Is there an “obesity gene?” Probably not. However, according to statistical geneticist Dr. Nilanjan Chatterjee, what we have instead is something that sounds very much like a far-reaching, all-encompassing synergy that makes a human body react in various ways. This has to do with the relationships between thousands of SNPs, pronounced snips, which are single nucleotide polymorphisms. The SNPs interact with each other in mind-bendingly complicated ways.
Several years ago, Dr. Chatterjee called attention to genome-wide association studies, or GWAS, as they are known in the trade, with their implication of “an extremely polygenic model for complex diseases.” He wrote,
Such models may involve thousands of susceptibility markers, each conferring only a modest risk, but collectively they could be explaining substantial variation in disease-risks in populations.
And this teaching page says,
Several large-scale projects have combined the efforts of many institutions to identify and catalog all of the SNPs in the 3-billion-base pair human genome. Each project involves hundreds of scientists, who compare the genomes of numerous individuals to identify the differences. These comparisons require a lot of computer-powered data analysis. As they work, scientists sort and catalog their results in databases that are available to anyone over the Internet, including other scientists and you.
In 2017 Dr. Chatterjee co-authored “Power Analysis Provides Bounds for Genetic Architecture and Insights to Challenges for Rare Variant Association Studies,” from which this is quoted:
Typically, power calculations for such tests require specification of large number of parameters, including effect sizes and allele frequencies of individual variants, making them difficult to use in practice. In this report, we approximate power to varying degree of accuracy using a smaller number of key parameters, including the total genetic variance explained by multiple variants within a locus. We perform extensive simulation studies to assess the accuracy of the proposed approximations in realistic settings.
You get the picture: It’s complicated. For anyone who really wants to go deep, the Dr. Chatterjee’s homepage is recommended.
The “power analysis” used in these studies is number-crunching on a massive scale. In spirit, it resembles a brute-force attempt to discover a computer password by hurling millions of possible combinations at the besieged machine. If the software is competently written, and the electricity stays on, sooner or later the safe can be cracked. Another comparison might be to a green vegetable. It requires plenty of spinach leaves to make a cup of spinach juice.
As writer Robin Scullin explains, the scientists utilized new statistical techniques to estimate the numbers of DNA variations that combine to create physical characteristics, traits, and diseases in the body. When you’re working with a thousand variables, it takes lots and lots of data to reach the point where the data can be presumed to mean anything.
But, wouldn’t you know it, the project eventually runs into another universal law known as “too much of a good thing.” Sample size is crucial. Like Goldilocks, you want enough, but not an excess.
This is scientifically known as the diminishing returns effect. Scullin writes,
For example, projections in the study suggest that for most traits and diseases […] the point of diminishing return for these studies only starts after a sample size reaches several hundred thousand. For psychiatric diseases and cognitive traits, with their “long-tail” distributions of gene effects, diminishing returns usually won’t kick in until sample sizes are even larger — possibly in the millions.
This bring us back to perhaps the most important discovery so far. For conditions that are more on the physical side, it takes fewer SNPs to bring on or exacerbate them. But with the mind, a whole different order of magnitude sets in. In areas like cognition and mental health, things happen with the collaboration of a much bigger number of SNPs.
Your responses and feedback are welcome!
Source: “Nilanjan Chatterjee,” UGA.edu,” 09/18/14
Source: “Making SNPs Make Sense,” Genetics.utah.edu
Source: “Power Analysis Provides Bounds for Genetic Architecture and Insights to Challenges for Rare Variant Association Studies,” bioRxiv.org, 04/21/17
Source: “Study Finds Genetic Basis of Common Diseases May Span Tens of Thousands of SNPs,” GenEngNews.com, 08/14/18
Source: “Understanding genetic architecture of different traits and diseases could lead to better-designed studies, researchers say,” JHU.edu, 08/13/18
Photo credit: Alex Sar on Visualhunt/CC BY