The article, DNA Variations That Make Us Who We Are talks about how small differences in the each person's DNA sequence make each individual different. However, how do scientists figure out which small differences in the 3 billion nucleotide base sequence of human are related to a particular trait, such as a predisposition to a disease?
The First Breast Cancer Genetic Connection
Some gene variations have been connected with particular diseases though intensive research combined with some luck. For example, two genes responsible for a high percentage of breast cancers, BRCA1 and BRCA2, were found in the 1990s, even before the human genome was fully sequenced. This discovery required many years of efforts analyzing DNA from families with a history of breast cancer.
Finding the BRCA Mutations Easier Than Most
However, it is rare to have single gene variants that produce such an obvious trait, like the early development of breast cancer, in high percentages of a small genetically related group like the families were in the BRCA studies. While the discovery of mutations in the BRCA1 and BRAC2 genes was a breakthrough since women with one of the mutations have about a 60% chance of developing breast cancer, actually these particular mutations only account for about 10% of breast cancers. There are likely other genetic mutations affecting the other 90% of breast cancer cases, probably many dozens at minimum, and they are harder to find than the BRCA genes since they are more spread out in the population.
What Makes Finding Mutations for a Disease So Difficult?
How do researchers find the single DNA nucleotides that are different in the 3 billion nucleotide bases that make up all the DNA in human cells (i.e., the human genome)? Well, first, just to briefly explain what most mutations are, DNA is a long polymer sequence built with four different nucleotide bases: adenine (A), cytosine (C), guanine (G),and thymine (T). Each "link" of the polymer chain has one of these nucleotides. Any two humans have gene sequences that are mostly identical (i.e., the same nucleotide base links), except occasionally—like maybe every thousand bases or so—there might be substitution of one nucleotide for another (e.g., A for T, G for C, etc.). This one letter difference is the type of mutation that researchers need to notice when they are trying to find genetic differences between individuals.
To simplify a bit the hunt for important mutations that influence some characteristic, researchers typically narrow the search down to just the active or expressed genes in the genome. This set of expressed genes is known as the exome, and cuts the number of nucleotides down to only about 60 million. Still not a cakewalk by any means, but a manageable number to analyze in a reasonable time using current DNA sequencing technology.
Even with limiting the analysis to the exome DNA sequences, however, finding the important single nucleotide base changes related to a specific trait is a real challenge. By simply looking closely at millions of DNA nucleotides, it is possible to find many of the occasional one-in-a-thousand base differences with a little patience. The problem, though, of course, is figuring out which ones are related to the trait being looked at. Which are associated with early breast cancer development, obesity, attention deficient disorder, a neurological disease, or even just a tendency to dislike some type of food?
How Are Gene Association Studies Done?
The approach to find the important DNA mutations for a specific disease or other trait is relatively straightforward in concept. Simply, look at the genes from a bunch of individuals with a certain characteristic, and see which particular DNA differences—that is which DNA nucleotide substitutions—appear more frequently in the affected population as compared to the general population. For example, it might be found that individuals with a sensitivity to penicillin usually have mutations in a couple of genes. Since a higher incidence of mutations in these genes has been associated with people that are penicillin sensitive, people with the particular gene variants are more likely to be sensitive to this drug. It also indicates that these genes might be involved in actually causing penicillin sensitivity, but this sort of causal connection can only be shown with further research.
The approach is not too different than the way the BRCA1 and BRCA2 breast-cancer mutations mentioned above were tracked down. However, with advances in DNA sequencing technology, it is possible to analyze many more genes faster and more precisely now than when the BRCA genes were discovered. These developments allow gene association studies to be done with a lot more people which makes it possible to find difference affecting fewer individuals.
Gene Associations Are Sometimes Not Straightforward
Gene association studies, however, can be tricky. Genetic associations just indicate that a test group which has an interesting trait, such as a disease, drug sensitivity or specific behavior, also has a tendency to have particular DNA mutations. It is just a correlation. Unless something is already known about how the gene with the mutation functions, it is not possible to establish a real causal biological connection between the gene mutations and a specific characteristic being investigated. In other words, people with certain mutations may have a tendency to develop schizophrenia but, unless the genes with the mutations have been shown to make or regulate a protein involved in brain function, or something connected with the physiology of schizophrenia, then the association is just two independent observations that seem to occur together more often than by chance. However, it is very much an overreach to say that the mutated gene causes schizophrenia without additional information.
Also, unless gene association studies are done very carefully, there may be some other aspects in common in the test group that can throw off the results. These are called confounders. For example, if all the individuals in the test group looking at penicillin resistance also happen to have dark hair, it is likely that some of the mutations that appear significant this group have to do with hair color, not penicillin resistance. This, of course, is a simplistic example, but it can be easy to overlook commonalities in a test group that can mess up the analysis.
Huge Research Efforts Focus on Genetic Associations
Much of current genomic research is focused on identifying which genetic changes correlate to phenotypic disorders. The sessions and poster boards at the yearly American Society of Human Genetic conference are filled with gene association studies. The recent 2012 conference had sessions on the genetics implicated in a wide range of human disorders, including cardiovascular disease, autism, cancer, metabolic disease, and neurologic disease. For information on published gene associations, though, the best resource is probably the Genetic Association Database which currently has over 130,000 entries.
The Advent of Personalized Medicine
Still, despite the massive amount of research, only a fraction of the several million DNA variations each of us possess have been linked to specific physical characteristics. Also, it is only a fraction of these that have associations in which we can explain the underlying biological connection between the gene and the characteristics it influences—the link between the genotype and phenotype in technical terms.
Even at this point, however, genetic associations are starting to impact healthcare. There are a number of gene association studies currently in clinical trials as to evaluate their utility as disease diagnostics, and companies such as Pathway Genomics and Personal Genome Diagnosticshave begun providing diagnostic tests to assist medical professionals in evaluating their patients' health risks and predispositions. As genetic research continues, its impact on personal healthcare will only expand.