Biocomputers talk back

Research at Harvard University and Princeton University has brought practical biological computers a step closer by having them produce an identifiable output under certain conditions.

Biocomputers, composed entirely of DNA, RNA and proteins, could provide information that will eventually help direct therapies only to diseased cells or tissues.

Human cells can build their own biocomputers independently when provided with a genetic blueprint of the machine.

The molecular mechanisms can detect anything from the presence of a mutated gene to the activity of genes within the cell by evaluating Boolean logic equations inside cells. The biocomputers' ‘input’ is RNA, proteins, and chemicals found in the cytoplasm. ‘Output’ molecules indicating the presence of the telltale signals are easily discernable with basic laboratory equipment, a first for the technology.

To date, the biocomputers have been tested in human kidney cells in culture. Research into the system's ability to monitor and interact with intracellular cues such as mutations and abnormal gene levels is still in progress.

A biocomputer's calculations, while mathematically simple, could allow researchers to build biosensors or medicine delivery systems capable of singling out very specific types or groups of cells in the human body. Molecular automata could allow doctors to specifically target only cancerous or diseased cells via a sophisticated integration of intracellular disease signals, leaving healthy cells completely unaffected.