Without an exhaustive review of the implications of Bayes' Theorem, suffice it to say that this isn't necessarily a good idea. The short version is: if you are unlikely to have a medical condition, a positive result may well be a false positive.
Much like a full-body MRI with no medical indication, there is a risk of a false positive leading to more expensive, unnecessary, and possibly dangerous procedures to chase down a condition that doesn't exist.
Going "directly to the lab" is, for many people, predicated on an overly simplistic and idealistic understanding of how lab tests work. The context of the result (health history, family history, physical exam, etc.) is important for accurate interpretation.
Finally, some tests border on the fraudulent. I'm a psychiatrist, and increasingly tests to make a psychiatric diagnosis based on a genetic screening are being marketed directly to patients. These tests are not generally accepted in the medical community as being accurate. In fact, some them are very likely to be totally misleading.
Caveat emptor.