruijver et al. (2000) looked at the number of neurons in BSTc instead of volumes. They found the same results as Zhou et al. (1995), but with even more dramatic differences. One MTF subject who had never gone on hormones was also included, and who matched up with the female neuron counts nonetheless.
In 2002, a follow-up study by Chung et al. found that significant sexual dimorphism (variation between sexes) in BSTc did not become established until adulthood. Chung et al. theorized that either changes in fetal hormone levels produce changes in BSTc synaptic density, neuronal activity, or neurochemical content which later lead to size and neuron count changes in BSTc, or that the size of BSTc is affected by the failure to generate a gender identity consistent with one's anatomic sex.
In a review of the evidence in 2006, Gooren confirms the earlier research as supporting the concept that transsexualism is a sexual differentiation disorder of the sex dimorphic brain. Dick Swaab (2004) concurs.
In 2008, a new region with properties similar to that of BSTc in regards to transsexualism was found by Garcia-Falgueras and Swaab: the interstitial nucleus of the anterior hypothalamus (INAH3), part of the hypothalamic uncinate nucleus. The same method of controlling for hormone usage was used as in Zhou et al. (1995) and Kruijver et al. (2000). The differences were even more pronounced than with BSTc; control males averaged 1.9 times the volume and 2.3 times the neurons as control females, yet once again, regardless of hormone exposure, MTF transsexuals lay within the female range and the FTM transsexual within the male range.
While the resolution of MRI tomographs in general can be fine enough, independent nuclei are not visible due to lack of contrast between different neurological tissue types. Therefore such images do not show detailed structures such as BSTc and INAH3, and studies on BSTC were done by bisecting brains postmortem.
However, MRI does much more easily allow the study of larger brain structures. In Luders et al. (2009), 24 MTF transsexuals not-yet treated with cross-sex hormones were studied via MRI. While regional gray matter concentrations were more similar to men than women, there was a significantly larger volume of gray matter in the right putamen compared to men. As with many earlier studies, they concluded that transsexualism is associated with a distinct cerebral pattern.
An additional feature was studied in a group of FTM transsexuals who had not yet received cross-sex hormones: fractional anisotropy values for white matter in the medial and posterior parts of the right superior longitudinal fasciculus (SLF), the forceps minor, and the corticospinal tract. Rametti et al. (2010) discovered that, "Compared to control females, FtM showed higher FA values in posterior part of the right SLF, the forceps minor and corticospinal tract. Compared to control males, FtM showed only lower FA values in the corticospinal tract."