Skip to main content

Parallel network-based biocomputation: technological baseline, scale-up and innovation ecosystem

Periodic Reporting for period 3 - Bio4Comp (Parallel network-based biocomputation: technological baseline, scale-up and innovation ecosystem)

Reporting period: 2019-07-01 to 2020-12-31

Many technologically and societally important mathematical problems – such as the design and verification of circuits, the folding and design of proteins and optimal network routing – are intractable for conventional, serial computers. Therefore, a significant need exists for parallel-computing approaches that are capable of solving such problems within reasonable time frames. Recently, part of our consortium demonstrated a proof-of-principle for a parallel-computation system – termed network-based biocomputation (NBC) – in which a given combinatorial problem is encoded into a graphical, modular network that is embedded in a nanofabricated planar device. The problem is then solved by a large number of independent biological agents, namely molecular-motor-propelled protein filaments, exploring the network in a highly parallel fashion (PNAS 113, 2591 (2016)). Notably, this approach uses orders of magnitude less energy than conventional computers, thus addressing issues related to power-consumption and heat-dissipation.

Within Bio4Comp we
(i) will establish the technological and scientific basis for robust scale-up of this approach,
(ii) will demonstrate scalability by systematically increasing the problem size by several orders of magnitude, and
(iii) will develop new algorithms with the aim to open up a wide range of applications. Additionally, we will
(iv) foster and structure an ecosystem of scientists and companies that will accelerate the path to market success, including the creation of a joint technological roadmap.

Benefits to society will include the ability to solve hitherto intractable practical problems, the development of a sustainable and energy-efficient computing approach that is radically different from current information and communications technology and shedding light on key unsolved problems in computer science.
We achieved the first major goal we set for Bio4Comp overall: to beat the most difficult combinatorial problem quantum computing has solved to date. Specifically, we solved several exact cover instances with up to 1024 potential solutions, which constitutes a 128-fold increase in problem size compared to the state-of-the-art. This was enabled by improvements in the network structures both for the actin-myosin and the kinesin-1-microtubule system that greatly enhanced the reliability of our computing devices.

We also made good progress towards Bio4Comp’s second major goal: solving satisfiability using network-based biocomputation:
(i) We have solved two proof-of-principle SAT instances using “space-encoded” SAT devices, that are large but do not require tagging.
(ii) We have demonstrated the reliable encoding and decoding of 4 bits of information for microtubules – an important step towards “agent-encoded” SAT devices that are very compact but require tagging.

We have made good progress towards further scale-up of network-based biocomputation devices. We have demonstrated
(i) error-free 3D pass-junctions with tunnels and bridges
(ii) filament multiplication inside network channels
(ii) reliable encoding and readout of 4 bits of data for moving microtubules enabling new encodings that result in more compact networks

In order to make network-based biocomputation more practically applicable, we are developing several methods to make the devices and the motor proteins programmable, as well as integrating detectors into the devices themselves.

The website https://bio4comp.org/ helps to communicate events such as workshops, awards and webinars. We have created a LinkedIn page in order to make use of an existing professional networking platform.
We solved several exact cover instances with up to 1024 potential solutions, which constitutes a 128-fold increase in problem size compared to the state-of-the-art. This was achieved within a timeframe of five years since the publication of the results that constitute the state-of-the-art (Nicolau et al. PNAS 2016). In comparison, according to Moore’s Law, electronic computers roughly double their performance every 18 months, which corresponds to a 11 times performance increase in five years.

We made good progress towards Bio4Comp’s second major goal: We have solved two instances of the satisfiability problem using network-based biocomputation.

We have developed several key technologies enabling further scale-up of our networks: error-free junctions, filament multiplication in networks, encoding of data into filaments, formal verification of the correctness of networks.

The Bio4Comp project will demonstrate the scale-up of an alternative parallel computing approach that uses orders of magnitude less energy than conventional computers and thus provides an important contribution towards green computing technology. This approach will be an excellent complement to existing electronic computers, particularly because heat generation, small-scale e.g. quantum effects, and rising costs will prevent further miniaturization of electronic computers by about 2024. Therefore, Bio4Comp will lay the foundation for a new technology that has the potential to disrupt the market of electronic computers and help us solve important practical problems such as the design and verification of circuits, the folding and design of proteins and optimal network routing.
Principle of network-based biocomputation. (Graphic/pictures: Till Korten, Cornelia Kowol)