The year 2014 was a good one for
Austen Heinz. Cambrian Genomics, his modest start-up of 11 employees, raised $10
million in investment money. For a small company, Cambrian has nailed down
contracts with some big-time partners, including pharmaceutical companies GlaxoSmithKline
(2013 revenue = $38 billion) and Roche (2014 revenue = $50 billion).
If these accomplishments don’t
convey the noise Heinz is beginning to generate, he also ended up in an
Instagram post with eccentric actor, musician and entrepreneur Jared Leto at the
Cambrian laboratory (total likes = 107,000).
So what exactly is Cambrian
Genomics doing to garner all of this attention? I’ll let Heinz explain in his
own words, as told to TechRepublic.com: “We print life. Life is very simple,
it’s just code. Four letters: we print that.”
Heinz is referring to the genetic ‘code’ stored in DNA, a series of linked molecules that, depending on the
pattern, can carry instructions for everything from a human to a jellyfish to
The code of DNA—composed of adenine
(A), threonine (T), cytosine (C) and guanine (G)—is much like binary computer
code. In computers, 1’s and 0’s correspond to physical changes in the hardware
of the computer. We use this interface between the abstract and physical world to
store our information and perform computations on it. In the eight decades since
the invention of the electronic computer, humans have gotten so good at writing
code that we can create a practically infinite array of virtual worlds: video
games, websites, operating systems for all of our diverse devices—even
this list barely scratches the surface.
The advances we’ve made in
computing are astronomical and immeasurably rapid, but DNA is quite different
from computers and computer code, mainly because we didn’t invent it. Progress
has moved slowly. Scientists weren’t even able to read the DNA code until 1977.
Consider the Human Genome Project,
an international endeavor to uncover the directions that guide human life from
birth to death. It took ten years and $3 billion to merely map out the chain of
letters in human DNA. That’s a dollar per letter sequenced. It now costs $5,000
to sequence an entire human genome—roughly a millionth of a cent per letter.
Cambrian’s technology, however, has
less to do with reading DNA and everything to do with writing it. The ability
to edit DNA and see the result is a tremendous advantage for scientists, who
have had to use workaround ways of changing organismal DNA to deduce the effect
of certain DNA sequences. And Heinz, a dropout from the Stanford electrical
engineering Ph.D. program, has found a way to make writing DNA much cheaper and
more accurate. For reference, the cost of writing a single letter of DNA has
fallen from $2 in 2003 ($6 billion for human sequence) to 12 cents in 2013 ($3
million for human sequence).
Of course, that’s still prohibitively
expensive, but the price is bound to keep dropping. The buzz about Cambrian is really
more attributable to what the enigmatic man at the company’s helm wants to do with
DNA synthesis once it’s cheap enough: democratize it.
Heinz wants anybody to be able to
send in DNA code and receive back the product. DNA is different from computer
code in that it carries both the instructions for manipulating the hardware of
life and the instructions for building the hardware itself. In Heinz’s vision, anyone with a computer could write DNA code and send it
into Cambrian to be grown in cells. This is almost entirely unprecedented
territory—life designed not by evolution but by human imagination.
As with most emerging technology,
the possibilities, though still far off, are endless, especially as 3D printers
move into the home. They are at once terrifying and invigorating. Want to wipe
out humanity? Google the code to the Ebola virus, make some modifications to
turn it airborne, and you’ve got it: mass murder at your fingertips. Want to
colonize Mars? It might help to create a plant that can survive in the planet’s
weak atmosphere and provide oxygen to colonizers.
Depending on who you are, the
promise of this vision may sound either remarkable or disturbing, but one
hurdle, aside from the price of synthesis, still stands in the way. To
understand it we must return to the computer code analogy. When programmers
tell computers what to do, they don’t write in actual computer code (1’s and
0’s). Instead, they use what are called abstract high-level languages to
execute their desired tasks. Some recognizable languages include Java and Python.
In high-level languages such as these, we can write “print ‘Hello world!’” and
the computer can interpret and execute our command, modifying the computer’s
1’s and 0’s to follow our orders.
But there is no abstract DNA
language. Scientists are still trying to understand how much of the hardware
built by DNA even works. Making controlled and directed changes on our whim
still represents a logistical and technical nightmare, and democratized creature
printing remains an infinite fantasy.
Still, if the history of computing
is any indication, humans will overcome these barriers. It’s tremendously
difficult to bet on when DNA printing will be in the hands of the many instead
of the few, but it’s all but trivial to bet that it will occur. If people like
Austen Heinz have their way, the ‘when’ may come faster than anybody is
Warren Szewczyk PO ’15 is hoping to spend the next couple years researching schizophrenia and continuing to write about science. If you’re reading this, you hopefully read the entire preceding column, so thanks for that.