Category Archives: data-science

In the year 2000, there was one single human genome. In 2010 there were fewer than 100 human genomes (you could look them up in a spreadsheet!). Today there are likely 1,000,000 human genomes. Good luck cataloging them all. Outside of the purview of our species, there are now efforts to sequence every animal on earth. And the sequencing revolution has not just changed our understanding of DNA, it has opened up the world of RNA to us, allowing scientists to track and trace gene expression in minute detail. Genomics is “eating biology.”

Whereas there was once a tiny data pond, today a substantial lake is swelling into a massive ocean. This is why we built GenRAIT — to help transition the burgeoning ecosystem of 21st-century genetics into the new age of genomics. Data offers the potential for insight and discovery. Data on life’s code — the genome — can potentially transform the future of human health outcomes. This makes “data” more than just a buzzword, but the key to unlocking the potential for a better world. But the influx and quantity of genome data in our new era threaten to overwhelm the capability of scientists to manage, utilize, and harness it, making that reality we wish to come into being unreachable. We want to push beyond that impasse with GenRAIT, and unlock the potential future.

But what happens when the data is finally brought under control? Data without an end is without purpose. What might the genetic future look like? Why do we at GenRAIT care so much?

One generation ago sequencing one’s genome was “blue sky” science, whereas today it’s a consumer commodity. Companies like Nebula genomics provide 30x high-quality medical-grade sequencing to consumers for $300 or less. With the average cost of health insurance for a family more than $1,000 a month, the cost of sequencing one’s genome is trivial. And whereas buying a car or other consumer item means acquiring a depreciating good, as its value declines over time, your genome sequence becomes more valuable as more research is published on the relationship between genetics and disease.

The more data you have in the pool the more results and findings you can obtain. Thirty years ago detecting a genetic variant that might cause a disease required tracking an inbred pedigree for decades. It was a project only viable for a hospital research group. But science moves forward. Fifteen years ago geneticists began to perform “genome-wide associations” that looked for common variation — those genes commonly causing disease within the population. This is the sort of result a company like 23andMe provides.

But there is more to the genome than things that are known and common. Many illnesses are caused by variations within families and narrow local lineages. If common variants are known unknowns, these are unknown unknowns, and only whole-genome sequences can give us insight. We have the technology, but we lack execution. Every individual’s joint medical and genomic information could be powerful, but only in the context of population-wide analysis of subtle but cumulatively significant patterns. You can only perceive the trees if you can see the forest. The value of one sequence goes up by orders of magnitude when you analyze it in the context of one billion sequences.

As we go into the 21st century, genomics will help us do more than diagnose and evaluate retrospectively. It will be essential to cure, treat, and anticipate the future. An individual’s genome can give doctors a map of how to cater to an individualized healthcare plan. That same genome can be used to prescribe lifestyle changes to improve that person’s future well-being and increase their longevity, impacting morbidity and mortality. It can be used to conserve and save endangered species and help them evolve to better adapt to the present and future environments. We now can imagine a future that can be edited and revised because of new technology.

In 2012 CRISPR genetic-engineering technology took the biological world by storm (and yielded Jennifer Doudna and colleagues a Nobel Prize), making gene-editing available to the broad masses of researchers. Though recombinant DNA technology has been utilized by scientists since the 1970s, it was a form of genetic engineering that was expensive and difficult to execute. CRISPR democratized genetic engineering, opening up the possibility that gene-editing could be a bespoke process, offering up the possibility of curing millions of people with congenital illnesses. Diseases like cystic fibrosis will likely be cured in the next twenty years through gene-editing technology.

Nevertheless, to get to that stage, we need the right environment in place to allow scientists to extract valuable information, patterns, and insights out of the genomes they receive. Before one can write to the genome, one must read the genome. Before one can develop engineering applications, one must master physics. We are already in the genomic age, as sequencing costs keep crashing and new technologies are on the horizon. But the flood of data threatens to overwhelm our capacity to use it rationally, intelligently, and effectively. As the NIH states, “Our ability to sequence DNA has far outpaced our ability to decipher the information it contains.” We must do better because the well-being of hundreds of millions is on the line. We have the data necessary to usher in a better future for healthcare and precision medicine. Now we need to unlock it.

Read more