August 31, 2002
Cheap SNP assay technology as occasion for massive study of humanity

Our first challenge with cheap SNP assays is to make them meaningful. Wealthy people today can already afford to have their SNP assays done. But the results will not be of much use to most people because we do not know what are the practical consequences of the vast bulk of the genetic variations. One way to make SNP maps useful to us as individuals is to compare the differences in the maps of large numbers of people while simultaneously comparing as many differences between those people as can be measured.

In many disciplines people have been arguing about the relative importance of genetic variations vs physical environment vs social environment for a long time. Among the questions that may yield interesting and useful answers:

  • Are there genetic variations that put individuals at such risk for certain types of injuries that those who have those variations would be well advised to avoid certain kinds of sports?
  • Are there people have genetic variations that allow them to smoke a couple of packs of cigarettes a day without any fear of getting cancer? (not that I'm advising anyone to smoke - bad idea)
  • Are there people who can eat a fatty steak every day without raising their risk of colon cancer or heart disease?
  • Will maps of genetic variations turn out to be road maps to tell each person what their ideal diet is?
  • Is shyness genetically caused?
  • Is one's degree of religiosity influenced by genetic traits?
  • Are some people genetically fated to be able to be world class athletes? (I'm not one of them!)
  • Did Mozart and Bach have special rare combinations of genetic variations that made them musical geniuses?
  • How much does environment (both phyiscal and social) vs genetics cause intelligence differences?
  • Are there genetic variations that make it inevitable that certain people will become criminals?

The reason these debates have gone on for so long is that humans vary in so many ways genetically and in their environments (and by environments I'm including everything from diet, climate, home life, school, and countless other influences). Many of these variations happen independently to each other and others vary together for reasons that are hard to track down (does A cause B or B cause A or is something else causing them both?). Since we never could measure the genetic influences accurately no matter how much we've measured on the other factors we have never been able to control for genetic differences. So as a result we've never had sufficient data to resolve these debates to the satisfaction of everyone.

Advances in biotechnology will eventually (my guess is in this decade) allow the collection of complete Single Nucleotide Polymorphism (single letter variations in the DNA code that differ between people - SNPs for short) mappings from millions of people. While SNPs are not the only way that genetic sequences vary between people they certainly are a major source of genetic variation. To make this data useful for sorting out the relative contributions of genetcs, diet, education, upbringing, chance, and other factors many other pieces of data will need to be collected about each person at the same time

When genetic assays become cheap that will be occasion to perform the most massive medical and social science research project that has ever been conducted. A single massive study could resolve long-standing debates on many issues and to produce information that is useful across many disciplines. Medicine, political science, and psychology are just a few of the disciplines that could be expected to benefit immeasureably from a massive data collection project that combined complete SNP data with other data about a large number of people.. To make this work SNP data collection will need to be done in conjunction with collection of many other types of data about each participant of the study. Among the types of data that should be collected:

  • Medical records
  • various medical tests
  • educational histories
  • assorted psychometric measures (ntelligence, aptitudes, personality, and any other tests that might be useful)
  • criminal records
  • work histories
  • many individual preferences (in music, food, sports, hobbies, types of pets, favorite colors, etc)
  • detailed polls about religious and political beliefs
  • reproductive, marital, and relationship histories

Once this massive amount of data is collected many separate teams of epidemiologists, medical doctors, biochemists, geneticists, social scientists of various types (psychologists, political scientists, etc) will need to employ teams of programmers to develop software to do various analyses to tease out relationships between genes and outcomes. Some genetic variations and combinations of genetic variations will produce outcomes so strongly and consistently that their contributions to characteristics will be easy to establish. Other genetic variations will seem to tip the odds toward getting a particular disease or having a particular behavioral characteristic but by themselves will not guarantee a particular outcome.


By Randall Parker at 2002 August 31 03:57 PM 
Comments
Post a comment
Name:

Email Address:

URL:

Comments:
Remember info?



Go Read More Posts On FuturePundit
Site Traffic Info