Traitwell's Future of Genomics Series: Trait and Disorder Description
Continuing our description of the future of genomics...
We’re continuing our analysis of the future of genomics. For those who are curious, please visit Traitwell.com. Be sure to check out our free apps.
Physical and mental health are traits like any other. So is behavior. They are all, as we have said above, under substantial genetic influence. Health traits can be comparatively simple and predictable in the case of Mendelian disorders, which are known to be causal; or more complex, since any number, from many to enormously many, of genes may have an influence. This last case is closely analogous to behavior, the difference there is that variation in behavior is not in itself a sign of disease or undesirable (we will treat mental pathology as health-related). Often people are just different. Mental abilities like general intelligence are included here—indeed they are predicted by some of the strongest genetic evidence to date—as are personality traits and other ‘personal differences’. Sporting abilities are included too, since those are a combination of mental and physical abilities strongly under genetic influence.
Many simple disorders can be detected from genome sequences and classified accurately. People inherently want to know about these, though each one is rare in the overall population, so unlikely to be present in a given individual.3 Empowering people to take an interest in their own health is an irresistible force. Existing genetic services only do this to a very limited extent. It is infeasible to continue to erect barriers between people and their own genetic information. Knowledge of genetics within the broader medical profession is known to be lacking (though users may be referred to specialists as a follow-up step). This sort of diagnosis will be democratized, and tools provided to help users understand the implications and possible actions for their case. The sophistication and usefulness of such tools, which can direct resulting actions by the diagnosed, will determine the success of services offering this information. This software approach will be a major growth area in the future.
More complex health and behavior traits are not yes/no diagnoses. Instead composite polygenic scores are calculated, and compared to population distributions where possible. (Sporting abilities fall under the same general scheme and considerations.) The percentile an individual occupies on that distribution is generally informative about how rarefied their case is. Again, people are naturally curious where they stand genetically. Where scores are not available yet, individual gene matches or simple counts may be shown. Several simplistic services already exist for describing this information to users. Knowledge of individual SNPs fuels zeal for further personal research, but that presumes considerable technical background.
The predictive power of scores where they can be calculated for an individual is modest at present, but this will improve over time.
There are legal applications of these trait descriptions that are becoming feasible. In order to determine compensation due to incapacitation caused by injury, or insurable events, it is useful to have an estimate of abilities prior to injury. Those measures may not be feasible using standard tests, like paper-and-pencil IQ tests, as the opportunity is gone: the subject is incapacitated. Genetic evidence about the native abilities of the individual maybe persuasive here, even at the purely descriptive level, as the best available estimate.
Such descriptive-only information drives initial adoption of services, but the next stage of growth requires more prescriptive value to be extracted from the genetic information. The aggregation techniques mentioned above are crucial for making such information actionable and therefore more valuable, but we will deal with those ideas separately.
Similar considerations apply to pet animals and livestock with regard to descriptive information. Pet owners are curious about their pets in the same way that they are curious about themselves, including behavioural traits. One should expect commercial expansion into this area in the short-term, though it will take more time for the science to mature—fewer studies have been done on animals to date. Tagging genetic groups will be useful here since general knowledge is already available about breed traits. Livestock requires more actionable trait information to be genuinely interesting and valuable to owners