Artificial neural networks made out of DNA are inspired by the human brain to program intelligent behaviours
Scientists at California Institute of Technology (Caltech) have developed an artificial neural network made out of DNA that can recognise highly complex and noisy molecular information like identifying handwritten numbers.
These artificial neural networks are mathematical models inspired by the human brain, function like networks of neurons and are capable of processing complex information. They are built with an aim to program intelligent behaviours like the ability to compute, make choices and much more.
Can recognise handwriting
To illustrate its capability, researchers chose a task for artificial neural networks to recognise the handwriting which proved to be a classic challenge for it. As human handwriting can vary widely, so when a person scrutinises a scribbled sequence of numbers, the brain performs complex computational tasks in order to identify them.
These networks must be “taught” how to recognise numbers, account for variations in handwriting, then compare an unknown number to their so-called memories and decide the number’s identity.
Molecular handwriting does not actually take the shape of a number. In fact, each molecular number is made up of 20 unique DNA strands chosen from 100 molecules, each assigned to represent an individual pixel in a 10 by 10 pattern. These DNA strands are mixed together in a test tube and carry out prescribed chemical reactions to identify molecular handwriting.
So these artificial neural networks can learn, form memories from examples added to the test tube. This way they can be trained to perform different tasks.
Common medical diagnostics detect the presence of a few biomolecules, for example, cholesterol or blood glucose. Using more sophisticated biomolecular circuits, diagnostic testing could one day include hundreds of biomolecules, with the analysis and response conducted directly in the molecular environment.