Reasons to Believe

Using Nature ’s Designs to Build a Better Mousetrap

Three examples of biological mimicry

I enjoy watching The New Yankee Workshop and Ask This Old House. Both of these programs showcase the skills of master craftsman Norm Abrams and others as they fabricate elegant home furniture or renovate homes. Almost without exception, I discover better designs and building techniques for the home projects I undertake. In similar and increasing fashion, scientists and engineers look to one source of superior designs––biological organisms––as they seek to build more efficient, elegant, robust, and functional devices. Three examples of such biological mimicry demonstrate the potential for exciting applications.

Mimicking Bacteria to Harness the Sun’s Energy

All the talk of global warming has generated interest in renewable energy resources. While the Sun provides a virtually inexhaustible energy supply, humans have struggled to harness that energy in a substantial and sustainable way. Solar cells continue to grow in efficiency, but their costs and technical limitations prevent widespread use. However, biological organisms employ a far superior design for harvesting the Sun’s energy.

Previous attempts to develop biologically inspired devices that efficiently and economically convert sunlight to energy have failed. Scientists’ futility resulted from an inadequate understanding of the mechanism behind the energy-harvesting mechanisms employed by the organisms. Recent advances are beginning to address that deficiency.

One group sought to understand how natural photosynthetic pigments in green bacteria self-assemble into an antenna system (called a chlorosome) that collects photons (basic units of light) and channels the energy to storage centers. Green bacteria live in lower layers of ponds, lakes, and oceans and are able to photosynthesize even in dim light. By combining the necessary chemical precursors from scratch (instead of trying to modify similar molecular systems), researchers were able to insert different functional groups into the hydrocarbon skeletons. (Visualize chains of carbon atoms with other groups of atoms branching off to form different molecules.)

Subsequent tests then measured how the different functional groups affected the amount of self-assembly that plays a critical role in harvesting the absorbed energy from dim light. Although the different functional groups affected the amount of assembly, all the trials indicated the “antenna system” for gathering sunlight self-assembled when dissolved.1 The research also demonstrated self-assembly of the light-harvesting structures on surfaces—a necessary requirement for widespread use. This result lays the foundation for the next generation of more-efficient solar-powered devices.

Mimicking the Brain’s Response to New Information

Computers have revolutionized modern society. Along with more mundane tasks (such as publishing articles like this one), they affect how we communicate (think Facebook, Twitter, and even cell phones), bring greater workplace efficiency, afford safer travel, and enable space exploration. While computers perform the tasks they were designed for well, reprogramming them for other purposes requires resources and is difficult to reverse. In contrast, humans show a remarkable ability to learn new skills and adapt to new situations (while retaining the older skills). Thus, MIT scientists have looked to the human brain’s architecture to design more flexible computer circuitry.

The behavior of neurons in the brain changes, depending on the information they receive. Known as plasticity, this adaptability provides a mechanism to learn new tasks and to function in different environments. Pairs of neurons and the gap between them (called a synapse) exchange various chemicals, which propagate information to other structures that generate appropriate responses. Ultimately, these processes produce electrical signals that operate in an analog fashion that allows the brain to exhibit plasticity. But most computers utilize digital signaling—either ON or OFF.

Using a suite of 400 transistors, the MIT-based team produced a computer chip that could mimic all the electrical signals (not just the ON-OFF switching) occurring between one pair of neurons in the brain.2 Although the brain contains around 100 billion neurons, this work represents a significant milestone in producing adaptable computer technology. No longer the domain of science fiction, some of the exciting foreseen applications include prosthetics that send and receive nervous system signals, interfacing machines to the brain, computers that operate like the brain, and machines that can learn.

Mimicking Decision Making of Bacterial Swarms

Robotic and unmanned devices often operate with great difficulty in unknown or hostile environments. One significant barrier pertains to making good decisions, a process that requires both anticipating the type of necessary information and having the equipment capable of making the appropriate measurements. In the natural realm, bacterial communities operate with a different approach that employs a much more limited set of information. Israeli scientists designed computer simulations to study the bacterial method and learned that it provides a superior capacity to survive.

Because of the limited “computational ability” of a bacterial community, they must adopt a different approach than gathering large amounts of information. Instead, each bacterium senses the local environment and communicates—through molecular, chemical, and mechanical means—with the rest of the swarm whether to proceed or alter course. Often, such a path produces disastrous results because a small group of individuals can lead the entire group in the wrong direction. For a bacterial community, this tendency might end with the swarm encountering a dangerous toxin or moving away from needed nutrients.

Computer simulations demonstrate how swarms avoid this pitfall.3 Specifically, each bacterium changes the amount of interaction with the rest of the swarm, depending on how beneficial its route is. For a beneficial path, a bacterium will still send its information, but it pays less attention to the information coming from the community. On a more strenuous or challenging path, the bacterium increases its interaction with the swarm to learn more about its terrain compared to other locations. This “self-confidence” exhibited by each bacterium drastically reduces the chance of a small group erroneously putting the community in danger.

These results show how to build robotic swarms to navigate and investigate unknown terrains reliably. Instead of trying to design highly sophisticated and intelligent (and consequently, expensive) units that communicate a wealth of information, a better plan utilizes a large number of simpler and cheaper units that operate similar to the bacterial communities.

Such an approach would greatly enhance the ability to explore planets, asteroids, comets, and moons in our solar system. The unpredictable nature of these environments makes a bacterial model far more likely to succeed and provide the information necessary to decide on the utility of a manned mission.

When humans need a better design, we seek the input of someone with more experience, knowledge, and resources. The fact that the best engineers look to biological organisms for superior designs comports well with the idea that a Grand Designer is behind all life here on Earth.

Subjects: Biochemical Design

Dr. Jeff Zweerink

While many Christians and non-Christians see faith and science as in perpetual conflict, I find they integrate well. They operate by the same principles and are committed to discovering foundational truths. Read more about Dr. Jeff Zweerink.

1. Olga Mass et al., “De Novo Synthesis and Properties of Analogues of the Self-assembling Chlorosomal Bacteriochlorophylls,” New Journal of Chemistry 35 (2011): 2671–90.

2. Adi Shklarsh et al., “Smart Swarms of Bacteria-inspired Agents with Performance Adaptable Interactions,” PLoS Computational Biology 7 (September 2011): e1002177.

3. Guy Rachmuth et al., “A Biophysically-based Neuromorphic Model of Spike Rate- and Timing-dependent Plasticity,” Proceedings of the National Academy of Sciences, USA 108 (2011): E1266–74.