There are genes that are on / off indicators but you are already likely to have found out about those genes by already suffering their ill effects. Eg haemophilia, cystic fibrosis, muscular dystrophy. There aren’t many genes that are suddenly activated later in life to cause a disease (Huntingdon’s, breast cancer – Braca1/2?) which it’s going to be useful to you to find out about when doing a 23&me test in your 40’s.
There will be other genes which may interact with other genes and environmental factors to increase the chance of something but this is hideously complicated and something we’re only just beginning to have a miniscule understanding of.
As an analogy think of a computer program with 50000 lines of code which runs in duplicate with another almost identical program with 50000 lines of code. Each line of code may have some affect or none on the other 99999 lines of code. Each line of code can also be affected by any number of external factors and the affect can change depending on the timing of the external factor. How many different feedback loops could you find in this system?
This makes it incredibly difficult to then conclude that the presence of gene x will increase your chance of disease y by z%.