System Bits: July 16

Test tube AI neural network
In a significant step towards demonstrating the capacity to program artificial intelligence into synthetic biomolecular circuits, Caltech researchers have developed an artificial neural network made out of DNA that can solve a classic machine learning problem: correctly identifying handwritten numbers.

The work was done in the laboratory of Lulu Qian, assistant professor of bioengineering.

Lulu Qian, Assistant Professor of Bioengineering at Caltech.
Source: Caltech

“Though scientists have only just begun to explore creating artificial intelligence in molecular machines, its potential is already undeniable,” Qian said. “Similar to how electronic computers and smart phones have made humans more capable than a hundred years ago, artificial molecular machines could make all things made of molecules, perhaps including even paint and bandages, more capable and more responsive to the environment in the hundred years to come.”

Artificial neural networks are mathematical models inspired by the human brain. Despite being much simplified compared to their biological counterparts, artificial neural networks function like networks of neurons and are capable of processing complex information. The Qian laboratory’s ultimate goal for this work is to program intelligent behaviors (the ability to compute, make choices, and more) with artificial neural networks made out of DNA.

She explained, “Humans each have over 80 billion neurons in the brain, with which they make highly sophisticated decisions. Smaller animals such as roundworms can make simpler decisions using just a few hundred neurons. In this work, we have designed and created biochemical circuits that function like a small network of neurons to classify molecular information substantially more complex than previously possible.”

To illustrate the capability of DNA-based neural networks, the team chose a task that is a classic challenge for electronic artificial neural networks: recognizing handwriting. This is because human handwriting can vary widely, so when a person scrutinizes a scribbled sequence of numbers, the brain performs complex computational tasks in order to identify them. Because it can be difficult even for humans to recognize others’ sloppy handwriting, identifying handwritten numbers is a common test for programming intelligence into artificial neural networks. These networks must be “taught” how to recognize numbers, account for variations in handwriting, then compare an unknown number to their so-called memories and decide the number’s identity.

Conceptual illustration of a droplet containing an artificial neural network made of DNA that has been designed to recognize complex and noisy molecular information, represented as ‘molecular handwriting.’
Source: Caltech

Specifically, the neural network made out of carefully designed DNA sequences carries out prescribed chemical reactions to accurately identify “molecular handwriting.” Unlike visual handwriting that varies in geometrical shape, each example of molecular handwriting does not actually take the shape of a number, they noted. Instead, each molecular number is made up of 20 unique DNA strands chosen from 100 molecules, each assigned to represent an individual pixel in any 10 by 10 pattern. These DNA strands are mixed together in a test tube.

The researchers plan to develop artificial neural networks that can learn, forming “memories” from examples added to the test tube. This way, Qian said, the same smart soup can be trained to perform different tasks.

Keeping time with constant, measurable rotation of molecules
In a development that could one day significantly improve the accuracy and performance of navigation on smartphones and other consumer devices, MIT researchers have developed the first molecular clock on a chip, which uses the constant, measurable rotation of molecules — when exposed to a certain frequency of electromagnetic radiation — to keep time.

The clock transmitter chip (pink) wired to a circuit board package. Connected is a metal gas cell (right), in which a 231.061 GHz signal generated from the chip excites the rotation of carbonyl sulfide molecules. Because the peak rotation of the molecules is constant, it can be used as a reference point to keep accurate time.
Source: MIT

The researchers reminded that today’s most accurate time-keepers are atomic clocks, which rely on the steady resonance of atoms, when exposed to a specific frequency, to measure exactly one second. Several such clocks are installed in all GPS satellites. By “trilaterating” time signals broadcast from these satellites — a technique like triangulation, that uses 3D dimensional data for positioning — smartphones and other ground receivers can pinpoint their own location.

The problem is that atomic clocks are large and expensive. A smartphone, therefore, has a much less accurate internal clock that relies on three satellite signals to navigate but can still calculate wrong locations. Errors can be reduced with corrections from additional satellite signals, if available, but this degrades the performance and speed of the navigation. When signals drop or weaken — such as in areas surrounded by signal-reflecting buildings or in tunnels — a phone primarily relies on its clock and an accelerometer to estimate a location and where the user going.

However, the MIT-developed on-chip clock exposes specific molecules — not atoms — to an exact, ultrahigh frequency that causes them to spin. When the molecular rotations cause maximum energy absorption, a periodic output is clocked — in this case, a second. As with the resonance of atoms, this spin is reliably constant enough that it can serve as a precise timing reference, they explained.
 
In experiments, the molecular clock averaged an error under 1 microsecond per hour, comparable to miniature atomic clocks and 10,000 times more stable than the crystal-oscillator clocks in smartphones, according to the team.

And because the clock is fully electronic and doesn’t require bulky, power-hungry components used to insulate and excite the atoms, it is manufactured with the low-cost, complementary metal-oxide-semiconductor (CMOS) integrated circuit technology used to make all smartphone chips.

The chip-scale molecular clock can also be used for more efficient time-keeping in operations that require location precision but involve little to no GPS signal, such as underwater sensing or battlefield applications, the team added.

Electrical contact to molecules in semiconductor structures
In a technique that promises to bring advances in sensor technology and medicine, a team of researchers from the University of Basel and IBM Research – Zurich in Rüschlikon have developed a new method that allows electrical contact to be established with simple molecules on a conventional silicon chip.

Given that electrical circuits are constantly being scaled down and extended with specific functions, the field of molecular electronics is seeking to manufacture circuit components from individual molecules instead of silicon, and because of their unique electronic properties, molecules are suited to applications that cannot be implemented using conventional silicon technology.

Tiny pores were filled with molecules and brought into electrical contact via a platinum electrode from below and a gold nanoparticle electrode from above. Section of a silicon wafer containing thousands of pores with contact pads for electrical characterization.
Source: IBM Research – Zurich

At the same time, this requires reliable and inexpensive methods for creating electrical contacts at the two ends of a molecule.

This technique allows electrical contact to individual molecules to be established, the team reported. Thousands of stable metal-molecule-metal components can be produced simultaneously by depositing a film of nanoparticles onto the molecules, without compromising the properties of the molecules, and has been demonstrated using alkane-dithiol compounds, which are made up of carbon, hydrogen, and sulfur.

The team said the technique largely resolves the issues that previously hampered the creation of electrical contacts to molecules – such as high contact resistance or short circuits by filaments penetrating the film.

Building blocks fabricated by this method can be operated under standard conditions, provide long-term stability, and can be applied to a variety of other molecular systems. The team believes this opens up new avenues for integrating molecular compounds into solid-state devices including new types of instruments in the fields of sensor technology and medicine.