Genomics

In almost every cell of a living organism lies a complete blueprint for that body. It consists of a long molecule called DNA, which can be imagined as a chain of four different building blocks. The order of these building blocks is the information. In humans, there are about three billion building blocks per cell. The entirety of this blueprint is called the genome. Genomics is the field that reads, compares, and evaluates such a genome as a whole.

From individual genes to the complete genetic material

In the past, research usually examined a single gene, that is, a short section of the blueprint. This helped with diseases that have exactly one clear cause. Most traits and diseases, however, arise from the interplay of many sections. That’s why it is far more useful to look at the complete genetic material all at once. This is precisely where genomics differs from classical genetics.

The practical benefit is now considerable. In cancer, doctors can search the tumor’s genetic material for the exact alteration and choose a suitable medication. In rare genetic diseases, a genome analysis can end a years-long search for the cause. In agriculture, it helps breed plant varieties with better traits.

An important and common misunderstanding needs addressing: the genome is not destiny. It describes possibilities and risks, not certain predictions. Diet, environment, and chance also play a role in what actually happens. Someone who carries a risk variant often never becomes ill.

Sequencing, assembling, comparing

The first step is called sequencing. In this process, machines break the DNA down into millions of short pieces and read out their sequence of building blocks. Each piece is only a tiny fragment of the whole. You can imagine it like a book that has been cut into snippets of ten words each.

The second step is pure computational work. Programs reassemble the snippets into the correct order based on their overlaps. The result is then compared with a reference genome, a kind of model version. Deviations from this model version are called variants. They are actually what’s interesting, since they account for the difference between two people.

This is where artificial intelligence comes into play. A genome yields hundreds of gigabytes of raw data and millions of variants. Trained models sort out which variants are harmless and which might disrupt a function. They also predict what shape a protein will take that a particular section describes. Without such software, these amounts of data would be practically impossible to evaluate.

Genomics in clinics, ancestry tests, and stock market news

Decoding the first human genome around the year 2003 still cost about three billion dollars and took over a decade. Today the price per genome is in the range of a few hundred dollars, and the analysis takes days. This drop in price is the reason genomics is suddenly turning up everywhere.

In everyday life, you encounter it in commercial ancestry tests, where you send in a saliva sample. Such tests usually read only selected sites in the genetic material, not the whole genome. In medicine, genome analyses for cancer and rare diseases are increasingly becoming routine. Genomics was also used during the COVID pandemic to identify new virus variants.

In business news, it is mainly the manufacturers of sequencing devices and biotech companies betting on genome-based therapies that appear. At the same time, there is ongoing debate about data privacy. A genome is the most personal information there is — it also reveals something about relatives. Anyone who collects such data is therefore under particular scrutiny from authorities and the public.