No biological tests are 100%. There are always some caveats; it’s a complicated process. Humans are probably the most complicated organisms on the planet. So we never have 100%. We will try to get 99% from our company in a fully penetrated single gene disorder. If we’re confident that these mutations will alter the protein, then we can give a more or less 99% score. The best way to prove something is to make an animal model, using exactly the same mutation and putting it in mice; we call it a humanized mouse. You genetically mutate the mouse in exactly the same position as the human. You will see if it’s male infertile, the same as the male or female infertile as the female. It’s probably one of the best ways to do it. But that would take years of work for each mutation, so we have to go on statistical predictions.
We use several reliable algorithms that can accurately predict whether a change in the DNA base will change the structure of proteins, whether that will inactivate the protein or remain active. We also use the rarity of the mutation because these mutations are naturally rare; if they were common, humans would be driven to extinction. We’d all be infertile; we wouldn’t have children; humans wouldn’t be here. So naturally, they are rare. We bring its rarity into factoring in our risk score. If a person has inherited two very rare mutations, either the same or different, then that’s a rare event; what were the chances of that? We would bring in forensics statistics that are used in rape and murder cases and except in court. We use them with the other algorithms to put them all together and develop a risk score.
