Prediction of risk
Analyses of a number of parameters such as blood sugar or white blood cell count are used to indicate the potential presence of a disease, e.g. diabetes and leukemia. However, in many cases a predisposition to developing one of several diseases is present in the genome for the very beginning (present in the embryo).
We have currently a huge list (according to the World Health Organization WHO more than 10,000, http://www.who.int/genomics/public/geneticdiseases/en/) of particular genomic mutations (often single nucleotide polymorphisms, SNPs, or short deletions or duplications) for which we know by experience that they are correlated with an increased risk to develop a particular disease. Well-known examples would be breast cancer, sickle cell anaemia (a deformation of red blood cell into a sickle form, which became endemic in Africa as it provides a partial protection against malaria), various forms of Asthma, and the juvenile type 1 diabetes. These diseases are all of the type of monogenic diseases, i.e. a single mutation in a single gene is sufficient to put the individual at an elevated risk for the associated disease. However, only in rare cases such mutations lead inevitably to the disease. In most cases people can stay healthy even when a disease-associated mutation is present in their genome. Reason are the diploid of the human genome, i.e. we have two copies of most genes and that truly monogenic diseases are extremely rare, in most cases a particular genetic background (i.e. other often unknown mutations) are required in addition to the tell-tale mutation before the actual disease manifests.
Complex or multigenic diseases
As always also genetically caused diseases fall on a continuum with respect to genes involved. From a count of 1 (true monogenic) to multiple genes (usually the exact numbers are unknown) every scenario is possible. Since it is possible in many cases to counteract a genetic risk by lifestyle, early knowledge of such preconditions can be vital. That is the reason why every newborn is automatically checked for a limited number of genetic risk factors (the well-know prick in the heel).
By comparing the sequence of the individual with healthy and known disease-risk variants it is possible to make a estimate of how plausible it is that this individual may develop the disease later on.
Figure 9: Prediction of genetic risk; example single mutation
Selected mutations vs all mutations
Modern molecular genetic diagnostics could use so-called whole genome-sequencing revealing not just a few but every mutation or variant in the whole genome. Ideally, this would cover all genetic risks of that person known and unknown. This “unknown” part represents the major problem with that approach. We all carry about 30 million genetic differences to other human beings, many of which are simple variants that may determine eye, skin or hair color without presenting a (known) disease risk. Unfortunately, we do not know in the vast majority of cases what purpose (if at all) a particular variant fulfills and whether there is a disease associated with it. Only in few cases this correlation is known. Thus, routine use of whole genome sequencing is not only avoided du to the high cost. A major reason is our inability to correctly interpret the findings.
Biomedical progress should not bypass currently living generations
Nevertheless, I would strongly advocate to make whole genome sequencing a routine test even now. The reason is simply that we discover more and more disease-associations as science progresses. Today, we can check for a multitude of disease-risk genes that were unknown a decade ago and this development continues. Once the whole genome sequence of a person is known, checking for newly detected risk-mutations is a lightning fast lookup, which can be fully automated. This way everybody could be kept current with respect to the detection of genetic disease risks.
What’s coming up next?
Next week I will deal with the logic extension of risk-prediction, which is risk monitoring - after a problem, has occurred or been spotted.