Since approximately 2000, scientists have been finding ways to build "biocomputers" and use them to diagnose and treat disease, to produce biofuels and pharmaceuticals, and to address environmental damage (Lu and Purcell 2016).1 In this new field of synthetic biology, researchers are building biological tools that have computing capacities: basic logical (AND, OR, etc) operations, if/then tests, and digital memory. With the right biological inputs, these systems generate predictable outcomes. These techniques follow from, and surpass, those of genetic engineering.
Biocomputing systems are designed to work with the natural information-processing mechanisms of DNA transcription and translation in cells, essentially running their own programs on biological wetware. Initial advances included a 1-bit digital memory, a rudimentary oscillator, and a "Goldilocks" circuit where a cell would light up only if the concentration of a compound was in a certain range. Biocomputing circuits have been designed to perform basic arithmetic operations such as addition and subtraction, evaluate Boolean operators (eg, AND, OR, XOR), compute ratios and logarithms, convert 2-bit digital signals to analog protein levels, and record and transmit on/off states for logic gates from the parent cell to its children (ie, memory). Many of these are described in detail in the earlier (2004 ) review in Scientific American, the more current review cited above,1 and more recently in Purcell and Lu's review.
Sensors, logic operators, and memory components are being combined into genetic circuits for some pretty interesting applications. For example, a diagnostic application developed at MIT sends a bacterial spy (Bacteroides thetaiotaomicron, a common human gut bacteria) into the mouse gut, where its biocomputing circuits are triggered if it encounters certain disease biomarkers. This leads to the production of luciferase, which can be detected on excretion. This type of system could be used to diagnose inflammatory bowel disease (IBD) or gastrointestinal cancer. (To download the full article, click here.)
A therapeutic application by Xie and colleagues forces a cell to self-destruct if it contains a specific cancerous signature. The presence of 6 different microRNA markers uniquely identifies HeLa cervical cancer cells (3 markers are present at higher than usual concentration and 3 at lower than usual concentration). This system is designed to trigger production of a protein that directs the cell to commit suicide if and only if the 6 types of microRNA are present in the specified high/low pattern.
Compared with conventional computing, biological "computing" has some distinct differences that pose challenges:
• It's slow. No gigahertz speeds here. Biological processes can take hours.
• It's noisy (especially in analog systems). Communication between different parts of the system is "wireless" and not synchronized by any clock.
• It's somewhat unpredictable. We don't know everything about cell behavior. Models are only as good as our knowledge.
As researchers forge ahead exploring this brave-new-world terrain, scholars and advisors at the Synthetic Biology Project, an initiative of the Science and Technology Innovation Program of the Woodrow Wilson International Center for Scholars debate the ethics of this new direction. See their 2009 presentation for an overview.
1 Lu TK, Purcell O. Machine life. Scientific American, April 2016;59-63.
Blogger: Ginny Fleming, Founder, Lucidize Medical & Scientific Editing. Chief capacities: medical, scientific, and technical writing and editing.