Xenobots: living nanorobots from frog cells

Back in 1495, Leonardo da Vinci created a drawing of living armor. And only after 425 years, Czech science fiction writer Karel Čapek first used the word “robot” in his play “R.U.R.”. Modern robots are much smarter, more complex and more mobile than the da Vinci robot, but they have common features. One of them is the material from which these amazing machines are made. When we talk about robots, we most often imagine something synthetic, not without reason in books and movies of robots are sometimes called synthetics.

However, robots can be created not only from metal, plastic or carbon fiber. Scientists from the University of Vermont (USA) decided to use dead frog cells as building materials. The resulting microscopic robots, called “xenobots,” are able to travel through the body of a living organism and perform their tasks. How exactly did scientists create artificial life, what talents can xenobots boast of, and where can such an unusual invention be applied? We learn about this from the report of the research group. Go.

Study basis

Creating a mechanism that will perform some tasks under the control of artificial intelligence is not so difficult nowadays. It is not difficult to rebuild the existing organism by changing its structure, functions or characteristics. However, creating life from scratch is not an easy task. Researchers at the University of Vermont say that synthetic materials are used in robotics for the most part because of the simplicity of their manufacture, implementation and integration. Exaggerated saying metal can always be melted, reforged, or sharpened. But living organisms, tissues and cells, i.e. living systems demonstrate the stability of structure and functions. They are very resistant to outside interference aimed at changing their behavior.

At the same time, living cells, especially embryonic ones, demonstrate amazing features that even the most developed synthetic robots cannot boast of. Embryonic cells are able to self-organize, realizing the processes of tissue development and regeneration, depending on the situation. Manipulations with this ability may allow the creation of a synthetic morphology through which new life forms can be realized, no matter how loud it sounds. Moreover, the process of cell self-organization can be supervised, thereby providing the future structure with the necessary functions and characteristics.

At the moment, there are already several methods for developing and creating individual living systems. For example, unicellular organisms can be modified by means of refactored (transformed) genomes, but this is not yet possible to implement in multicellular systems.

You can also modify the cell strand by changing the culture conditions. But in this case, control over the processes and over the structure and functions will be minimal. In contrast, there are developments in the field of bioengineering, where three-dimensional frameworks are studied. This option will give more control. But the inability to predict the behavior of an arbitrary biological structure limits this technique to the assembly of biological machines based on existing ones. In other words, it will be the same modification of what is already there, but not the creation of a new living organism.

Despite all the difficulties and obstacles, there are ways. One of them is computational search in conjunction with three-dimensional printing. Unlike machine learning, search is an evolutionary process that allows you to design the physical structure of a machine and its behavior from scratch. In addition, this method is not tied to any specific types of the structure being created or to any specific functions. The same evolutionary algorithm can be used to develop different systems: drugs, metamaterials, and even autonomous machines.

In our study today, scientists demonstrated a scalable approach to the design of living systems using an evolutionary algorithm.


Image No. 1

The new method is organized as a linear conveyor, which takes as input the description of the used biological building blocks and the desired behavior that the manufactured system should demonstrate. The conveyor continuously displays healthy living systems that implement the specified behavior in different ways. The resulting living systems are new collections of cells that have very little to do with existing organs or organisms.

Research results

The conveyor is organized as a sequence of generators and filters. The first generator is an evolutionary algorithm that discovers various ways of combining biological building blocks to realize the desired behavior. To begin with, a population of random variants of future system models is created. Each model is then recreated in a virtual environment, after which a performance rating is automatically assigned. Less productive models are deleted and overwritten by accidentally modified copies of more productive models. The repetition of this process leads to the formation of populations of diverse and non-repeating patterns.

Video presentation design process of reconfigurable organisms.

Since there are likely to be many differences between the simulated and the target physical media, effective models are passed through a stability (reliability) filter that allows only those that support the desired behavior despite noise (changes in the environment) to pass through.

Surviving noise-resistant models are then passed through an assembly filter, which removes models that are not suitable for the current assembly method or cannot scale to more complex tasks in the future. Manufacturability depends on the minimum size of concavity, which will be preserved in clusters of developing stem cells, which tend to close small gaps in their general geometry. The scalability of the model depends on its share of passive tissue, which provides space for future organ systems or payloads (i.e., space for holding a transportable substance, such as a medicine).


Xenopus laevis

Models that successfully pass through the assembly filter are then formed from living tissue. Pluripotent * stem cells are first harvested from embryos Xenopus laevis (smooth spur frog) in the blastula stage, then dissociate and combine to achieve the desired number of cells.

Pluripotent cells * able to differentiate (transform) into all types of cells, except cells of extra-germ organs.

After the incubation period, the aggregated tissue is manually formed by subtraction using a combination of forceps for microsurgery and a cauterization electrode with a 13 μm needle.

In addition, the contractile tissue is layered on the body by introducing frog heart progenitor cells, which naturally develop into cardiomyocytes (cardiac muscle). These cells will create contractile waves in certain places of the created organism.

The result of all these manipulations was a three-dimensional live representation of the model, which has the ability to independently move through the aquatic environment for several days and even weeks without additional nutrients.

Video presentation is the process of creating reconfigurable organisms.

The resulting organisms are subsequently introduced into the real physical environment to monitor their behavior.

Then, scientists compared the observations with the simulation results to identify the fact of the transition of behavior from “synthetics” to “organic”.

An important aspect of the technique is the continuation of the evolutionary algorithm even after the introduction of ready-made organisms into the environment. Some of them, despite their usefulness at the development stage, may not display the exact behavior that was expected. Such models of organisms are removed from the algorithm, which leads to the creation of next time more stable and environmentally appropriate organisms.

Thus, 4 runs of the conveyor were carried out. The result of this was 4 types of organisms, demonstrating the following features: locomotion, manipulation of objects, transfer of objects and collective behavior. Now let’s talk more about each of the features.


Image No. 2

To obtain a diverse set of models, 100 independent tests of the evolutionary algorithm (2A2C), each of which began with a different set of initial random models. During each test, models were selected based on the net displacement (displacement) achieved over a 10-second period (with a randomized, phase-modulated contraction, cyclic at 2 Hz). In the course of each test, additional selection parameters were applied in the form of stimulating competition within and between unique genetic lines, which led to the formation of unique environmental dynamics. At the end of each test, the most appropriate models were extracted (1A) and passed through stability and assembly filters. In the process of this filtering, those models were selected that retained fast locomotion during scaling and building-up (image No. 3).


Image No. 3

It should be noted that cilia (cilia, thin hairs on the cell surface) were not modeled during the design process and were suppressed during practical tests by embryonic microinjection of mRNA transcribing the intracellular Notch domain. Thus, all the movements of the organisms were carried out exclusively through the reduction of cardiac muscle tissue.

The trajectories of simulated and realized decylated (without cilia) organisms were compared in two orientations: vertical and inverted, i.e. inverted 180 ° relative to the transverse plane. Observations showed that at least one of the variants of the body’s models successfully realized the given behavior in a vertical orientation, but not in an inverted one (image No. 4).


Image No. 4

The direction of movement of organisms with a vertical orientation coincided with the direction of the model under random disturbances. This suggests that successful movement in space is not random, but is the result of the design of the model of the organism itself.

The second feature is the manipulation of objects. When there were solid particles in the environment of the test organisms, the former began to unite spontaneously with them, both in modeling and in practice (3F)

This behavior can hardly be called structured, since the necessary task parameters were not set. This can be more accurately realized by adding more precise data: for example, indicate the area that needs to be cleaned of particles or indicate a specific type of particles that need to be removed, while ignoring all the others. The second behavior was implemented, but so far at a primitive level.

At the next stage, organisms showing the ability to transport objects were tested. Some of the organisms have been designed to reduce hydrodynamic drag through a hole in the center of their transverse plane. However, there were no contractile tissues in this area of ​​the body.

This hole during the subsequent cycles of the evolutionary algorithm can be transformed into a kind of bag for transferring objects, which can be used for localized delivery of drugs.

Another feature is collective behavior. During practical implementation, a collision of two organisms was observed, leading to the formation of a temporary mechanical connection. These two organisms begin to rotate around each other, and after a few revolutions again separate along the tangential trajectories. Such behavior will be much more distinct if you do not suppress the development of cilia, as they lead to “intimidation” of the two organisms among themselves, i.e. their connection becomes much longer.

For a more detailed acquaintance with the nuances of the study, I recommend a look at report of scientists and Additional materials to him.

Epilogue

This study is not difficult to call unique, given its results. Scientists took frog cells and created from them new multicellular organisms that perform specified functions in varying degrees of success. Even the authors of this work understand that many are frightened by the prospect of autonomous, somewhat thinking, robots, let alone new forms of life. However, in their opinion, this study allows a better understanding of life itself as a phenomenon. In addition, the developed xenobots can serve in medicine, becoming living nano-surgeons, removing harmful and pathogenic cells from the patient’s body, or nano-couriers of drugs, delivering them directly to where they will most effectively fight the disease.

Life in all its forms and manifestations is impeccable, despite all the shortcomings. Every living creature has evolved to adapt to changing living conditions. Modern science is capable of creating life, but only within the framework of changing existing organisms. But to create something new from scratch is a completely different task, more difficult, more ambitious. There are examples of artificially created organisms, but they cannot be compared with those described by science fiction writers. Despite this, scientists from all over the world do not stop their research, hoping to create a new life form. It is still difficult to say how dangerous it is to play with nature. However, according to science fiction writers, such discoveries will not bring to good. Nevertheless, works of science fiction literature, although in many respects predict the future of society in general and science in particular, are only the figment of the author’s imagination. How exactly the technologies currently being developed, including xenobots, will be used, depends solely on ourselves.

Thank you for your attention, stay curious and have a great weekend everyone, guys! 🙂

A bit of advertising 🙂

Thank you for staying with us. Do you like our articles? Want to see more interesting materials? Support us by placing an order or recommending to your friends, cloud VPS for developers from $ 4.99, A unique analogue of entry-level servers that was invented by us for you: The whole truth about VPS (KVM) E5-2697 v3 (6 Cores) 10GB DDR4 480GB SSD 1Gbps from $ 19 or how to divide the server correctly? (options are available with RAID1 and RAID10, up to 24 cores and up to 40GB DDR4).

Dell R730xd 2 times cheaper at the Equinix Tier IV data center in Amsterdam? Only here 2 x Intel TetraDeca-Core Xeon 2x E5-2697v3 2.6GHz 14C 64GB DDR4 4x960GB SSD 1Gbps 100 TV from $ 199 in the Netherlands! Dell R420 – 2x E5-2430 2.2Ghz 6C 128GB DDR3 2x960GB SSD 1Gbps 100TB – from $ 99! Read about How to Build Infrastructure Bldg. class using Dell R730xd E5-2650 v4 servers costing 9,000 euros for a penny?

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *