Not only have we learned to read and write the genetic code; now we can put it in digital form and translate it back into synthesized life. In theory, that gives our species control over biological design. We can write DNA software, boot it up to a computer converter and create unlimited variations of the gene sequences of biological life.
The most anxiously and immediately awaited outcome of this new capacity is its application for correcting genetic deficiencies that lead to a range of maladies, from cancer to diabetes to Alzheimer’s. Here, some humility must temper hope. What we know is surely considerable, but it is dwarfed by what we still have to learn.
What we know
The human genome, which represents each individual’s entire set of genetic information, was first decoded in 2000 by two competing teams. When the results were published in 2001, the biggest single finding was that we only had 20,000 or so genes, not the hundreds of thousands predicted by many scientists. That surprising discovery led us to rethink our assumptions about human biology and disease. Early genetic findings, such as the discovery of the Huntington’s disease gene and the cystic fibrosis gene, had led many to believe that most human diseases were caused by a single defective gene. Not so.
The cost of sequencing the first genomes — determining the exact order of the base pairs of a segment of DNA — were astronomical. The private effort that I led spent over $100 million on the first genome, while the U.S. government costs were over $2 billion. That meant that even understanding the genome by sequencing more than the reference genome was unlikely — until the technology changed.
The discovery of a smaller-than-expected number of genes also caused some to question whether DNA was the full genetic material or if there was something more. A small team from my nonprofit research institute set out to answer that question in a definitive manner by synthesizing a chromosome from the four chemicals that make up the DNA chemical code. The project was much harder and more complicated than expected. But in 2010, we finally succeeded in booting up our synthesized chromosome, proving that a synthetic cell could be produced.
Six years later, we produced the first cell where the genome was designed in the computer. It turned out that every cell function and every molecule in the cell derived directly or indirectly from the synthetic DNA chromosome — proving that DNA is in fact the complete genetic material.
Interpreting the code
The human genome thus understood has spawned hundreds of businesses, all based on trying to interpret the code to tell people their geographic ancestry, disease risks and even what foods to eat based on their genome. Most of these commercial genome shops don’t sequence the genome, they simply use so-called “gene chips” that give a readout on a very small portion of the genome.
This is the approach taken by popular products like 23andMe, Ancestry and My Heritage. Many labs only sequence the exome, which represents about 2 percent of the genome but contains most of the protein coding areas. Very few actually sequence the genome to cover all 6.4 billion bases.
We are still at the earliest stages of understanding the human genome, and most people get little value from genome snippet analysis. Those offering diet information are outside the realm of documented science. Cancer risk is currently one of the most useful areas of genome analysis, but we are at the very beginnings of being able to use only genome data for life predictions.
At Human Longevity, we focus on generating clinical and phenotype data to aid us in the interpretation of the genome and to improve our ability to make predictions from it. In the process of collecting data from presumed healthy individuals, we have found that how you feel is not a good indication of your actual health. For example, 5 percent of all people that we test over the age of 50 have a major cancer that they are unaware of. The good news is we have had 100 percent success in treating these cancers due to their early detection. One percent of all clients have a brain aneurysm; 27 percent have fatty livers; and 12 percent are at high risk for a cardiac event.
We are using machine learning tools to correlate this data with the complete genome sequence to discover the precise genetic cause of diseases. I predict that within a decade, the human genome sequence will provide sufficient predictive knowledge to make it a worthwhile standalone test. Unfortunately, we are not there yet.
Correcting genetic defects — and the ethical limits
Many believe that as we learn the cause of genetic defects, CRISPR or other editing tools could correct them. However, gene therapy has mostly failed so far — it has proven impossible to get the corrected gene into the right 100 trillion cells in the human body. Ex vivo gene therapy — where cells are treated outside of the body and then returned — has had some success. There have been some encouraging results with cancer treatments using immune cells that have been edited with CRISPR. I think we will see even more success in cancer treatments using ex vivo gene editing.
There are some who want to take genome editing much further by editing the germline genome so that it will forever change a given trait or disease. This is a complex area because there are clearly devastating diseases that we would like to eliminate from humans. But to do so, we need to do human experimentation without knowing the consequences. Also, the current editing tools are not as precise as indicated in the press and have many unintentional effects, where other genes are changed along with the intended one.
The world agreed at the end of World War II to stop all direct human experimentation. Human germline editing would cross that boundary and take us back into random human genome editing, just to see what happens. We should not let this happen.
The prospects of eliminating disease and improving longevity are within our grasp. The way to reach that aspiration is to continue enhancing our knowledge of the genome itself so that genome editing can become a legitimate part of the future of medicine.