Open and remotely accessible Neuroplatform for research in wetware computing

Wetware computing and organoid intelligence is an emerging research field at the intersection of electrophysiology and artificial intelligence. The core concept involves using living neurons to perform computations, similar to how Artificial Neural Networks (ANNs) are used today. However, unlike ANNs, where updating digital tensors (weights) can instantly modify network responses, entirely new methods must be developed for neural networks using biological neurons. Discovering these methods is challenging and requires a system capable of conducting numerous experiments, ideally accessible to researchers worldwide. For this reason, we developed a hardware and software system that allows for electrophysiological experiments on an unmatched scale. The Neuroplatform enables researchers to run experiments on neural organoids with a lifetime of even more than 100 days. To do so, we streamlined the experimental process to quickly produce new organoids, monitor action potentials 24/7, and provide electrical stimulations. We also designed a microfluidic system that allows for fully automated medium flow and change, thus reducing the disruptions by physical interventions in the incubator and ensuring stable environmental conditions. Over the past three years, the Neuroplatform was utilized with over 1,000 brain organoids, enabling the collection of more than 18 terabytes of data. A dedicated Application Programming Interface (API) has been developed to conduct remote research directly via our Python library or using interactive compute such as Jupyter Notebooks. In addition to electrophysiological operations, our API also controls pumps, digital cameras and UV lights for molecule uncaging. This allows for the execution of complex 24/7 experiments, including closed-loop strategies and processing using the latest deep learning or reinforcement learning libraries. Furthermore, the infrastructure supports entirely remote use. Currently in 2024, the system is freely available for research purposes, and numerous research groups have begun using it for their experiments. This article outlines the system's architecture and provides specific examples of experiments and results.

Evolutionary approach to construct robust codes for DNA-based data storage

DNA is a practical storage medium with high density, durability, and capacity to accommodate exponentially growing data volumes. A DNA sequence structure is a biocomputing problem that requires satisfying bioconstraints to design robust sequences. Existing evolutionary approaches to DNA sequences result in errors during the encoding process that reduces the lower bounds of DNA coding sets used for molecular hybridization. Additionally, the disordered DNA strand forms a secondary structure, which is susceptible to errors during decoding. This paper proposes a computational evolutionary approach based on a synergistic moth-flame optimizer by Levy flight and opposition-based learning mutation strategies to optimize these problems by constructing reverse-complement constraints. The MFOS aims to attain optimal global solutions with robust convergence and balanced search capabilities to improve DNA code lower bounds and coding rates for DNA storage. The ability of the MFOS to construct DNA coding sets is demonstrated through various experiments that use 19 state-of-the-art functions. Compared with the existing studies, the proposed approach with three different bioconstraints substantially improves the lower bounds of the DNA codes by 12-28% and significantly reduces errors.

Cell-Membrane-Anchored DNA Logic-Gated Nanoassemblies for In Situ Extracellular Bioimaging

Extracellular K⁺ and adenosine triphosphate (ATP) levels are significantly elevated in the tumor microenvironment (TME) and can be used as biomarkers for early cancer detection and tumor localization. Most reported TME sensors only respond to single abnormal factors, resulting in a lack of accuracy and specificity for the detection of complex environments. Thus, precisely locating the TME remains challenging. In this work, we aimed to develop an intelligent DNA nanoassembly controlled by a "YES-AND" logic circuit using a bimolecular G-quadruplex (G4) and ATP aptamer as logical control units. As a proof of concept, in the presence of K⁺ (input 1) and ATP (input 2), the YES-AND Boolean operator returned a true value and the output was the fluorescence resonance energy transfer (FRET) signal, indicating high sensitivity and selectivity. After being anchored to living cell surfaces, this logic nanosensor imaged extracellular K⁺ and ATP present at abnormal levels in situ. Owing to diverse disease markers in the TME, this novel logic sensor might hold great promise for the targeted delivery of intelligent anticancer drugs and Boolean logic-controlled treatment.

Large and stable: actin aster networks formed via entropic forces

Biopolymer networks play a major role as part of the cytoskeleton. They provide stable structures and act as a medium for signal transport. These features encourage the application of such networks as organic computation devices. While research on this topic is not advanced yet, previous results are very promising. The protein actin in particular appears advantageous. It can be arranged to various stable structures and transmit several signals. In this study aster shaped networks were self-assembled via entropic forces by the crowding agent methyl cellulose. These networks are characterised by a regular and uniquely thick bundle structure, but have so far only been accounted in droplets of 100 μm diameter. We report now regular asters in an area of a few mm² that could be observed even after months. Such stability outside of an organism is striking and underlines the great potential actin aster networks display.

PPDTS: Predicting potential drug-target interactions based on network similarity

Identification of drug–target interactions (DTIs) has great practical importance in the drug discovery process for known diseases. However, only a small proportion of DTIs in these databases has been verified experimentally, and the computational methods for predicting the interactions remain challenging. As a result, some effective computational models have become increasingly popular for predicting DTIs. In this work, the authors predict potential DTIs from the local structure of drug–target associations' network, which is different from the traditional global network similarity methods based on structure and ligand. A novel method called PPDTS is proposed to predict DTIs. First, according to the DTIs’ network local structure, the known DTIs are converted into a binary network. Second, the Resource Allocation algorithm is used to obtain a drug–drug similarity network and a target–target similarity network. Third, a Collaborative Filtering algorithm is used with the known drug–target topology information to obtain similarity scores. Fourth, the linear combination of drug–target similarity model and the target–drug similarity model are innovatively proposed to obtain the final prediction results. Finally, the experimental performance of PPDTS has proved to be higher than that of the previously mentioned four popular network‐based similarity methods, which is validated in different experimental datasets. Some of the predicted results can be supported in UniProt and DrugBank databases.

Encoding scheme for data storage and retrieval on DNA computers

There has been exponential growth in the amount of data being generated on a daily basis. Such a huge amount of data creates a need for efficient data storage techniques. Due to the limitations of existing storage media, new storage solutions have always been of interest. There have been recent developments in order to efficiently use synthetic deoxyribonucleic acid (DNA) for information storage. DNA storage has attracted researchers because of its extremely high data storage density, about 1 exabyte/mm3 and long life under easily achievable conditions. This work presents an encoding scheme for DNA-based data storage system with controllable redundancy and reliability, the authors have also talked about the feasibility of the proposed method. The authors have also analysed the proposed algorithm for time and space complexity. The proposed encoding scheme tries to minimise the bases per letter ratio while controlling the redundancy. They have experimented with three different types of data with a value of redundancy as 0.75. In the randomised simulation setup, it was observed that the proposed algorithm was able to correctly retrieve the stored data in our experiments about 94% of the time. In the situation, where redundancy was increased to 1, the authors were able to retrieve all the information correctly in the proposed experiments.

RNA secondary structured logic gates for profiling the microRNA cancer biomarkers

Deregulation of microRNAs expression is symptomatic of cancer disease and occurs before the awareness of cancer signs. Early detection of cancer disease can improve or drop the disease entirely. DNA computing is an emerging field of detecting microRNAs based on toehold-mediated strand displacement reactions, which is a more efficient method than the commonly used method like real-time PCR. Accuracy and cost of diagnostic applications are essential criteria that are achieved by using the DNA logic gates based on the DNA computing method. In this study, the authors proposed the multi-input liver cancer biosensor with the RNA secondary structure motifs as the computational module and two approaches are suggested.

Ensemble of Deep Recurrent Neural Networks for Identifying Enhancers via Dinucleotide Physicochemical Properties

Enhancers are short deoxyribonucleic acid fragments that assume an important part in the genetic process of gene expression. Due to their possibly distant location relative to the gene that is acted upon, the identification of enhancers is difficult. There are many published works focused on identifying enhancers based on their sequence information, however, the resulting performance still requires improvements. Using deep learning methods, this study proposes a model ensemble of classifiers for predicting enhancers based on deep recurrent neural networks. The input features of deep ensemble networks were generated from six types of dinucleotide physicochemical properties, which had outperformed the other features. In summary, our model which used this ensemble approach could identify enhancers with achieved sensitivity of 75.5%, specificity of 76%, accuracy of 75.5%, and MCC of 0.51. For classifying enhancers into strong or weak sequences, our model reached sensitivity of 83.15%, specificity of 45.61%, accuracy of 68.49%, and MCC of 0.312. Compared to the benchmark result, our results had higher performance in term of most measurement metrics. The results showed that deep model ensembles hold the potential for improving on the best results achieved to date using shallow machine learning methods.

Biocomputing with Nanostructures on Lipid Bilayers

Biocomputation is the algorithmic manipulation of biomolecules. Nanostructures, most notably DNA nanostructures and nanoparticles, become active substrates for biocomputation when modified with stimuli-responsive, programmable biomolecular ligands. This approach-biocomputing with nanostructures ("nano-bio computing")-allows autonomous control of matter and information at the nanoscale; their dynamic assemblies and beneficial properties can be directed without human intervention. Recently, lipid bilayers interfaced with nanostructures have emerged as a new biocomputing platform. This new nano-bio interface, which exploits lipid bilayers as a chemical circuit board for information processing, offers a unique reaction space for realizing nanostructure-based computation at a previously unexplored dimension. In this Concept, recent advances in nano-bio computing are briefly reviewed and the newly emerging concept of biocomputing with nanostructures on lipid bilayers is introduced.

A CRISPR/Cas9-based central processing unit to program complex logic computation in human cells

Controlling gene expression with sophisticated logic gates has been and remains one of the central aims of synthetic biology. However, conventional implementations of biocomputers use central processing units (CPUs) assembled from multiple protein-based gene switches, limiting the programming flexibility and complexity that can be achieved within single cells. Here, we introduce a CRISPR/Cas9-based core processor that enables different sets of user-defined guide RNA inputs to program a single transcriptional regulator (dCas9-KRAB) to perform a wide range of bitwise computations, from simple Boolean logic gates to arithmetic operations such as the half adder. Furthermore, we built a dual-core CPU combining two orthogonal core processors in a single cell. In principle, human cells integrating multiple orthogonal CRISPR/Cas9-based core processors could offer enormous computational capacity.

High-Performance Biocomputing in Synthetic Biology-Integrated Transcriptional and Metabolic Circuits

Biocomputing uses molecular biology parts as the hardware to implement computational devices. By following pre-defined rules, often hard-coded into biological systems, these devices are able to process inputs and return outputs—thus computing information. Key to the success of any biocomputing endeavor is the availability of a wealth of molecular tools and biological motifs from which functional devices can be assembled. Synthetic biology is a fabulous playground for such purpose, offering numerous genetic parts that allow for the rational engineering of genetic circuits that mimic the behavior of electronic functions, such as logic gates. A grand challenge, as far as biocomputing is concerned, is to expand the molecular hardware available beyond the realm of genetic parts by tapping into the host metabolism. This objective requires the formalization of the interplay of genetic constructs with the rest of the cellular machinery. Furthermore, the field of metabolic engineering has had little intersection with biocomputing thus far, which has led to a lack of definition of metabolic dynamics as computing basics. In this perspective article, we advocate the conceptualization of metabolism and its motifs as the way forward to achieve whole-cell biocomputations. The design of merged transcriptional and metabolic circuits will not only increase the amount and type of information being processed by a synthetic construct, but will also provide fundamental control mechanisms for increased reliability.

Targeting Amyloid Aggregation: An Overview of Strategies and Mechanisms

Amyloids result from the aggregation of a set of diverse proteins, due to either specific mutations or promoting intra- or extra-cellular conditions. Structurally, they are rich in intermolecular β-sheets and are the causative agents of several diseases, both neurodegenerative and systemic. It is believed that the most toxic species are small aggregates, referred to as oligomers, rather than the final fibrillar assemblies. Their mechanisms of toxicity are mostly mediated by aberrant interactions with the cell membranes, with resulting derangement of membrane-related functions. Much effort is being exerted in the search for natural antiamyloid agents, and/or in the development of synthetic molecules. Actually, it is well documented that the prevention of amyloid aggregation results in several cytoprotective effects. Here, we portray the state of the art in the field. Several natural compounds are effective antiamyloid agents, notably tetracyclines and polyphenols. They are generally non-specific, as documented by their partially overlapping mechanisms and the capability to interfere with the aggregation of several unrelated proteins. Among rationally designed molecules, we mention the prominent examples of β-breakers peptides, whole antibodies and fragments thereof, and the special case of drugs with contrasting transthyretin aggregation. In this framework, we stress the pivotal role of the computational approaches. When combined with biophysical methods, in several cases they have helped clarify in detail the protein/drug modes of interaction, which makes it plausible that more effective drugs will be developed in the future.

Hybrid Mesoporous Nanocarriers Act by Processing Logic Tasks: Toward the Design of Nanobots Capable of Reading Information from the Environment

Here, we present the design of smart nanodevices capable of reading molecular information from the environment and acting accordingly by processing Boolean logic tasks. As proof of concept, we prepared Au-mesoporous silica (MS) nanoparticles functionalized with the enzyme glucose dehydrogenase (GDH) on the Au surface and with supramolecular nanovalves as caps on the MS surface, which is loaded with a cargo (dye or drug). The nanodevice acts as an AND logic gate and reads information from the solution (presence of glucose and nicotinamide adenine dinucleotide (NAD⁺)), which results in cargo release. We show the possibility of coimmobilizing GDH and the enzyme urease on nanoparticles to mimic an INHIBIT logic gate, in which the AND gate is switched off by the presence of urea. We also show that such nanodevices can deliver cytotoxic drugs in cancer cells by recognizing intracellular NAD⁺ and the presence of glucose.

Smart materials on the way to theranostic nanorobots: Molecular machines and nanomotors, advanced biosensors, and intelligent vehicles for drug delivery

BACKGROUND

Theranostics, a fusion of two key parts of modern medicine - diagnostics and therapy of the organism's disorders, promises to bring the efficacy of medical treatment to a fundamentally new level and to become the basis of personalized medicine. Extrapolating today's progress in the field of smart materials to the long-run prospect, we can imagine future intelligent agents capable of performing complex analysis of different physiological factors inside the living organism and implementing a built-in program thereby triggering a series of therapeutic actions. These agents, by analogy with their macroscopic counterparts, can be called nanorobots. It is quite obscure what these devices are going to look like but they will be more or less based on today's achievements in nanobiotechnology.

SCOPE OF REVIEW

The present Review is an attempt to systematize highly diverse nanomaterials, which may potentially serve as modules for theranostic nanorobotics, e.g., nanomotors, sensing units, and payload carriers.

MAJOR CONCLUSIONS

Biocomputing-based sensing, externally actuated or chemically "fueled" autonomous movement, swarm inter-agent communication behavior are just a few inspiring examples that nanobiotechnology can offer today for construction of truly intelligent drug delivery systems.

GENERAL SIGNIFICANCE

The progress of smart nanomaterials toward fully autonomous drug delivery nanorobots is an exciting prospect for disease treatment. Synergistic combination of the available approaches and their further development may produce intelligent drugs of unmatched functionality.

Muscle-Cell-Based "Living Diodes"

A new type of diode that is made entirely of electrically excitable muscle cells and nonexcitable fibroblast cells is designed, fabricated, and characterized. These two cell types in a rectangular pattern allow the signal initiated on the excitable side to pass to the nonexcitable side, and not in the opposite direction.

An Enzyme-based 1:2 Demultiplexer Interfaced with an Electrochemical Actuator

An enzyme-based 1:2 demultiplexer is designed in a flow system composed of three cells where each one is modified with a different enzyme: hexokinase, glucose dehydrogenase and glucose-6-phosphate dehydrogenase. The Input signal activating the biocatalytic cascade is represented by glucose, while the Address signal represented by ATP is responsible for directing the Input signal to one of the output channels, depending on the logic value of the Address. The biomolecular 1:2 demultiplexer is extended to include two electrochemical actuators releasing entrapped DNA molecules in the active output channel. The modular design of the system allows for easy exchange and extension of the functional elements. The present demultiplexer can be easily integrated in various biomolecular logic systems, including different logic gates based on the enzyme- or DNA-based reactions, as well as containing different chemical actuators, for example, with a biomolecular release function.

DNA Nanostructures as Smart Drug-Delivery Vehicles and Molecular Devices

DNA molecules can be assembled into custom predesigned shapes via hybridization of sequence-complementary domains. The folded structures have high spatial addressability and a tremendous potential to serve as platforms and active components in a plethora of bionanotechnological applications. DNA is a truly programmable material, and its nanoscale engineering thus opens up numerous attractive possibilities to develop novel methods for therapeutics. The tailored molecular devices could be used in targeting cells and triggering the cellular actions in the biological environment. In this review we focus on the DNA-based assemblies - primarily DNA origami nanostructures - that could perform complex tasks in cells and serve as smart drug-delivery vehicles in, for example, cancer therapy, prodrug medication, and enzyme replacement therapy.

Bioelectronic Interface Connecting Reversible Logic Gates Based on Enzyme and DNA Reactions

It is believed that connecting biomolecular computation elements in complex networks of communicating molecules may eventually lead to a biocomputer that can be used for diagnostics and/or the cure of physiological and genetic disorders. Here, a bioelectronic interface based on biomolecule-modified electrodes has been designed to bridge reversible enzymatic logic gates with reversible DNA-based logic gates. The enzyme-based Fredkin gate with three input and three output signals was connected to the DNA-based Feynman gate with two input and two output signals-both representing logically reversible computing elements. In the reversible Fredkin gate, the routing of two data signals between two output channels was controlled by the control signal (third channel). The two data output signals generated by the Fredkin gate were directed toward two electrochemical flow cells, responding to the output signals by releasing DNA molecules that serve as the input signals for the next Feynman logic gate based on the DNA reacting cascade, producing, in turn, two final output signals. The Feynman gate operated as the controlled NOT gate (CNOT), where one of the input channels controlled a NOT operation on another channel. Both logic gates represented a highly sophisticated combination of input-controlled signal-routing logic operations, resulting in redirecting chemical signals in different channels and performing orchestrated computing processes. The biomolecular reaction cascade responsible for the signal processing was realized by moving the solution from one reacting cell to another, including the reacting flow cells and electrochemical flow cells, which were organized in a specific network mimicking electronic computing circuitries. The designed system represents the first example of high complexity biocomputing processes integrating enzyme and DNA reactions and performing logically reversible signal processing.

The Essential Component in DNA-Based Information Storage System: Robust Error-Tolerating Module

The size of digital data is ever increasing and is expected to grow to 40,000 EB by 2020, yet the estimated global information storage capacity in 2011 is <300 EB, indicating that most of the data are transient. DNA, as a very stable nano-molecule, is an ideal massive storage device for long-term data archive. The two most notable illustrations are from Church et al. and Goldman et al., whose approaches are well-optimized for most sequencing platforms - short synthesized DNA fragments without homopolymer. Here, we suggested improvements on error handling methodology that could enable the integration of DNA-based computational process, e.g., algorithms based on self-assembly of DNA. As a proof of concept, a picture of size 438 bytes was encoded to DNA with low-density parity-check error-correction code. We salvaged a significant portion of sequencing reads with mutations generated during DNA synthesis and sequencing and successfully reconstructed the entire picture. A modular-based programing framework - DNAcodec with an eXtensible Markup Language-based data format was also introduced. Our experiments demonstrated the practicability of long DNA message recovery with high error tolerance, which opens the field to biocomputing and synthetic biology.

Realization of Associative Memory in an Enzymatic Process: Toward Biomolecular Networks with Learning and Unlearning Functionalities

We report a realization of an associative memory signal/information processing system based on simple enzyme-catalyzed biochemical reactions. Optically detected chemical output is always obtained in response to the triggering input, but the system can also "learn" by association, to later respond to the second input if it is initially applied in combination with the triggering input as the "training" step. This second chemical input is not self-reinforcing in the present system, which therefore can later "unlearn" to react to the second input if it is applied several times on its own. Such processing steps realized with (bio)chemical kinetics promise applications of bioinspired/memory-involving components in "networked" (concatenated) biomolecular processes for multisignal sensing and complex information processing.