What if the Geneticists are Wrong?

Genetic disease may be as different as different kaleidoscope pictures. This complicates finding the cause of most genetic disease.

Scientists have thought that they had a pretty good handle on how genetic diseases work. A certain DNA difference causes a certain gene to work incorrectly leading to a specific genetic disease. And importantly, people with a disease tend to share the same DNA difference.

A disease like sickle cell anemia (SCA) is a perfect example. A key DNA difference in the hemoglobin gene causes sickle cell anemia and many people with SCA have the exact same difference.

But what if diseases like SCA are the outliers? What if instead most people or families had unique DNA differences that led to their disease? Then scientists have been going about finding the causes of genetic disease in the wrong way.

Right now scientists tend to do something called genome wide association studies (GWAS) to find the causes of genetic diseases. Basically scientists compare the DNA of people with a disease to people without the disease. If they compare enough people, they should be able to pinpoint the differences that matter.

This strategy is based on people with the same disease having the same DNA differences. If people have unique differences, then this strategy won’t work. And so far, most GWAS studies have not found a lot.

They have tended to find a few DNA differences that can explain a very small part of any disease when they look at related groups like Icelanders or Northern Europeans. And even these small effects sometimes go away when scientists try to apply them to larger, more diverse groups.

This is just what you might expect with unique, recent DNA changes causing disease. And if you stop and think about it, the idea of lots of unique DNA changes leading to common diseases makes some sense.

Each person in each generation has around 175 new DNA changes that their parents didn’t have. Spread out over the human race over hundreds or thousands of years, that is a lot of relatively recent differences.

Any DNA change that has too severe an effect on a person shouldn’t survive long in human DNA. People with the change will be at a disadvantage and so won’t do as well as people without the change. Over time, the change will disappear.

The exception to this is if the change can have both good and bad effects. SCA is obviously bad but in areas with a lot of malaria, it is actually helpful. People who carry SCA but do not have the disease are more resistant to malaria and do better. Most bad DNA changes won’t come with benefits though.

So instead of a set of static DNA differences that are passed down we might need to think about human DNA as constantly changing. Many DNA differences come and go over time. In this scenario, most human disease is caused by DNA changes in the process of being eliminated. Or they are neutral changes that have turned bad now that we live in a new environment. (Think having lots of food, living longer, having fewer babies, etc.)

If this is how things work, then GWAS won’t find much. And as I said before, they really haven’t. But luckily there are other ways to skin a genome.

Even if people have unique DNA changes that lead to disease, there will be overlap in the affected genes. One approach is to look at the DNA of people with severe forms of a disease to find the gene involved.

Then you’d have to sequence the whole gene and look for changes. The sequencing will be easy…we can sequence a gene in no time these days. The hard part will be figuring out which changes matter. There are bound to be lots of differences in most genes with hardly any of them having any effect. So scientists will need to come up with some way to tell which changes matter.

If this is how genetic disease works, more traditional sorts of DNA tests are in real trouble. And we’ve spent an awful lot of time and money heading down blind alleys. But that’s how science works, two steps forward, one step back.

A couple of in-depth looks at this problem:

Genetic Heterogeneity in Human Disease

Beyond genome-wide association studies

37.7749295 -122.4194155