Biological processing devices and computer systems, and related methods

Abstract

A biological computer system can include an interface. The interface can include a field-programmable signal transceiver (FPST) and an interface controller coupled to the FPST. The system can further include a synthetic biological circuit (SBC) disposed on the FPST. The interface controller can be configured to encode data in signals provided to the synthetic biological circuit by the FPST (e.g., base-N electrical signals, wherein N > 2). The SBC can perform a computational task based on the data encoded in the signals, and can provide electrical signals encoding a result of the computational task. The computational task can be an ill-posed inverse task. The SBC can include a confluent array of neurons. The SBC can include one or more biologic qubits, which can perform portions of the computational task using quantum computing.

Description

Docket No. 229303-701601 / PCTBIOLOGICAL PROCESSING DEVICES AND COMPUTER SYSTEMS,AND RELATED METHODS BACKGROUND

[0001] Machine learning (ML) and artificial intelligence (Al) are computational paradigms designed to enable systems to learn patterns from data and make predictions or decisions without explicit programming. Al broadly encompasses techniques that simulate aspects of human cognition, while ML focuses on algorithms that improve performance through experience. Among these, artificial neural networks (ANNs) are a class of models inspired by the structure and function of biological neural systems. ANNs consist of interconnected nodes, or “neurons,” organized in layers that process input signals through weighted connections, enabling tasks such as classification, pattern recognition, and decision-making.

[0002] Biological neural networks (BNNs), in contrast, are complex systems of living neurons, which transmit information via electrochemical signals across synapses. These networks exhibit adaptive and dynamic properties, such as plasticity and self-organization, which allow BNNs to learn and respond to environmental stimuli. The conceptual parallels between ANNs and biological neural networks have driven research into hybrid systems and bio-inspired computational architectures, aiming to leverage the robustness and adaptability of biological systems with the scalability and precision of artificial models.SUMMARY

[0003] According to an aspect of the present disclosure, a biological computer system is disclosed. The biological computer system includes an interface including a multi-electrode array (MEA) and an interface controller communicatively coupled to the MEA; and a synthetic biological circuit disposed on the MEA. The interface controller is configured to encode data in base-N electrical signals provided to the synthetic biological circuit by the MEA, where N > 2.

[0004] According to another aspect of the present disclosure, a method is disclosed. The method includes providing, by a multi-electrode array (MEA) to a synthetic biological circuit disposed on the MEA, one or more base-N electrical signals encoding data, where N > 2.

[0005] According to another aspect of the present disclosure, a biological computer system is disclosed. The biological computer system includes an interface including a multi-electrode array (MEA) and an interface controller communicatively coupled to the MEA; and a syntheticDocket No, 229303-701601 / PCTbiological circuit disposed on the MEA and configured to: perform an ill-posed inverse task based on data encoded in one or more signals provided by the MEA, and provide one or more signals encod ing a result of the ill-posed inverse task.

[0006] Accord ing to another a spect of the present disclosure, a method is d isclosed. The method includes providing, by a multi-electrode array (MEA) to a synthetic biological circuit disposed on the MEA, one or more signals encoding data; performing, by the synthetic biological circuit, an ill-posed inverse task based on the data encoded in the one or more signals provided by the MEA; and providing, by the synthetic biological circuit, one or more signals encoding a result of the ill-posed inverse task,

[0007] According to another aspect of the present disclosure, a biological computer system is provided. The biological computer system includes an interface including a multi-electrode array (MEA) and an interface controller communicatively coupled to the MEA; and a synthetic biological circuit disposed on the MEA and configured to perform a computational task based on data encoded in one or more signals provided by the MEA, wherein the synthetic biological circuit includes a plurality of neurons, and one or more neurons in the plurality of neurons are genetically engineered.

[0008] According to another aspect of the present disclosure, a method is disclosed. The method includes providing, by a multi -electrode array (MEA) to a synthetic biological circuit disposed on the MEA, one or more signals encoding data; performing, by the synthetic biological circuit, a computational task based on the data encoded in the one or more signals provided by the MEA; and providing, by the synthetic biological circuit, one or more signals encoding a result of the computational task, wherein the synthetic biological circuit includes a plurality of neurons, and one or more neurons in the plurality of neurons are genetically engineered.

[0009] According to another aspect of the present disclosure, a biological processing device is disclosed. The device includes a multi-electrode array (MEA); and a synthetic biological circuit disposed on the MEA, wherein the synthetic biological circuit is configured to decode data encoded in base-N electrical signals provided by the MEA, perform a computational task based on the decoded data, and provide one or more signals encoding a result of the computational task to the MEA, wherein N > 2.

[0010] According to another aspect of the present disclosure, a biological processing device is disclosed. The device includes a multi-electrode array (MEA); and a synthetic biological circuitDocket No. 229303-701601 / PCTdisposed on the MEA, wherein the synthetic biological circuit is configured to perform an ill-posed inverse task based on data encoded in one or more signals provided by the MEA, and provide one or more signals encoding a result of the ill-posed inverse task to the MEA.

[0011] According to another aspect of the present disclosure, a biological processing device is disclosed. The device includes a multi-electrode array (MEA); and a synthetic biological circuit disposed on the MEA, wherein the synthetic biological circuit includes a plurality of neurons, and one or more neurons in the plurality of neurons are genetically engineered. The synthetic biological circuit is configured to perform a computational task based on data encoded in one or more signals provided by the MEA, and provide one or more signals encoding a result of the computation to the MEA.

[0012] According to another aspect of the present disclosure, a method of fabricating a biological processing device is disclosed. The method includes (a) culturing a population of neuronal cells, wherein the population of neuronal cells comprise at least one neuron cell subtype and at least one glial cell subtype; (b) dispersing and seeding a population of neuronal cells on a surface of a microelectrode array (MEA) comprising a plurality of microelectrodes, wherein the neuronal cell seeding density is sufficient to form a confluent neuronal array, and wherein the population of neuronal cells form one or more synthetic biological circuits on the plurality of microelectrodes; and (c) communicatively coupling the MEA with an interface controller configured to encode data in base-N electrical signals provided to the confluent neuronal array by the MEA, wherein N > 2.

[0013] According to another aspect of the present disclosure, a method of fabricating a biological processing device is provided. The method includes (a) culturing a population of neuronal cells, wherein the population of neuronal cells comprise at least one genomic disruption in a target sequence of a gene; (b) dispersing and seeding a population of neuronal cells on a surface of a microelectrode array comprising a plurality of microelectrodes, wherein the neuronal cell seeding density is sufficient to form a confluent neuronal array, and wherein the population of neuronal cells form one or more synthetic biological circuits on the plurality of microelectrodes; and (c) communicatively coupling the MEA with an interface controller configured to encode data in base-N electrical signals provided to the confluent neuronal array by the MEA, wherein N > 2.

[0014] According to another aspect of the present disclosure, a biological computer system is disclosed. The system includes an interface including a field-programmable signal transceiverDocket No. 229303-701601 / PCT(FPST) and an interface controller communicatively coupled to the FPST; and a synthetic biological circuit disposed on the FPST and configured to perform an inverse computational task based on data encoded in one or more signals provided by the FPST, and provide one or more signals encoding a result of the inverse computational task.

[0015] According to another aspect of the present disclosure, a method is disclosed. The method includes providing, by a field-programmable signal transceiver (FPST) to a synthetic biological circuit disposed on the FPST, one or more signals encoding data; performing, by the synthetic biological circuit, an inverse computational task based on the data encoded in the one or more signals provided by the FPST; and providing, by the synthetic biological circuit, one or more signals encoding a result of the inverse computational task.BRIEF DESCRIPTION OF DRAWINGS

[0016] The accompanying drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures is represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing.

[0017] FIG. 1A illustrates an example system architecture for a biological computational system, in accordance with some embodiments.

[0018] FIG. 1B is a flowchart of an example method for performing a computational task with one or more synthetic biologic circuits (SBCs).

[0019] FIG. 1C is a flowchart of an example method for training one or more synthetic biologic circuit (SBCs) to perform a computational task.

[0020] FIG. 1D illustrates examples of two types of stimulation that can be applied to neurons.

[0021] FIG. 1E illustrates an example of quantum ion superposition for synaptic signaling.

[0022] FIG. 1F illustrates, by example, quantum principles underlying neural processing.

[0023] FIG. 1G illustrates an example of entangled phosphorous atoms.

[0024] FIG. 1H illustrates an example of entangled Posner clusters.

[0025] FIG. 2A illustrates a truth table for a two-input adder with 3 -valued inputs and a 5- valued output.

[0026] FIG. 2B illustrates a truth table for a set of multi-input 4-valued logic gates.

[0027] FIG. 3 shows an exemplary protocol for preparing a synthetic biological circuit or SBC and glutamatergic neurons after 18 days in culture.Docket No. 229303-701601 / PCT

[0028] FIG. 4 shows a MaxOne electrical activity scan after 15 days.

[0029] FIG. 5 shows a MaxOne electrical activity scan after 29 days.

[0030] FIG. 6 shows the MaxOne electrical activity scan of neurons alone and neurons in co¬ culture with astrocytes as indicated, 9 days post-cell seeding.

[0031] FIG. 7 illustrates an example of Biological Neural Network (BNN) hardware, including components configured to interface with biological substrates for computational tasks.

[0032] FIG. 8 depicts a schematic representation of an example system architecture for integrating biological computing elements with electronic control modules.

[0033] FIG. 9 shows a block diagram of signal processing pathways within an example biological computing system, including input / output channels and feedback loops.

[0034] FIG. 10 illustrates an embodiment of a chamber configuration for housing biological substrates and associated microelectrode arrays,

[0035] FIG. 11 depicts a flow diagram of data acquisition and stimulation phases in an example biological computing process.

[0036] FIG. 12 shows multiple example configurations for electrode placement and signal routing within a biological computing chamber.

[0037] FIG. 13 illustrates an example of a biological computing substrate integrated with a microelectronic interface.

[0038] FIG. 14 depicts a schematic of an example interface controller for managing bidirectional communication between biological and electronic components.

[0039] FIGS. 15A, 15B, and 15C show various embodiments of biological computing substrates, including neuron-only and neuron-plus-astrocyte configurations. FIG. 15D shows an exemplary training session of the different reading zones of a biological circuit on the MEA chip across different trials.

[0040] FIGS. 16A, 16B, and 16C illustrate example stages of biological substrate preparation and integration with microelectrode arrays.

[0041] FIG. 17 depicts an example high-level architecture for scaling biological computing elements across multiple chambers.

[0042] FIG. 18 shows a schematic of example signal amplification and conditioning circuitry for biological data streams.Docket No. 229303-701601 / PCT

[0043] FIG. 19 illustrates an embodiment of a biological computing chamber with integrated fluidic control systems.

[0044] FIG. 20 depicts an example configuration for multi-chamber biological computing systems connected via a network interface.

[0045] FIG. 21 A-21 J show example phases and components of biological qubit operation. FIG.21 A shows an example microelectrode array (MEA) layout. FIG. 21B shows an example chip architecture for hosting biological substrates. FIG. 21C shows an example of neurons cultured on a chip. FIG. 21D shows an example representation of a biological qubit. FIG. 21E shows example spatially coded digit representations within a biological network. FIG. 21F shows an example of an ambient recording phase. FIG. 21G shows an example of a sensory stimulation phase. FIG. 21H shows an example of a monitoring phase. FIG. 21I shows an example of a feedback stimulation phase. FIG. 21J shows an example scaling approach for multiple biological qubits on a single chip.DETAILED DESCRIPTION

[0046] The scale of computational resources used to train cutting-edge artificial intelligence (Al) models is immense. Many state-of-the-art generative AI models are trained for months on vast arrays of graphics processing units (GPUs) with computational bandwidth on the order of tens of exaflops, and those computational demands are increasing by orders of magnitude with each generation of Al technology. For example, Open Al’s GPT-4 model, released in 2023, is reported to have cost approximately $80M to train using 25,000 Nvidia A100 GPUs for 90-100 days. Less than two years later, the training of Open Al’s GPT-5 model is expected to cost between SIB and S2.5B using 500,000 Nvidia H100 GPUs for 90-100 days. Operating trained generative Al models is also computationally demanding and expensive.

[0047] The amount of electrical power used by the GPUs and other silicon-based integrated circuits (“chips”) on which generative Al models are trained and deployed is also significant. During the training of the GPT-4 model, the A100 GPUs alone used up to 7.5 MW of power, and the H100 GPUs used to train the GPT-5 model are expected to use up to 350 MW of power. As Al models continue to grow in complexity and the integration of Al applications into business workflows and consumer products continues apace, new energy sources are likely to be needed to power the data centers in which such models are trained and deployed. For example, Microsoft recently reached an agreement to purchase up to 800 MW of power from Constellation Energy,Docket No. 229303-701601 / PCTwhich plans to provide the power by reviving a shuttered nuclear reactor at the infamous Three Mile Island nuclear power plant.

[0048] The sharp increases in scale, cost, and power demands of the chips used to train successive versions of the GPT model indicate that the silicon-based approach to delivering ever- increasing computational bandwidth at ever-decreasing cost (i.e., integrated circuit scaling in accordance with Moore’s Law) is not keeping pace with Al systems’ staggering demand for computational resources. In some respects, silicon-based chips are approaching fundamental physical limits to further scaling. For example, transistor widths in high-performance GPUs and CPUs are already less than 10 nm, and it is unlikely that transistor width can be reduced below the width of a single atom (e.g, 0.1 nm).

[0049] The deep neural networks (DNNs) used in generative Al and other cutting-edge applications are designed to model or approximate biological neural networks (BNNs), but DNNs are incredibly complex models which require enormous amounts of computational resources and electrical power to train and run in silico on the binary logic gates of conventional chips. Historically, Moore’s law has provided exponentially more computational resources at fixed or decreasing operational cost, but the limits of Moore’s law are rapidly approaching. Accordingly, there is a need for alternative platforms that can provide scalable, energy-efficient computational resources in the Al era.

[0050] Theoretically, the energy efficiency of in silico DNNs can be improved by running the DNNs on programmable chips built from multi-valued logic gates (<?.g., N- valued logic gates, where N > 3) rather than binary logic gates, because devices built from multi-valued logic gates are much more efficient and less complex than devices built from binary (2-valued) logic gates. See, e.g., [1], However, efforts to design reliable, scalable multi-valued logic (MVL) gates in silico have met with limited success, despite increasing interest from the research community over the last half-century. See [1], [3],

[0051] On the other hand, in vivo biological computational systems (e.g., brains) are many orders of magnitude more energy-efficient than silicon-based computational systems (see [2]), and are capable of processing information at a scale and speed adequate for general intelligence (e.g., at the speed of quantum computing). The fundamental building blocks of such systems (e.g., cells, such as neurons) have evolved over millions of years into reliable, scalable, computational units that communicate with each other using analog signaling. The inventors have recognized andDocket No. 229303-701601 / PCTappreciated that these biological computational units can be configured and / or trained to provide energy-efficient computational resources at scale in vitro. For example, the inventors have recognized and appreciated that cultures of multiple biological computational units (e.g., neurons) can be configured and / or trained to function as programmable (e.g., trainable), energy-efficient, scalable, synthetic biological circuits (e.g., biologic qubits, deconvolution circuits, circuits capable of performing ill-posed inverse computational tasks, multi-valued logic circuits and / or gates (sometimes referred to herein as “wetware MVL circuits” or “wetware MVL gates”), etc. in vitro, and have further recognized and appreciated that such synthetic biological circuits can be communicatively coupled to form programmable (e.g., trainable), energy-efficient, scalable, processing systems (e.g., multi-valued processing systems (sometimes referred to herein as “wetware MVL processing systems” or “wetware MVL systems”), arithmetic-logic units (ALU), general-purpose processors, etc.) in vitro.

[0052] The present disclosure describes techniques for configuring and / or training synthetic biological circuits (e.g., biological computational units, biologic qubits, etc.) to perform computational tasks (e.g., ill-posed inverse computational tasks, deconvolution tasks, or any other suitable computational tasks), function as wetware MVL circuits (e.g., adders, logic gates, etc.), etc. The present disclosure also describes techniques for connecting such synthetic biological circuits to form biological computer systems (e.g., programmable, multi-valued ALUs) in vitro, and techniques for interfacing the biological computer systems with conventional in silico computers and networks.

[0053] The present disclosure describes examples of biological computer systems (e.g., wetware multi-valued logic circuits) implemented using the techniques described herein. These biological computer systems, which include a 3-valued adder, 4-valued logic gates (e.g., AND, OR, NAND, and NOR gates), and a deconvolution circuit can be used to construct even more complex biological computer systems (e.g., 4-valued ALUs).

[0054] Briefly, provided herein are (1) examples of biological computational systems; (2) cells for biological circuitry; and (3) genetic modifications.Definitions

[0055] The terminology used herein is for the purpose of description and should not be regarded as limiting.Docket No. 229303-701601 / PCT

[0056] The term “generative model” as used herein may generally refer to a type of machine learning model that is trained on existing data to enable the generative model to generate, based on an input or prompt, new data that shares characteristics similar to that of the training data. In some examples, a generative model may handle text. In these examples, the generative model may accept text prompts and produce text outputs. Any suitable type of Al model can be used, including predictive models, generative Al (“Gen Al”) models, etc. Predictive models can analyze historical data, identify patterns in that data, and make inferences (e.g., produce predictions or forecast outcomes) based on the identified patterns. Some non-limiting examples of predictive models include neural networks (e.g., deep neural networks (DNNs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), learning vector quantization (L. VQ) models, etc.), regression models (e.g., linear regression models, logistic regression models, linear discriminant analysis (LDA) models, etc.), decision trees, random forests, support vector machines (SVMs), naive Bayes models, classifiers, etc.

[0057] “Machine learning” may refer to the application of certain techniques (e.g., pattern recognition and / or statistical inference techniques) by computer systems to perform specific tasks. Machine learning techniques (automated or otherwise) may be used to build data analytics models based on sample data (e.g, “training data”) and to validate the models using validation data (e.g., “testing data”). The sample and validation data may be organized as sets of records (e.g., “observations” or “data samples”), with each record indicating values of specified data fields (e.g., “independent variables,” “inputs,” “features,” or “predictors”) and corresponding values of other data fields (e.g, “dependent variables,” “outputs,” or “targets”). Machine learning techniques may be used to train models to infer the values of the outputs based on the values of the inputs. When presented with other data (e.g, “inference data”) similar to or related to the sample data, such models may accurately infer the unknown values of the targets of the inference dataset.

[0058] The term “approximately,” the phrase “approximately equal to,” and other similar phrases, as used in the specification and the claims (e.g, “X has a value of approximately Y” or “X is approximately equal to Y”), should be understood to mean that one value (X) is within a predetermined range of another value (Y). The predetermined range may be plus or minus 20%, 10%, 5%, 3%, 1%, 0.1%, or less than 0.1%, unless otherwise indicated.

[0059] Measurements, sizes, amounts, etc. may be presented herein in a range format. The description in range format is merely for convenience and brevity and should not be construed asDocket No. 229303-701601 / PCTan inflexible limitation on the scope of the invention. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range. For example, description of a range such as 10-20 inches should be considered to have specifically disclosed subranges such as 10-11 inches, 10-12 inches, 10-13 inches, 10-14 inches, 11 -12 inches, 11-13 inches, etc.

[0060] The indefinite articles “a” and “an,” as used in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.” The phrase “and / or,” as used in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and / or” should be construed in the same fashion, i.e,, “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and / or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and / or B”, when used in conjunction with open-ended language such as “comprising” can refer, in some embodiments, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

[0061] As used in the specification and m the claims, “or” should be understood to have the same meaning as “and / or” as defined above. For example, when separating items in a list, “or” or “and / or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of or “exactly one of,” or, when used in the claims, “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used shall only be interpreted as indicating exclusive alternatives (i.e., “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.” “Consisting essentially of,” when used in the claims, shall have its ordinary meaning as used in the field of patent law.

[0062] As used in the specification and in the claims, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every’ element specifically listed within the list of elements and not excluding anyDocket No. 229303-701601 / PCTcombinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and / or B”) can refer, in some embodiments, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

[0063] The use of “including,” “comprising,” “having,” “containing,” “involving,” and variations thereof, is meant to encompass the items listed thereafter and additional items.

[0064] Use of ordinal terms such as “first,” “second,” “third,” etc,, in the claims to modify a claim element does not by itself connote any priority, precedence, or order of one claim element over another or the temporal order in which acts of a method are performed. Ordinal terms are used merely as labels to distinguish one claim element having a certain name from another element having a same name (but for use of the ordinal term), to distinguish the claim elements.(1) Biological Computational Systems

[0065] Described herein are embodiments of a biological computational system (or “biological computer”) usable to perform general-purpose computing. In some embodiments, a biological computational system includes one or more synthetic biological circuits. As used herein, a “synthetic biological circuit” or “SBC” can include one or more biological computational units (e.g, cells, neurons, etc.) configured to perform one or more logical functions in vitro. For example, a synthetic biological circuit can include one or more biological neurons (e.g., a biological neural network) configured to function as a logic circuit (e.g., a multi-valued logic circuit) or a logic gate (e.g., a multi-valued logic gate).

[0066] Biological neurons are near infinitely scalable, energy efficient (especially as compared to silicon based processors or in silico processing), small, and produce very little heat (e.g., as compared to silicon based processors or in silico processing). For example, a biological neural network in a multi-electrode array or other neural processing unit (e.g., a cell excitation andDocket No. 229303-701601 / PCTmeasurement device) has an energy use per synapse of about 2E-10 Joules. In contrast, the energy use per transistor in an example silicon processing device is about 2E-7 Joules - an increase in energy use of 3 orders of magnitude relative to a synapse. Additionally, biological neural networks are fault tolerant, and in some instances can withstand the destruction of large portions of the biological neural network while continuing to function properly. Biological neural networks also exhibit neuroplasticity, which enables highly adaptable intelligence that is suitable for many different applications.

[0067] In some embodiments, the biological computational system includes an interface configured to generate input signals (e.g., electrical, chemical, and / or optical signals) that can be interpreted by the SBCs, and to interpret output signals (e.g., electrical, chemical, and / or optical signals) generated by the SBCs. The input signals can encode data and / or instructions, and the output signals can encode data produced by the SBCs in response to the input signals. Thus, the interface can function as an encoder / decoder. In some embodiments, a biological computer and an in silico computer can communicate with each other via the interface.

[0068] In some embodiments, the interface includes a multi-electrode array (MEA), an optics¬ based equivalent to an MEA that uses optical input and / or output signals (“optical MEA”), and / or a chemical-based equivalent to an MEA that uses chemical emitters and / or chemical sensors (“chemical MEA”). The MEA, optical MEA, chemical MEA, and hybrids of MEAs, optical MEAs, and / or chemical MEAs are referred to herein as cell excitation and measurement devices, or simply as multi-input arrays (MI As).

[0069] In some embodiments, the biological computer’s SBCs can include cortical neuronal cells differentiated from human induced pluripotent stem cells (hIPSCs) or harvested from embryonic animals such as embryonic mice. In a biological neural network, data can be encoded as spikes of action potentials (e.g., voltage variation across a cell membrane) of a population of biological neurons. Neural cells communicate using spiking electrical activity via a biological process called an action potential, or more colloquially ‘firing.’ The above-described cortical neuronal cells can form dense connections with rich spiking activity when plated on, for example, a CMOS-based high-density MEA.

[0070] In some embodiments, the interface is configured to encode base-N logic values in the input signals provided to the biological computer, and to interpret the signals generated by the biological computer as signals encoding base-K logic values, where N > 3, K > 3, and N and KDocket No. 229303-701601 / PCTcan have the same value or different values. In some examples, the input signals provided to the biological computer are base-N, and the output signals generated by the biological computer are base-K. For example, with an electrode of an MEA capable of generating and / or sensing electrical signals with voltages between zero and eight volts, the interface can encode base-4 logic values (e.g., 0, 1, 2, and 3) as values within four discrete voltage ranges (e.g., 0-2 V, 2-4 V, 4-6 V, and 6- 8 V), and can interpret action potentials of the neurons as base-8 logic values by decoding eight discrete voltage ranges (0-1 V, 1-2 V, 2-3 V, 3-4 V, 4-5 V, 5-6 V, 6-7 V, and 7-8 V) as eight base- 8 logic values (e.g., 0, 1, 2, 3, 4, 5, 6, and 7). Likewise, any suitable properties of optical signals (e.g., intensity, duration, frequency, etc.) can be used to encode base-N (or decode base-K) logic values, and any suitable properties of chemical signals can be used to encode base-N (or decode base-K) logic values.

[0071] Thus, the interface can receive binary input signals (e.g., from an in silico computer) representing instructions or input data for the biological computer, convert the binary input signals into base-N logic values representing the instructions or input data, encode the base-N logic values as corresponding electrical, optical, and / or chemical input signals, and provide those electrical, optical, and / or chemical input signals to the biological computer. Likewise, the interface can sense electrical, optical, and / or chemical signals produced by the biological computer, decode those electrical, optical, and / or chemical signals into corresponding base-K logic values, convert the base-K logic values into binary output signals representing the data produced by the biological computer, and provide those binary output signals to the in silico computer.

[0072] In some examples, the interface includes a two-dimensional (2D) or three-dimensional (3D) grid of excitation sites and a processing device or integrated circuit configured to encode the input signals to the biological computer and decode the output signals produced by the biological computer. The SBCs of the biological computer may be disposed on the grid of excitation sites. Alternatively, the interface may be connected to a processing device or integrated circuit (e.g., via a printed circuit board). In some embodiments, the processing device is a component of a system on a chip (SoC) that includes a network adapter, an analog to digital converter and / or a digital to analog converter.

[0073] Referring now to the drawings, FIG. 1A illustrates an example system architecture for a biological computational system 100, which includes one or more synthetic biological circuits (SBCs) in accordance with some embodiments. As shown, the biological computational systemDocket No. 229303-701601 / PCT100 includes an interface 105 connected to one or more computing devices 110 (e.g., in silico computing devices) via a network 120. The network 120 may be a local area network, a wide area network, a private network (e.g, an intranet), a public network (e.g, the Internet), or any suitable combination thereof. The connection between the interface 105 and the computing device(s) 110 may include wired connections, wireless connections, or a combination thereof.

[0074] The computing devices 110 may include physical machines and / or virtual machines hosted by physical machines. The physical machines may be rackmount servers, desktop computers, or other computing devices.

[0075] The interface 105 can include an MIA 130, which can include an MEA, an optical MEA, and / or a chemical MEA. An MIA 130 can contain multiple plates or shanks through which neural signals are obtained and / or delivered. In some embodiments, high-density MEA (HD-MEA) are used. The plates or shanks can be arranged in a grid or other array, and can function as neural interfaces that connect SBCs 135 to electronic circuitry. The interface 105 can include a chamber 140 suitable for housing one or more SBCs 135 and / or a solution or other medium (e.g., a saline solution). These SBCs 135 can include cultured neurons (e.g, cultured from stem cells) and / or extracted neurons (e.g., extracted from a rat brain). The biological neurons can be from a generic cell line, or from a cell line with specific traits to be tested. For example, the biological neurons may be cultured from stem cells of a person having a particular genotype, or from a particular person for whom a test is to be performed, or from a person having a particular pathology. In some embodiments, the neurons include cortical cells from embryonic rodent sources. In some embodiments, the neurons include cortical cells from human induced pluripotent stem cell (hIPSC) sources.

[0076] The neurons of the SBCs 135 can be grown or harvested from numerous sources via any suitable methods. Most in vitro electrophysiological investigations on neural cells have been conducted on primary neurons. This process involves disassociating cortical cells from the dissected cortices of embryos e.g., rodent embryos). These cells are then grown in nutrient rich medium and can be maintained for time periods on the order of months. These cultures can develop complicated morphology, with numerous dendritic and axonal connections, leading to functional biological neural networks (BNNs). In some embodiments, such cultures are developed from embryos (e.g, mouse embryos).Docket No. 229303-701601 / PCT

[0077] Alternatively, advances in stem cell engineering allow for stem cells (e.g., induced PSCs, embryonic PSCs, neural precursor cells, etc.) to be efficiently differentiated into monolayers of active cortical neurons which display mature functional properties. This method has the capability of differentiating both upper and lower layer cortical neurons as well as other neural phenotypes. This protocol uses a defined neural induction and maintenance media under specific culture conditions to generate a heterogeneous culture of cortical progenitor cells. Pluripotent cells can be differentiated using variety of techniques, including but not limited to the use of small molecules to recapitulate natural ontogeny, direct reprogramming through the use of viral or other vectors to insert or modify the expression of genes in a cell line to give rise to a specific or varied neural phenotype, or the use of other genetic modification techniques that give rise to a specified or varied neuronal cell types.

[0078] In some embodiments, neuron cultures (e.g., of long-term cortical neurons and / or other types of neurons) from hIPSCs and / or other sources are impl emented to form networks comparable to in vivo neuronal networks within organisms and / or in vitro networks found in primary neuronal cell cultures, along with appropriate biomarkers showing that cells are not only neural but also more specifically cortical. In addition, hIPSC-derived cells can survive for longer than 6 months with maintained activity and can be grown on an exponential scale, rendering the cost per cell relatively low at high volumes. This approach allows neuronal ‘wetware’ for computation to be grown and maintained in a functional way.

[0079] In some embodiments, the SBCs 135 include sparse, two-dimensional neural cultures (e.g., with thousands of non-overlapping neurons). In some embodiments, the SBCs 135 include much denser arrangements of neurons (e.g., with tens of thousands, hundreds of thousands, or millions of neurons overlapping one another to form a three-dimensional arrangement in which multiple neurons may be stacked vertically in addition to being arranged on a two-dimensional grid). The dense arrangement of neurons enables the neurons to form spontaneous three-dimensional (3D) structures such as neurospheres, effectively increasing the intelligence of the biological neural network that incorporates the neurons. In some embodiments, the dense arrangement of neurons includes at least 10,000 cells per square millimeter, at least 20,000 cells per square millimeter, or at least 50,000 cells per square millimeter. In some embodiments, the dense arrangement of neurons enables development of computational assemblies of the neurons.Docket No. 229303-701601 / PCT

[0080] In some examples, an SBC 135 includes two or more biological computational units (e.g., neurons) configured (e.g, trained, engineered, connected, etc.) to perform the function of a multi-valued logic circuit (e.g., gate). For example, an SBC 135 can be include neurons configured to perform the function a 3-valued adder or a 4-valued AND, OR, NAND, or NOR gate. In other examples, an SBC 135 can include neurons configured to perform the function of any suitable 4- valued logic circuit (e.g., gate) or any suitable N-valued logic circuit (e.g., gate), where N > 3. An SBC configured to perform the function of a multi-valued logic circuit (e.g., gate) may be referred to herein as a “wetware MVL circuit” or “wetware MVL gate.” In some examples, two or more of the SBCs 135 can be configured to implement a more complex computational function, such as the function of an N-valued encoder, decoder, adder, multiplier, programmable arithmetic-logic unit (ALU), etc. A set of SBCs configured to perform a complex computational function may be referred to herein as a “wetware MVL system.”

[0081] In some examples, an SBC includes two or more neurons. The neurons can be integrated in a confluent neuronal array. In some examples, the confluent neuronal array is two-dimensional (2D). In a 2D confluent neuronal array, the neurons may form a monolayer on the MEA. In some examples, the neurons are integrated in a three-dimensional (3D) structure (e.g., a brain organoid or a brain tissue). Such a structure can include two or more cell types. The cell types can include sensory neurons, motor neurons, interneurons, projection neurons, astrocytes, oligodendrocytes, pyramidal neurons, Purkinje cells, granule cells, microglial cells, radial glial cells, ependymal cells, endothelial cells, pericytes, choroid plexus epithelial cells, glutamatergic neurons, GABAergic neurons, dopaminergic neurons, cholinergic neurons, serotonergic neurons, neural progenitor cells, adult neural stem cells, and / or any combination thereof. In some examples, the neurons of the 3D structure include vitro-differentiated human neurons, human induced-pluripotent stem cell derived cells, human embryonic stem cell derived cells, or a combination thereof.

[0082] In some examples, the neurons of an SBC or a set of SBCs can be engineered (e.g., genetically engineered). Such genetic engineering can, for example, enhance performance (e.g., processing speed, electrical conductivity' of neurons, firing rate of neurons, robustness of neurons to electrical current, etc.) and / or enhance the SBC’s ability to reliably perform specific functions. In some examples, one or more of the neurons are genetically engineered to knock out one or more GABA receptors and / or one or more glutamate receptors. In some examples, the knock outDocket No. 229303-701601 / PCTincludes a genomic disruption in a target sequence of a gene encoding for the GABA receptor or a gene encoding for a glutamate receptor. In some examples, the knock out includes a CRISPR / Cas mediated genomic disruption in a target sequence of a gene. Additional examples of genetic engineering are described further below.

[0083] The biological neurons of the SBCs 135 can be disposed on the interface 105 (e.g., within the chamber 140 and on the MIA 130). The MIA 130 may include electrodes and / or light sources to provide electrical and / or optical stimulation of neurons. Additionally, or alternatively, the MIA 130 may include chemical generators or emitters that can release chemicals at locations on the MIA 130. Electrodes may provide electrical stimulation of neurons, light sources may provide optical or light-based stimulation of neurons, and chemical generators or emitters can provide chemical stimulation of neurons. In some embodiments, the electrodes, light sources, and / or chemical generators / emitters are arranged in a grid. This arrangement may enable targeted stimulation of neurons with pinpoint accuracy.

[0084] In some embodiments, many light emitting diodes (LEDs) may be arranged in a grid. In some embodiments, a screen may be interposed between one or more light source and the neurons. The screen may be opaque in areas where the neurons are not to be exposed to light, and the screen may be open or otherwise transparent to the light at areas where the neurons are to be exposed to light. Which regions of the screen are opaque and which regions are transparent may be adjusted as appropriate. In some embodiments, a display (e.g., a liquid crystal display or organic light emitting diode display) is used as the light source.

[0085] In some embodiments, the light sources comprise one or more lasers that may be movable to project light at target coordinates (e.g, at target neurons). For example, the laser may be attached to an actuator or servo-motor that can rotate the laser around multiple axes. In another example, the laser may be fixed, but one or more movable mirrors may direct light from the laser to target neurons or locations.

[0086] In some embodiments, a grid of chemical emitters is arranged on the MEA. Some non¬ limiting examples of chemical compounds that may be released by the chemical generators / emitters include neurotransmitters, dopamine, serotonin, glutamate, GABA, ACH, etc. Neurotransmitters are chemical compounds that condition neurons. For example, neurotransmitters may up regulate or down regulate the internal firing capacity of neurons exposed to those neurotransmitters.Docket No. 229303-701601 / PCT

[0087] In some embodiments, the interface 105 can apply multiple types of stimulus to neurons. For example, any combination of electrical, optical, and / or chemical stimulus may be applied to neurons sequentially and / or m parallel.

[0088] Responsive to one or more neurons of the SBCs 135 being excited, those neurons may generate or emit an electrical current, a voltage, a chemical, light, or any combination thereof. These signals may trigger other nearby neurons to generate or emit an electrical current, a voltage, a chemical, and / or light. This process may repeat, where excited neurons then excite still other neurons, and so on.

[0089] The MIA 130 may further include one or more electrical, optical, and / or chemical sensors suitable for detecting neuron activity. In some embodiments, the MIA 130 includes a grid of electrodes (e.g., an MEA) that can measure voltage and / or current at locations of neurons. In some embodiments, the same grid of electrodes can be used both for excitation of neurons and for measuring electrical activity of neurons responsive to such excitation. In some embodiments, a grid of chemical sensors is arranged on the MIA 130 to detect locations at which particular chemicals are present.

[0090] Neurons can be designed to fluoresce under certain conditions (e.g., responsive to stimulus). Optical sensors may be used to detect locations on the MIA 130 at which neurons are fluorescing (e.g, to detect which neurons have been stimulated and are generating an output). In some embodiments, the MIA 130 includes a grid of optical sensors. In some embodiments, the MIA 130 includes one or more cameras. Different regions within the fields of view of the cameras may be associated with different neurons and / or MIA coordinates. Images generated by the camera(s) can be used to determine locations on the MIA at which neurons have been activated. For example, each pixel in an image may be associated with a particular location on MIA 130. The camera can generate an image which can be analyzed to determine which locations on the MIA 130 have neurons that fluoresced at a given time.

[0091] In some embodiments, the MIA 130 includes an optical source capable of providing optical impulses to specified locations (e.g., 2D coordinates) in the 2D grid. The optical source may include light emitting elements (e.g., light emitting diodes (LEDs), light bulbs, lasers, etc.) capable of emitting light having one or more specified wavelengths. Accordingly, optogenics may be used to manipulate neural activity'. Additionally, lasers of specific wavelengths may be used for highly accurate targeting of specific neurons. The response to optical stimulation may then beDocket No. 229303-701601 / PCTmeasured by the sensors of the MIA 130. Light stimulation can manipulate specific cells (e.g., neurons) that may express a targeted opsin protein, thereby making it possible to investigate the role of a subpopulation of neurons in a neural circuit. Accordingly, in some embodiments the MIA 130 provides optical stimulation to specified 2D coordinates and measures electrical signals generated by neurons of the SBC(s) 135 in response.

[0092] In some embodiments, the MIA 130 provides electrical stimulation to specified 2D coordinates in the 2D grid, and optical signals are measured. MIA 130 may include one or more optical sensors capable of optically detecting excitation (e.g., electrical excitation) of neurons and generating signals (e.g., optical signals) based on such detected excitation of the neurons. Accordingly, optogenics may be used to detect neural activity. The optical sensors may include charge coupled devices (CCDs), complementary metal oxide (CMOS) devices, and / or other types of optical sensors.

[0093] Mechanisms for optically detecting neural activity are discussed in greater detail below. In some embodiments, immunofluorescence of specifically modified calcium that get cleaved and activated when they enter the neurons can be paired with one or more image sensors to image activation of neurons. In some embodiments genetically encoded voltage detectors may be introduced into cells at a given point and used to detect activation of neurons when stimulated with light. In some embodiments luciferase based reactions may be introduced into the cells and paired with another method of detecting voltage changes in neurons to detect changes in voltage without the need for external light stimulation.

[0094] In some embodiments, the interface 105 includes a fully optical system instead of an MIA 130. In such an embodiment, the interface 105 may include an optical source capable of providing optical impulses to specified 2D coordinates in a 2D grid. The optical source may include light emitting elements (e.g., light emitting diodes (LEDs), light bulbs, lasers, etc.) capable of emitting light having one or more specified wavelengths. Additionally, lasers of specific wavelengths may be used for highly accurate targeting of specific neurons. Additionally, the substrate and / or other components may include one or more optical sensors capable of optically detecting electrical excitation of neurons and generating optical signals based on such detected electrical excitation of the neurons. Accordingly, optogenics may be used to manipulate and detect neural activity.Docket No. 229303-701601 / PCT

[0095] Biological neurons can be designed to fluoresce, generate a current, generate a voltage, release a chemical compound, etc. via various mechanisms. In some embodiments, excitation of the biological neurons stimulates changes in cell membrane characteristics, which can cause them to fluoresce, generate a current, generate a voltage, release a chemical compound, and so on. In some embodiments, ion channels, proteins, intramembrane structures, extra membrane structures, and / or transmembrane structures generate a current, voltage, light and / or a chemical compound responsive to stimulation of a neuron. In some embodiments, channels (e.g., ion channels) are opened and / or closed (e.g., responsive to exposure to light, to a voltage, to a current, to a chemical compound, etc.) in cell membranes to generate a current. In some embodiments, neurons may be designed to directly generate a voltage (e.g., via a protein).

[0096] In some embodiments, biological neurons create ion currents through their membranes when excited, causing a change in voltage between the inside and the outside of the cell. When sensing, the electrodes on an MEA transduce the change in voltage from the environment carried by ions into currents carried by electrons (electronic currents). When stimulating, electrodes may transduce electronic currents into ionic currents through the MEA. These ionic currents can trigger the voltage-gated ion channels on the membranes of the excitable neurons, causing the neuron to depolarize and trigger an action potential.

[0097] In some embodiments, neurons express a reporter (e.g., a gene reporter) responsive to stimulation. The expressed reporter may cause the neurons to fluoresce at a certain wavelength and / or to release a chemical compound. For example, neurons may be designed to have a fluorescent protein that fluoresces when stimulated. In another example, neurons may be designed to cleave to release another protein or chemical when stimulated (e.g., when stimulated via light). In some embodiments, light can be used to target an organelle of a neuron cell. In some embodiments, light can trigger a reaction in, on or through a cell membrane of a neuron cell. In some embodiments, stimulation of a neuron cell can open or close ion channels, activate, inflate, inhibit, or cleave a protein in the neuron cell, and so on.

[0098] The size and shape of a sensed signal may depend upon the nature of the medium (e.g., solution) in which the neuron or neurons are located (e.g. the medium’s electrical conductivity, capacitance, and homogeneity), the nature of contact between the neurons and the electrodes (e.g. area of contact and tightness), the nature of the electrodes (e.g. its geometry, impedance, and noise), the analog signal processing (e.g. the system's gain, bandwidth, and behavior outside ofDocket No. 229303-701601 / PCTcutoff frequencies), and data sampling properties (e.g. sampling rate and digital signal processing). For the sensing of a single neuron that partially covers a planar electrode, the voltage at the contact pad is approximately equal to the voltage of the overlapping region of the neuron and electrode multiplied by the ratio the surface area of the overlapping region to the area of the entire electrode.

[0099] The MIA 130 may be an active MIA that includes a controller 145. The controller 145 may include processing logic (e.g., a general purpose or special purpose processor), a network adapter, a digital to analog converter (DAC), an analog to digital converter (ADC), and / or other components. The network adapter may be a wired network adapter (e.g., an Ethernet network adapter) or a wireless network adapter (e.g., a Wi-Fi network adapter), and may enable the interface 105 to connect to network 120. In some embodiments, the controller 145 includes a processing device, which may be a general purpose processor, a microcontroller, a digital signal processor (DSP), a programmable logic controller (PLC), a microprocessor or programmable logic device such as a field programmable gate array (FPGA) or a complex programmable logic device (CPLD). In some embodiments, controller 145 includes a memory, which may be a non-volatile memory (e.g., RAM) and / or a volatile memory (e.g., ROM, Flash, etc.). In some embodiments, the controller 145 is a system on a chip (SoC) that includes the processing device, memory, network adapter, DAC, and / or ADC,

[0100] In the case of an active MIA 130, on-chip signal multiplexing may be used to provide a large number of electrodes to achieve a high spatio-temporal resolution in sensing of electrical and / or optical signals and providing of electrical impulses (e.g., as with an HD-MEA). Moreover, weak neuronal signals can be conditioned at the electrodes by dedicated circuitry configured to provide a large signal-to-noise ratio. Finally, analog-to-digital conversion may performed on chip, so that stable, digital signals are generated. In some embodiments, the MIA 130 is a passive MIA that is connected to a controller via one or more leads and / or a printed circuit board (PCB).

[0101] The interface 105 can be used to perform electrophysiological experiments on dissociated cell cultures (e.g., cultures of biological neurons). With dissociated neuronal cultures, the neurons spontaneously form biological neural networks (BNNs), which can be trained to perform the function of a MVL circuit or gate. The spontaneous formation of BNNs can be facilitated by using very dense neural cultures, as set forth above. The interface 105 may include a multi-input array (MIA) 130 and the chamber 140 that contains the living cultures of biological neurons configured to function as synthetic biological circuits (SBCs) 135 in a nutrient richDocket No. 229303-701601 / PCTsolution that keeps the biological neurons alive. The MIA 130 may include a planar array (e.g., a two-dimensional (2D) grid) or a three-dimensional (3D) array (e.g., a 3D matrix) of emitters (e.g., electrical, chemical, and / or optical signal emitters) and / or detectors (e.g., electrical, chemical, and / or optical signal detectors). The detectors of the MIA 130 may be used to detect (e.g., sample or measure) signals (e.g., electrical, chemical, and / or optical signals) at 2D coordinates (or 3D coordinates) at high spatial and temporal resolution at excellent signal quality. Additionally, the emitters of the MIA 130 may be used to apply signals (e.g., electrical, chemical, and / or optical signals) at the 2D coordinates or 3D coordinates.

[0102] Along with sensing changes in electrical activity brought about from action potentials, the MIA 130 is capable of stimulating cells at a range of voltages. Providing external electrical stimulation is relatively non-invasive to cells, and effectively elicits action potentials or responses in a comparable manner to internal electrical stimulation. With an appropriate coding scheme, external electrical stimulations are able to convey a range of information. Thus, the MIA 130 is capable of not only reading information from the SBCs, but also writing data into the SBCs.

[0103] In some examples, each of the computing device(s) 110 may include a neural interface responsible for translating between inputs / outputs of the computing device 110 and the inputs / outputs of the interface 105. Since neurons communicate with each other using a shared ‘language’ of electrical activity, links between silicon and biological computer systems can be formed through electrical stimulation. For this reason, electrical stimulation (as well as optical and chemical stimulation) may be used to induce neuronal plasticity in vitro or to provide structured information to the SBCs 135. In some examples, the neural interface performs encoding, which includes arranging information output by processing logic of the computing device 110 into a format that can be delivered to the SBCs 135 and understood by the SBCs 135. In some examples, the neural interface performs decoding, which includes arranging information output by the SBCs 135 into a format that can be delivered to and understood by the processing logic of the computing device 110.

[0104] In some embodiments, the neural interface provides a vectorized bridge that can convert temporal encoding, rate encoding, place encoding, and / or position encoding into vectors and / or tensors (e.g., static sets of values). The neural interface may convert these static sets of values into actual potential spiking of electrodes of the MIA 130, where the potential spiking can be performed according to rate-based coding, place-based coding and / or mixed coding schemes. Accordingly,Docket No. 229303-701601 / PCTthe potential spiking (e.g., stimulus patterns) can be performed according to rate coding and also in terms of a 2D or 3D spatial layout. In this way, the neural interface can provide a biologically compatible mechanism for choosing when and where to provide stimulation (e.g., which electrodes to apply electrical signals to, which light sources to apply signals to for generation of optical signals, which voltages to use, which currents to use, which frequencies to use, and so on). Thus, a vector or tensor may include a set of values that capture voltage levels in time and by electrode, for example.

[0105] In some examples, place-based or position-based encoding schemes may be used, such that a different meaning is associated with stimulation (e.g., electrical, chemical and / or optical stimulation) in different locations on the MIA 130. For example, stimulation m a first region of the MIA 130 may represent the value of a first input to a wetware MVL system, stimulation in a second region of the MIA 130 may represent the value of a second input to the wetware MVL system, and so on. Similarly, for decoding, different regions of the MIA 130 in which neural activity is detected may correspond to different outputs of a wetware MVL system. In a particular example, stimulation in a first region of the MIA 130 may represent the value of a first input to a 3-valued adder or 4-valued logic gate, stimulation in a second region of the MIA 130 may represent the value of a second input to the 3 -valued adder or 4-valued logic gate, and neural activity detected by a third region of the MIA 130 may represent the output of the 3-valued adder or 4-valued logic gate. In another particular example, stimulation in a first region of the MIA 130 may represent an instruction (e.g., an opcode) to be performed by a wetware ALU of the biological computer 100, stimulation in one or more second regions of the MIA 130 may represent one or more operands of the instruction, and neural activity detected by a third region of the MIA 130 may represent a result of the performing the indicated instruction with the indicated operands.

[0106] In some embodiments, the neural interface uses a rate-based encoding scheme. For a rate-based encoding scheme, the frequency of MIA stimulation or detected neuron activation may indicate a numeric value (e.g., a value of an input to a MVL circuit or gate, an opcode representing an instruction, a value of an operand of an instruction, a value of a result of an instruction, etc.) For example, a first frequency of MIA stimulation or detected neuron activation may indicate a first numeric value, and a second frequency of MIA stimulation or detected neuron activation may indicate a second numeric value.Docket No. 229303-701601 / PCT

[0107] In some embodiments, a mixed encoding scheme may be used, in which some information is conveyed based on position of electrical, chemical and / or optical signals provided as stimulus to the SBCs 135 by the MIA 130 or detected as activity of the SBCs 135 by the MIA 130, and other information is conveyed based on frequency of electrical, chemical and / or optical signals provided as stimulus to the SBCs 135 by the MIA 130 or detected as activity of the SBCs 135 by the MIA 130. Coding and / or decoding schemes may also be at least in part based on voltage levels and / or current levels. Accordingly, a mixed coding scheme may convey information based on position, rate, voltage and / or current for encoding and / or decoding of information,

[0108] Returning to FIG. 1A, the computing device(s) 110 may provide one or more application programming interfaces (APIs) that enable users to upload a program to be executed by the biological computer 100, encode the program (e.g., using an encoding scheme as described herein) and upload the encoded program to the biological computer via the MIA 130. In some examples, uploading information (e.g., an encoded program or encoded data) to the biological computer via the MIA 130 involves the computing device 110 generating a digital input signal and the neural interface converting the digital input signal into instructions for one or more electrical, chemical and / or optical impulses according to an encoding scheme, where each electrical, chemical and / or optical impulse instruction is associated with a 2D coordinate or a 3D coordinate in the MIA. Each electrical, chemical and / or optical impulse instruction may further include information on an amplitude or intensity of the impulse to apply, a frequency or wavelength of the impulse to apply, timing of when to apply the electrical, chemical and / or optical impulse and / or a current of the impulse to apply. Accordingly, the information for each impulse may be a tuple that includes coordinates (e.g., 2D or 3D coordinates), intensity / amplitude values, frequency / wavelength values, a current amplitude value, or other coded information.

[0109] After the neural interface converts the digital input signal into information for one or more optical, chemical and / or electrical impulses (referred to herein as encoding), it may send the information to the MIA 130. As discussed above, such encoding may be performed according to an encoding scheme, which may be a place-based encoding scheme, a rate-based encoding scheme, a hybrid encoding scheme, or another type of encoding scheme. An interface controller 145 of the interface 105 may convert the information into one or more analog signals for the optical or electrical impulses (e.g., using a digital-to-analog converter or “DAC”) The MIA 130 may apply the one or more analog signals to appropriate electrodes 130 (or light emitting elements orDocket No. 229303-701601 / PCTchemical emitters) to apply the optical, chemical and / or electrical impulses at the specified coordinates and / or with the specified intensity / amplitude, frequency / wavelength, and so on.

[0110] In some embodiments, one or more cameras are used to measure neuron activation. In some embodiments, neurons may be modified to fluoresce when they fire, and the fluorescence may be captured by image sensors (e.g., cameras).

[0111] In some embodiments, genetically encoded voltage indicators (GEVI) are used. GEVIs are fluorescent protein reporters of membrane potential. A GEVI is a protein that can sense membrane potential in a cell and relate the change in voltage to a form of output. In some embodiments, the protein generates the output by fluorescing. A GEVI can have many configuration designs in order to realize a voltage sensing function. In some embodiments, the GEVI is on or in the cell membrane. The GEVI senses a voltage difference as part of a voltagesensitive domain (VSD)-based sensor to report the voltage difference by a change in fluorescence. In another embodiment the GEVI can be a rhodopsin (a G-protein coupled receptor found in rod cells in the retina) based sensor. In another embodiment the GEVI could be a rhodopsin-fluorescence resonance energy transfer (FRET) sensor.

[0112] In some embodiments, a Bioluminescence resonance energy transfer (BRET) is used. In some embodiments, BRET is based on the energy derived from a luciferase reaction that can be used to excite a fluorescent protein if the fluorescent protein is near the luciferase enzyme. A BRET includes a fusion of donor (luciferase) and acceptor (fluorescent) molecules to proteins of interest. Energy is transferred through non-radiative dipole-dipole coupling from the donor to the acceptor when in proximity, resulting in fluorescence emission at a specific wavelength. The energy emitted by the acceptor relative to that emitted by the donor is termed the BRET signal. It is dependent upon the spectral properties, ratio, distance and relative orientation of the donor and acceptor molecules as well as the strength and stability of the interaction between the proteins of interest.

[0113] In some embodiments, GEVIs are expressed alongside Bioluminescence resonance energy transfer (BRET) techniques to enable emission (e.g., a luciferase based emission) of light when voltage changes occur in a cell. This technique enables observation of firing cells without additional fluorescence imaging. In some embodiments, the in vitro biological neurons in the SBCs 135 each comprise a genetically encoded voltage indicator (GEVI) and a bioluminescence resonance energy transfer (BRET) that fluoresce to generate the signals. In some embodiments, through genetic and / or protein engineering either a rhodopsin based voltage sensitive unit orDocket No. 229303-701601 / PCTanother VSD acts to cleave or transfer a bound luciferase from a unit with an embedded fluorophore when voltage changes, thereby emitting a wavelength of light that marks a voltage change - such as an action potential - inside or around a cell. In another embodiment the change in membrane voltage influences the local electrochemical potential triggering the release of energy bound m the luciferase and thereby exciting the fluorophore.

[0114] In some embodiments, the cameras can detect the fluorescence and determine a location from which the fluorescence originated. Alternatively, the interface 105 or computing device 110 can receive an image from the camera and determine a location from which the fluorescence originated. In some examples, the interface 105 or computing device 110 can generate a digital representation of coordinates of any locations at which light was detected (e.g., locations that exhibited immunofluorescence).

[0115] In some examples, the SBCs 135 use quantum phenomena to perform computational tasks. The use of quantum phenomena by SBCs to perform computational tasks may be referred to herein as “biologic quantum computing” or simply “quantum computing.” FIGS. 1E-1H illustrate the principles of biologic quantum computing.

[0116] FIG. 1E illustrates an example of quantum ion superposition for synaptic signaling. At the chemical and electrical synaptic junctions, ions such as Ca2+and Na+exist within an environment of extreme nanometer confinement and ultra-short timescales (e.g., femtoseconds). Under these conditions, the ions transiently occupy quantum superposition states of position or momentum before decoherence collapses them into a definite state during ion channel transit or synaptic fusion. However, ion superposition in neurons is thought to be unobservable using existing technology due to rapid decoherence m warm, wet biological systems.

[0117] Referring to FIG. IF, quantum (or quantum-like) dynamics of neuronal arrays can enable rapid, parallel processing of computational tasks. As one of ordinary skill in the art of quantum computing will appreciate, principles of quantum computing include superposition (particles exist in multiple states simultaneously, enabling parallel computation), entanglement (quantum states of particles become linked, allowing instantaneous correlations), and interference (quantum wavefunctions combine to amplify correct outcomes and cancel errors). At atomic scales, ionic signal may briefly exhibit quantum behaviors, such as transient superposition or entanglement, enhancing the precision and speed of neural processing, echoing quantum computing principles. These fleeting quantum effects may enhance signaling efficiency,Docket No. 229303-701601 / PCTunderpinning the rapid and complex information processing capabilities of biologic quantum computing systems (e.g., biological computer systems 100, synthetic biological circuits 135, etc.) - an emergent analogue to quantum computation.

[0118] Referring to FIGS. 1G and 1H, arrays of neurons (in the brain or in SBCs) may contain Posner clusters, which may contain six phosphorous atoms whose nuclear spin states can be quantum entangled, influencing how neurons process and store information. When phosphorous atoms in Posner clusters are entangled, changing the spin state of one entangled phosphorous atom causes the state of its entangled partner to change as well, regardless of how far apart they are. Thus, entangled Posner clusters involved in chemical signaling in one neuron may induce similar reactions in another neuron. For additional information regarding biologic quantum computing, see [4] -[9],

[0119] The performance of biologic quantum computing by synthetic biologic circuits (SBCs) has significant implications. For example, quantum computing is useful for computational tasks involved in neural network training and inference (e.g., large matrix multiplication). Relative to conventional quantum computing, biologic quantum computing can be equally fast but far more energy efficient, and operating biologic quantum computers can be much more straightforward then programming conventional quantum computers because the quantum processes used by biologic quantum computers are managed by the neurons themselves. In addition, the use of multi¬ valued logic as described herein is highly compatible with biologic quantum computing because quantum computing relies on multivalued logic.

[0120] In some examples, an SBC may include one or more biologic qubits. A biologic qubit may include a confluent array of neurons coupled to an electrode array. In some examples, the biologic qubit functions as a configurable logic block of a biological computer system. The energy intensity of biologic qubits may be lower (e.g., much lower) than the energy intensity of conventional qubits.

[0121] Some examples have been described in which a biological computer 100 includes an interface controller 145 and interface 105, which includes an MIA and one or more SBCs 135. In some examples, components of the biological computer 100 can be scaled in parallel. For example, the biological computer 100 can include multiple interfaces 105 in communication with an interface controller 145, multiple interface controllers 145 in communication with multiple sets of interfaces 105, etc.Docket No. 229303-701601 / PCT

[0122] Referring to FIG. IB, a method 150 for performing a computational task with one or more synthetic biologic circuits (SBCs) may include one or more of steps 152-160. In step 152, one or more first signals encoding data may be provided by a multi-input array 130 (e.g., a multielectrode array) of a biological computer system 100 to one or more synthetic biological circuits 135. In some examples, the first signals encoding the data include base-N electrical signals, where N > 2. Additionally or alternatively, the first signals encoding the data can include base-2 electrical signals.

[0123] In step 154, the SBCs may perform a computational task based on the data encoded in the first signals. In some examples, performing the computational task includes sensing the first signals and decoding the data encoded in the first signals. In some examples, performing the computational task includes tokenizing the data using neural maps.

[0124] The computational task performed by the SBCs may be any suitable computational task and may include any set of operations, calculations, processing steps, etc. In some examples, the computational task is an ill-posed inverse task. Inverse problems can involve determining the underlying causes or parameters of a system or process from its observed effects or outputs. An inverse task is considered ill-posed when it violates one or more of the conditions of Hadamard’s definition of a well-posed problem: (i) a solution exists, (li) the solution is unique, and (iii) the solution depends continuously on the input data. Ill-posed inverse tasks often exhibit instability, meaning small perturbations in input data can lead to large variations in the solution. They may also suffer from non-uniqueness, where multiple candidate solutions fit the observed data equally well, or from incomplete or noisy measurements that make exact reconstruction impossible without additional constraints or regularization.

[0125] Examples of ill-posed inverse tasks may include image reconstruction in medical imaging (e.g., computed tomography or MRI), where limited or noisy measurements are used to infer high-resolution images; deconvolution in signal processing, where the original signal is recovered from blurred or degraded observations; and parameter estimation in complex biological systems, where sparse experimental data is used to infer kinetic constants or network topology. These tasks may use specialized algorithms, such as regularization techniques, Bayesian inference, or machine learning approaches, to stabilize solutions and incorporate prior knowledge. More generally, examples of ill-posed tasks may include image reconstruction, image processing, inverse scattering, matrix inversion, deconvolution, etc.Docket No. 229303-701601 / PCT

[0126] In some examples, the computational task performed by the SBCs includes deconvolution. Deconvolution is a computational process used to reverse the effects of convolution on observed data, thereby recovering an estimate of the original signal or structure that produced the measurements. Convolution often occurs when a system blurs, mixes, or distorts input signals during acquisition, such as through optical, acoustic, or biochemical processes. Deconvolution seeks to separate these overlapping contributions and restore resolution by mathematically modeling the system’s transfer function or point-spread function. This technique is widely applied to various data types, including images (e.g., removing blur in microscopy or astronomical imaging), time-senes signals (e.g., recovering original waveforms in audio or seismic data), and spatial or spectral measurements (e.g., disentangling overlapping peaks in spectroscopy or gene expression profiles).

[0127] In some examples, the SBCs perform at least some portions of the computation task using biologic quantum computing. In some examples, the SBCs include one or more biologic qubits.

[0128] Referring again to FIG. IB, in step 156, the SBCs may provide one or more second signals encoding a result of the computational task (e.g., the data produced by performing the computational task). These second signals may be provided by the firing of neurons of the SBCs, or by other neuronal communication mechanisms. In step 158, the MIA may sense the second signals. In step 160, the interface controller may decode the data encoded in the second signals sensed by the MEA.

[0129] In some examples, the SBCs 135 may be trained to perform specific functions (e.g., the functions of MVL circuits or gates). Training an SBC 135 may involve providing input (e.g, stimulus) to the SBC 135 representing instructions and / or data, detecting a response of the SBC, and determining whether the response of the SBC matches an expected response. If the response of the SBC 135 matches the expected response, the computing device 110 may use an API of the neural interface to send a positive reinforcement training signals to SBC 135. Alternatively, in some embodiments no positive reinforcement training signal is generated or sent to the SBC. Also, if the response of the SBC fails to match the expected response, the computing device 110 may use the API of the neural interface to send a negative reinforcement training signal to the SBC 135. Alternatively, in some embodiments no negative reinforcement training signal is generated or sent. Instead, all inputs to the SBC 135 may be paused for a brief time period if the SBC responses failDocket No. 229303-701601 / PCTto match the expected response. In some embodiments, positive reinforcement signals are used, but negative reinforcement signals are not used. In some embodiments, both positive and negative reinforcement signals are used. In some embodiments, negative reinforcement signals but not positive reinforcement signals are used.

[0130] In some embodiments, an SBC 135 is trained without using any reward or punishment stimulus. There may be a steady or periodic stream of signals to the SBC 135 during standard operation. Each set of signals may include analog signals delivered to appropriate electrodes (or light emitting elements or chemical emitters) of the MIA 130 to apply the optical, chemical and / or electrical impulses at specified coordinates and / or with specified intensity / amplitude, frequency / wavelength, and so on. For each set of signals, the SBC 135 may generate responses (e.g., by generating electrical impulses / signals, fluorescing, emitting chemical compounds, etc,). If the signals generated by the SBC 135 represent a target value (or a value within a target range), the system may continue to provide the stream of signals to the SBC 135. However, if the signals generated by the SBC 135 do not represent the target value, the stream of signals may be paused for a period (e.g., 1-5 seconds), thus depriving the SBC 135 of any stimulus. Accordingly, the system may cease to deliver a stimulus to the vitro biological neurons of the SBC 135 for a time period in response to the SBC’s output signals failing to satisfy criteria. Ceasing delivery of the stimulus may elicit self-organizing behavior of the in vitro biological neurons. Experimentation has shown that neurons effectively desire a stimulus, and will operate in a manner to increase the chance of receiving a stimulus. Accordingly, neurons can be trained to perform tasks by depriving the neurons of stimuli when they fail to act as desired.

[0131] Referring to FIG. 1C, a method 170 for training one or more synthetic biologic circuit (SBCs) to perform a computational task may include steps 172-182. In some examples, the training method is initiated and / or controlled by a computing device 110 via an interface controller 145 of a biological computer system 100.

[0132] In step 172, the interface controller measures, via the MIA, first neuronal activity of the neurons of the SBC(s). These measurements may be obtained while the neurons are not being stimulated by MIA, or while the only stimulation applied by the MIA is a baseline level of stimulation that the MIA continually applies to the neurons. Thus, the first neuronal activity may be a baseline level of neuronal activity. Some examples of this phase of the training process may be referred to herein as an “ambient recording phase.”Docket No. 229303-701601 / PCT

[0133] In step 174, the interface controller stimulates, via the MIA, the neurons. Such stimulation may include providing, by the MIA, one or more input signals encoding input data for the computational task. Some examples of this phase of the training process may be referred to herein as a “sensory stimulation phase.”

[0134] In step 176, the interface controller measures, via the MIA, second neuronal activity of the neurons of the SBCs. These measurements may be obtained while the neurons are being stimulated by MIA (e.g., in parallel with step 174) or shortly after the neurons have been stimulated. Some examples of this phase of the training process may be referred to herein as a “monitoring phase.”

[0135] In step 178, the interface controller 145 or the computing device 110 determines, based on the second neuronal activity, whether the SBC(s) have performed the computational task satisfactorily. In some examples, satisfactory performance is achieved if the accuracy of the data produced by the SBC(s) exceeds a threshold accuracy level. If satisfactory performance is achieved, in step 180, the interface controller may provide a first type of stimulus to the neurons via the MIA or provide no stimulus to the neurons via the MIA. Alternatively, if satisfactory performance is not achieved, in step 182, the interface controller may provide a second type of stimulus to the neurons via the MIA. Some examples of this phase of the training process may be referred to herein as a “feedback stimulation phase.”

[0136] In some examples, the first type of stimulus is deterministic, predictable, and / or non-punitive. The first type of stimulus can be used to indicate positive feedback. Signal 184 shown in FIG. 1D is a non-limiting example of the first type of stimulus. In the example of FIG. 1D, the signal 184 has voltage spikes of 100 mV at a frequency of 100 Hz for a duration of 100 ms, such that the total number of spikes is 10.

[0137] In some examples, the second type of stimulus is stochastic, unpredictable, and / or punitive. The second type of stimulus can be used to indicate negative feedback. Signal 186 shown in FIG. 1D is a non-limiting example of the second type of stimulus. In the example of FIG. 1D, the signal 186 has larger voltage spikes of 150 mV at a lower frequency of 5 Hz for a longer duration of 4 s, such that the total number of spikes is greater (e.g., more than 10).

[0138] The steps of the method 170 can be repeated any suitable number of times to train the SBC(s) on a training data set of any suitable size.Docket No. 229303-701601 / PCT

[0139] Some examples of multi-electrode arrays and multi-input arrays have been described. Multi-electrode arrays and multi-input arrays are examples of field-programmable signal transceivers (FPSTs). In some examples, a biological computer system 100 can include any suitable type of FPST; some embodiments are not limited to using MEAs or MIAs. In some examples, a FPST includes a set (e.g., grid or array) of emitters (e.g., electrical, chemical, and / or optical signal emitters) and / or detectors (e.g., electrical, chemical, and / or optical signal detectors). The detectors may be used to detect (e.g., sample or measure) signals (e.g., electrical, chemical, and / or optical signals) emitted by the SBCs. Additionally, the emitters may be used to apply signals (e.g., electrical, chemical, and / or optical signals) to stimulate the SBCs.(2) Cells for Biological Circuitry

[0140] The synthetic biological circuits provided herein can be engineered from any cell type that can, for example, (1) form junctions between cells (e.g., gap junctions or tight junctions); (2) elicit an electrical signal via an ion channel (e.g., voltage-gated ion channels); and / or (3) can be genetically modified to elicit an electrical signal (e.g., via optogenetic modifications). Nonlimiting examples of cells that can transmit electrical signals between cells include neurons, astrocytes, glial cells, cardiomyocytes, skeletal muscle cells, smooth muscle cells, vascular smooth muscle cells, kidney cells, leukocytes, lymphocytes, pancreatic cells (e.g, beta cells), hair cells of the ear, photoreceptor cells, bacterial cells, yeast cells, fungal cells, plant cells, insect cells, mammalian cells, human cells, stem cells, in vztro-differentiated cells (e.g., iPSC derived neurons) or any combinations thereof. The SBC can comprise one cell type or multiple different cell types (e.g, neurons and astrocytes). The cells can be primary cells obtained from a subject, commercially produced cells, or induced-pluripotent stem cell derived (in vztro-differentiated cells). The cells can be from a healthy subject or a subject with a disease. In some embodiments, the SBC comprises neurons and astrocytes. Astrocytes can regulate the extracellular environment around neurons, including the balance of ions and neurotransmitters, providing metabolic support, and actively participating in synaptic plasticity, essentially acting as a support system that ensures optimal neuronal activity and communication within the network of neurons in culture. Astrocytes can remove neurotransmitters like glutamate from the synaptic cleft, preventing excessive neuronal excitation, and can also release gliotransmitters which can modulate neuronal activity. Astrocytes assist neurons by maintaining proper ion balance in the extracellular space by taking up excessDocket No. 229303-701601 / PCTpotassium ions. Moreover, astrocytes can phagocytose synapses, alter neurotrophin secretion, and clear debris in the SBCs provided herein.

[0141] The cells provided herein are seeded on the devices provided herein, for example, a microelectrode array or a patch clamp apparatus, to form a network of cells that produce an electrical signal, called a field potential. Methods of seeding cells on a solid support or a membrane are known and further provided in FIG. 3.

[0142] Generally, on the single cell level, an action potential occurs when the membrane potential of a specific cell rapidly rises and falls. This depolarization then causes adjacent locations or cells to similarly depolarize. Action potentials occur naturally in several types of excitable cells, which include animal cells like neurons and muscle cells, as well as some plant cells. Certain endocrine cells such as pancreatic beta cells, and certain cells of the anterior pituitary gland are also excitable cells. Notably, in neurons, action potentials play a role in cell-cell communication by providing for — or with regard to saltatory conduction, assisting — the propagation of signals along the neuron's axon toward synaptic boutons situated at the ends of an axon; these signals can then connect with other neurons at synapses, or to motor cells or glands. In other types of cells, their primary function is to activate intracellular processes. In muscle cells, for example, an action potential is the first step in the chain of events leading to contraction. In beta cells of the pancreas, they provoke release of insulin. Action potentials in neurons are also known as "nerve impulses" or "spikes", and the temporal sequence of action potentials generated by a neuron is called its "spike train". A neuron that emits an action potential, or nerve impulse, is often said to "fire". The rate of firing is a molecular signal that can be measured and used by the systems provided herein.

[0143] In excitable cells, such as neurons or cardiomyocytes, action potentials are generated by voltage-gated ion channels embedded in a cell's plasma membrane. These channels are shut when the membrane potential is near the (negative) resting potential of the cell, but they rapidly begin to open if the membrane potential increases to a precisely defined threshold voltage, depolarizing the transmembrane potential. When the channels open, they allow an inward flow of sodium and / or calcium ions, which changes the electrochemical gradient, which in turn produces a further rise in the membrane potential towards zero. This then causes more channels to open, producing a greater electric current across the cell membrane and so on. The process proceeds explosively until all of the available ion channels are open, resulting in a large upswing in the membrane potential. The rapid influx of sodium ions causes the polarity of the plasma membrane to reverse, and the ionDocket No. 229303-701601 / PCTchannels then rapidly inactivate. As the positive ion channels (e.g., sodium channels) close, the positive ions can no longer enter the neuron, and they are then actively transported back out of the plasma membrane. Potassium channels are then activated, and there is an outward current of potassium ions, returning the electrochemical gradient to the resting state. After an action potential has occurred, there is a transient negative shift, called the afterhyperpolarization, the electrical signals of one cell can be transferred to other cells via a cell junction, such as a gap junction or a tight junction. Methods of measuring action potentials, electrical potentials, and voltage-gated ion channel activity are known and For multiple cells seeded on a microelectrode array (MEA), electrical potentials are pooled across groups of neurons near the recording electrode as a local field potential. Local field potentials (LFPs) offer another approach to study neural circuits through correlated neural activity. LFP signals reflect the presence of correlations in the activity of many neurons that can be used for the SBCs provided herein.

[0144] The cells provided herein can be manipulated to transmit a specific signal as part of the MIA systems provided herein. For example, the manipulation can be via a chemical, physical, mechanical, genetic, or molecular methods or agents. In some embodiments, the cells provided herein are contacted with an agent that modulates the level or the activity of a cell surface protein, an ion channel, or a neurotransmitter. Non-limiting examples of chemical agents that can be used to modulate a cell electrical signal or neurotransmitter signal provided herein can include a serotonin reuptake inhibitor (SSRI), memantine, N-acetylcysteine (NAC), riluzole, topiramate, lamotrigine, minocycline, baclofen, phenibut, a benzodiazepine, progabide, topiramate, valproic acid, lithium, clonazepam, gabapentin, phenobarbital, carbamazepine, phenytoin, oxcarbazepine, zonisamide, brivaracetam (Briviact), cannabidiol (Epidyolex), cenobamate (Ontozry), clobazam (Frisium), eslicarbazepine acetate (Zebinix), levetiracetam (Keppra), perampanel (Fycompa), pregabalin (Lyrica), sodium valproate (Epilim), tiagabine (Gabitril), aripiprazole, haloperidol, risperidone, clozapine, cariprazine, chlorpromazine, fluphenazine, olanzapine, prochlorperazine, quetiapine, loxapine, promazine, thioridazine, trifluoperazine, perphenazine, flupentixol, pimozide, mesoridazin, paliperidone, lidocaine, tetrodotoxin, tricyclic antidepressants (TCAs), anticonvulsants, cocaine, disopyramide, ranolazine, amlodipine, diltiazem, felodipine, isradipine, nicardipine, nifedipine, nisoldipine, verapamil, dofetalide, amiodarone, repaflinide, amifampridine, dalfampridine, tetraethylammonium, linopirdine, nateglinide, niflumic acid, stilbene disulphonate derivatives, diphenylamine-2-carboxylate derivatives, triphenyl-Docket No. 229303-701601 / PCTnonsteroidal anti-estrogens, pyrethroids, NS004, xanthine derivatives, omeprazole, tamoxifen. The agents provided herein can be used to modulate the biologic signals and downstream computing functions provided herein.(3) Genetic Modifications of the Biological Circuit

[0145] The cells of the synthetic biologic circuits provided herein can also be modified to express, reduce, or eliminate the level or activity of a cellular protein by genomic or molecular modifications. The modification can be made to modulate the electrical field potentials produced by the SBCs provided herein (c.g., increase amplitude and communication across a network of cells relative to cells that are not modified). The modifications can be made to modulate the processing speed and / or efficiency of the systems provided herein.

[0146] Genetic modification of a cell provided herein (e.g., knocking-in transgenes or knocking-out undesirable genes) can be achieved by any known genetic engineering techniques, for instance, but not restricted to endonucleases, including but are not limited to zinc-finger nucleases (ZFN), transcription activator-like effector nucleases (TALENs), and CRISPR-Cas nuclease (Clustered Regular Interspaced Short Palindromic Repeats) genetic engineering.CRISPR-Cas genomic modifications

[0147] The methods of making genetically engineered cells described herein can take advantage of a CRISPR system, including but not limited to knockout of glutamate in a neuron.

[0148] There are at least five types of CRISPR systems which all incorporate RNAs and Cas proteins. Types I, III, and IV assemble a multi-Cas protein complex that is capable of cleaving nucleic acids that are complementary to the crRNA. Types I and 111 both require pre-crRNA processing prior to assembling the processed crRN A into the multi-Cas protein complex. Types II and V CRISPR systems comprise a single Cas protein complexed with at least one guiding RNA.

[0149] The general mechanism and recent advances of CRISPR system are discussed in Cong, L. et al, " Multiplex genome engineering using CRISPR systems," Science, 339(6121): 819-823 (2013); Fu, Y. et al., " High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells," Nature Biotechnology, 31, 822-826 (2013); Chu, VT et al. " Increasing the efficiency of homology-directed repair for CRISPR-Cas9-induced precise gene editing in mammalian cells," Nature Biotechnology 33, 543-548 (2015); Shmakov, S. et al, " Discovery andDocket No. 229303-701601 / PCTfunctional characterization of diverse Class 2 CRISPR-Cas systems," Molecular Cell, 60, 1-13 (2015); Makarova, KS et al, " An updated evolutionary classification of CRISPR-Cas systems,", Nature Reviews Microbiology, 13, 1-15 (2015). Site-specific cleavage of a target DNA occurs at locations determined by both 1) base- pairing complementarity between the guide RNA and the target DNA (also called a protospacer) and 2) a short motif in the target DNA referred to as the protospacer adjacent motif (PAM). For example, an engineered cell can be generated using a CRISPR system, e.g., a type II CRISPR system. A Cas enzyme used in the methods disclosed herein can be Cas9, which catalyzes DNA cleavage. Enzymatic action by Cas9 derived from Streptococcus pyogenes or any closely related Cas9 can generate double stranded breaks at target site sequences which hybridize to 20 nucleotides of a guide sequence and that have a protospacer-adjacent motif (PAM) following the 20 nucleotides of the target sequence.

[0150] A CRISPR system can be introduced to a cell or to a population of cells using any means. In some embodiments, a CRISPR system may be introduced by electroporation or nucleofection. Electroporation can be performed for example, using the Neon® Transfection System (ThermoFisher Scientific) or the AMAXA® Nucleofector (AMAXA® Biosystems). Electroporation parameters may be adjusted to optimize transfection efficiency and / or cell viability. Electroporation devices can have multiple electrical wave form pulse settings such as exponential decay, time constant and square wave. Every cell type has a unique optimal Field Strength (E) that is dependent on the pulse parameters applied (e.g., voltage, capacitance and resistance). Application of optimal field strength causes electropermeabilization through induction of transmembrane voltage, which allows nucleic acids to pass through the cell membrane. In some embodiments, the electroporation pulse voltage, the electroporation pulse width, number of pulses, cell density, and tip type may be adjusted to optimize transfection efficiency and / or cell viability.Cas protein

[0151] A vector can be operably linked to an enzyme-coding sequence encoding a CRISPR enzyme, such as a Cas protein (CRISPR- associated protein). In some embodiments, a nuclease or a polypeptide encoding a nuclease is from a CRISPR system (e.g., CRISPR enzyme). In some embodiments, the CRISPR enzyme directs cleavage of one or both strands at a target sequence. In some embodiments, the CRISPR enzyme mediates cleavage of both strands at a target DNA sequence (e.g., creates a double strand break in a target DNA sequence).Docket No. 229303-701601 / PCT

[0152] Non-limiting examples of Cas proteins can include Casl, CaslB, Cas2, Cas3, Cas4, Cas5, Cas6, Cas7, Cas8, Cas9 (also known as Csnl or Csxl2), CaslO, Csyl, Csy2, Csy3, Csel, Cse2, Cscl, Csc2, Csa5, Csn2, Csm2, Csm3, Csm4, Csm5, Csm6, Cmrl, Cmr3, Cmr4, Cmr5, Cmr6, Csbl, Csb2, Csb3, Csxl7, Csxl4, CsxlO, Csxl6, CsaX, Csx3, Csxl, CsxlS, Csfl, Csf2, CsO, Csf4, Cpfl (Cas 12a), c2cl, c2c3, Cas9HiFi, homologues thereof, or modified versions thereof. In some embodiments, a catalytically dead Cas protein can be used (e.g., catalytically dead Cas9 (dCas9)) or a nickase. An unmodified CRISPR enzyme can have DNA cleavage activity, such as Cas9. In some embodiments, a nuclease is Cas9. In some embodiments, a polypeptide encodes Cas9. In some embodiments, a nuclease or a polypeptide encoding a nuclease is catalytically dead. In some embodiments, a nuclease is a catalytically dead Cas9 (dCas9). In some embodiments, a polypeptide encodes a catalytically dead Cas9 (dCas9). A Cas protein can be a high fidelity Cas protein such as Cas9HiFi.

[0153] While S. pyogenes Cas9 (SpCas9) is commonly used as a CRISPR endonuclease for genome engineering, it may not be the best endonuclease for every target excision site. For example, the PAM sequence for SpCas9 (5' NGG 3') is abundant throughout the human genome, but an NGG sequence may not be positioned correctly to target a desired gene for modification. In some embodiments, a different endonuclease may be used to target certain genomic targets. In some embodiments, synthetic SpCas9-derived variants with non-NGG PAM sequences may be used. Additionally, other Cas9 orthologues from various species have been identified and these "non-SpCas9s" bind a variety of PAM sequences that can also be useful for the present disclosure. For example, the relatively large size of SpCas9 (approximately 4kb coding sequence) means that plasmids carrying the SpCas9 cDNA may not be efficiently expressed in a cell. Conversely, the coding sequence for Staphylococcus aureus Cas9 (SaCas9) is approximately 1 kilo base shorter than SpCas9, possibly allowing it to be efficiently expressed in a cell. Similar to SpCas9, the SaCas9 endonuclease is capable of modifying target genes in mammalian cells in vitro and m mice in vivo.

[0154] Alternatives to S. pyogenes Cas9 may include RNA-guided endonucleases from the Cpf 1 family that display cleavage activity in mammalian cells. Unlike Cas9 nucleases, the result of Cpfl -mediated DNA cleavage is a double-strand break with a short 3' overhang. Cpfl's staggered cleavage pattern may open up the possibility of directional gene transfer, analogous to traditional restriction enzyme cloning, which may increase the efficiency of gene editing. Like the Cas9Docket No. 229303-701601 / PCTvariants and orthologues described above, Cpfl may also expand the number of sites that can be targeted by CRISPR to AT-rich regions or AT-rich genomes that lack the NGG PAM sites favored by SpCas9.

[0155] A vector that encodes a CRISPR enzyme comprising one or more nuclear localization sequences (NLSs), such as at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, NLSs can be used. For example, a CRISPR enzyme can comprise at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, NLSs at or near the amino- terminus, at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, NLSs at or near the carboxyl- terminus, or any combination of these (e.g., one or more NLS at the amino-terminus and one or more NLS at the carboxyl terminus). When more than one NLS is present, each can be selected independently of others, such that a single NLS can be present in more than one copy and / or in combination with one or more other NLSs present in one or more copies. The NLS can be located anywhere within the polypeptide chain, e.g., near the N- or C-terminus. For example, the NLS can be within or within about 1, 2, 3, 4, 5, 10, 15, 20, 25, 30, 40, 50 amino acids along a polypeptide chain from the N- or C-terminus. Sometimes the NLS can be within or within about 50 amino acids or more, e.g., 100, 200, 300, 400, 500, 600, 700, 800, 900, or 1000 amino acids from the N- or C-terminus.

[0156] Any functional concentration of Cas protein can be introduced to a cell. For example, 15 micrograms of Cas mRNA can be introduced to a cell. In other cases, a Cas mRNA can be introduced from 0.5 micrograms to 100 micrograms. A Cas mRNA can be introduced from 0.5, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, or 100 micrograms.

[0157] In some embodiments, a dual nickase approach may be used to introduce a double stranded break or a genomic break. Cas proteins can be mutated at known amino acids within either nuclease domains, thereby deleting activity of one nuclease domain and generating a nickase Cas protein capable of generating a single strand break. A nickase along with two distinct guide RNAs targeting opposite strands may be utilized to generate a double strand break (DSB) within a target site (often referred to as a "double nick" or "dual nickase" CRISPR system). This approach can increase target specificity because it is unlikely that two off-target nicks will be generated within close enough proximity to cause a DSB.Guiding polynucleic acids (gRNA or gDNA)

[0158] A guiding polynucleic acid (or a guide polynucleic acid) can be DNA (gDNA) or RNA (gRNA). A guiding polynucleic acid can be single stranded or double stranded. In someDocket No. 229303-701601 / PCTembodiments, a guiding polynucleotide can contain regions of single stranded areas and double stranded areas. A guiding polynucleotide can also form secondary structures.

[0159] In some embodiments, the guide nucleic acid is a gRNA. In some embodiments, the gRNA comprises a guide sequence that specifies a target site and guides an RNA / Cas complex to a specified target DNA for cleavage. Site-specific cleavage of a target DNA occurs at locations determined by both 1) base-pairing complementarity between a gRNA and a target DNA (also called a protospacer) and 2) a short motif in a target DNA referred to as a protospacer adjacent motif (PAM). Similarly, a gRNA can be specific for a target DNA and can form a complex with a nuclease to direct its nucleic acid-cleaving activity.

[0160] In some embodiments, the gRNA comprises two RNAs, e.g., CRISPR RNA (crRNA) and transactivating crRNA (tracrRNA), In some embodiments, the gRNA comprises a single-guide RNA (sgRNA) formed by fusion of a portion (e.g., a functional portion) of crRNA and tracrRNA. In some embodiments, the gRNA comprises a dual RNA comprising a crRNA and a tracrRNA. In some embodiments, the gRNA comprises a crRNA and lacks a tracrRNA. In some embodiments, the crRNA hybridizes with a target DNA or protospacer sequence.

[0161] In some embodiments, the gRNA targets a nucleic acid sequence of or of about 20 nucleotides. In some embodiments, the gRNA targets a nucleic acid sequence of or of about 5, 10, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30 or more nucleotides. In some embodiments, the gRNA binds a genomic region from about 1 base pair to about 20 base pairs away from a PAM. In some embodiments, the gRNA binds a genomic region from about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or up to about 20 base pairs away from a PAM. In some embodiments, the gRNA binds a genomic region within about 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 base pairs away from a PAM.

[0162] A guide RNA can also comprise a dsRNA duplex region that forms a secondary structure. For example, a secondary structure formed by a guide RNA can comprise a stem (or hairpin) and a loop. The length of a loop and a stem can vary. For example, a loop can range from about 3 to about 10 nucleotides in length, and a stem can range from about 6 to about 20 base pairs in length. A stem can comprise one or more bulges of 1 to about 10 nucleotides. The overall length of a second region can range from about 16 to about 60 nucleotides in length. For example, a loop can be or can be about 4 nucleotides in length and a stem can be or can be about 12 base pairs. ADocket No. 229303-701601 / PCTdsRNA duplex region can comprise a protein-binding segment that can form a complex with an RNA-binding protein, such as a RNA-guided endonuclease, e.g., Cas protein.

[0163] In some embodiments, a Cas protein, such as a Cas9 protein or any derivative thereof, is pre-complexed with a gRNA to form a ribonucleoprotein (RNP) complex. In some embodiments, the RNP complex is introduced into a cell to mediate editing.

[0164] In some embodiments, the gRNA is modified. The modifications can comprise chemical alterations, synthetic modifications, nucleotide additions, and / or nucleotide subtractions. The modifications can also enhance CRISPR genome engineering. A modification can alter chirality of a gRNA, In some embodiments, chirality may be uniform or stereopure after a modification. In some embodiments, the modification enhances stability of the gRNA.

[0165] In some embodiments, the modification is a chemical modification, A modification can be selected from 5' adenylate, 5' guanosine-triphosphate cap, 5' N7-Methylguanosine-triphosphate cap, 5' triphosphate cap, 3' phosphate, 3' thiophosphate, 5' phosphate, 5' thiophosphate, Cis-Syn thymidine dimer, trimers, Cl2 spacer, C3 spacer, C6 spacer, dSpacer, PC spacer, rSpacer, Spacer 18, Spacer 9, 3'-3' modifications, 5'-5' modifications, abasic, acridine, azobenzene, biotin, biotin BB, biotin TEG, cholesteryl TEG, desthiobiotin TEG, DNP TEG, DNP-X, DOTA, dT-Biotin, dual biotin, PC biotin, psoralen C2, psoralen C6, TINA, 3' DABCYL, black hole quencher 1, black hole quencher 2, DABCYL SE, dT-DABCYL, IRDye QC-1, QSY-21, QSY-35, QSY-7, QSY-9, carboxyl linker, thiol linkers, 2' deoxyribonucleoside analog purine, 2' deoxyribonucleoside analog pyrimidine, ribonucleoside analog, 2'-0-methyl ribonucleoside analog, and sugar modified analogs, wobble / universal bases, fluorescent dye label, 2’ fluoro RNA, 2' O-methyl RNA, methylphosphonate, phosphodiester DNA, phosphodiester RNA, phosphothioate DNA, phosphorothioate RNA, UNA, pseudouridine-5'-triphosphate, and 5-methylcytidine-5'-triphosphate, and any combination thereof.

[0166] In some embodiments, the modification comprise a phosphorothioate internucleotide linkage. In some embodiments, the gRN A comprises from 1 to 10, 1 to 5, or 1-3 phosphorothioate. In some embodiments, the gRNA comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 phosphorothioate linkages. In some embodiments, the gRNA comprises phosphorothioate internucleotide linkages at the N terminus, C terminus, or both N terminus and C terminus. For example, in some embodiments, the gRNA comprises phosphorothioateDocket No. 229303-701601 / PCTinternucleotide linkages between the N terminal 3-5 nucleotides, the C terminal 3-5 nucleotides, or both.

[0167] In some embodiments, the modification is a 2'-O-methyl phosphorothioate addition. In some embodiments, the gRNA comprises 1-10, 1-9, 1-8, 1-7, 1-6, 1-5, 1-4, 1-3, or 1-22'-O-methyl phosphorothioates. In some embodiments, the gRNA comprises from 1 to 10, 1 to 5, or 1-3 2'-O-methyl phosphorothioates. In some embodiments, the gRNA comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 202'-O-methyl phosphorothioates. In some embodiments, the gRNA comprises 2'-O-methyl phosphorothioate internucleotide linkages at the N terminus, C terminus, or both N terminus and C terminus. For example, in some embodiments, the gRNA comprises 2'-O-methyl phosphorothioate internucleotide linkages between the N terminal 3-5 nucleotides, the C terminal 3-5 nucleotides, or both.

[0168] A gRNA can be introduced at any functional concentration. In some embodiments, 0.5 micrograms to 100 micrograms of the gRNA is introduced into a cell. In some embodiments, 0.5, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, or 100 micrograms of the gRNA is introduced into a cell.Other Endonucleases

[0169] Other endonuclease based gene editing systems known in the art can be used to make an engineered cell described herein. For example, zinc finger nuclease systems and TALEN systems.

[0170] ZFNs are targeted nucleases comprising a nuclease fused to a zinc finger DNA binding domain. A "zinc finger DN A binding domain" or " ZFBD" is a polypeptide domain that binds DNA in a sequence-specific manner through one or more zinc fingers. A zinc finger is a domain of about 30 amino acids within the zinc finger binding domain whose structure is stabilized through coordination of a zinc ion. Examples of zinc fingers include, but are not limited to, C₂H₂ zinc fingers, C₃H zinc fingers, and C4 zinc fingers. A "designed" zinc finger domain is a domain not occurring in nature whose design / composition results principally from rational criteria, e.g., application of substitution rules and computerized algorithms for processing information in a database storing information of existing ZFN designs and binding data. A "selected" zinc finger domain is a domain not found in nature whose production results primarily from an empiricalDocket No. 229303-701601 / PCTprocess such as phage display, interaction trap or hybrid selection. The most recognized example of a ZFN in the art is a fusion of the FokI nuclease with a zinc finger DNA binding domain.

[0171] A TALEN is a targeted nuclease comprising a nuclease fused to a TAL effector DNA binding domain. A "transcription activator-like effector DNA binding domain", " TAL effector DNA binding domain", or " TALE DNA binding domain" is a polypeptide domain of TAL effector proteins that is responsible for binding of the TAL effector protein to DNA. TAL effector proteins are secreted by plant pathogens of the genus Xanthomonas during infection. These proteins enter the nucleus of the plant cell, bind effector-specific DNA sequences via their DNA binding domain, and activate gene transcription at these sequences via their transactivation domains. TAL effector DNA binding domain specificity depends on an effector-variable number of imperfect 34 amino acid repeats, which comprise polymorphisms at select repeat positions called repeat variable-di-residues (RVD). The most recognized example of a TALEN in the art is a fusion polypeptide of the FokI nuclease to a TAL effector DNA binding domain,

[0172] Another example of a targeted nuclease that finds use in the methods described herein is a targeted Spoil nuclease, a polypeptide comprising a Spoil polypeptide having nuclease activity fused to a DNA binding domain, e.g., a zinc finger DNA binding domain, a TAL effector DNA binding domain, etc. that has specificity for a DNA sequence of interest. Additional examples of targeted nucleases suitable for the present invention include, but are not limited to Bxbl, phiC31, R4, PhiBTi, and WO / SPBc / TP9Ol-l, whether used individually or in combination.

[0173] Any one of the aforementioned methods comprising genomically editing via use of an endonuclease can result in a genomic disruption. The genomic disruption can be sufficient to result in reduction or elimination of expression of the protein encoded by the gene. In some cases, a genomic disruption can also refer to the incorporation of an exogenous transgene into the cellular genome. In such cases, an exogenous transgene can also be detected. The genomic disruption can be detected in at least about 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or 100% of cells tested. Detection can be performed by evaluating the disruption at the genomic level via sequencing, at the mRNA level, or protein level. Suitable methods include PCR, qPCR, flow cytometry, imaging, ELISA, NGS, and any combination thereof. In some cases, protein expression can be reduced by about 1 fold, 2 fold, 3 fold, 5 fold, 10 fold, 20 fold, 30 fold, 50 fold, 70 fold, 100 fold, 125 fold, 150 fold, 200 fold, 250 fold, 300 fold, 350 fold, 500 fold, or up to about 1000Docket No. 229303-701601 / PCTfold as compared to a comparable method that lacks the use of the gene editing, such as with CRISPRTransgenes

[0174] A transgene polynucleic acid encoding an exogenous protein or polypeptide that is knocked into a cell described herein can be DNA or RNA, single-stranded or double stranded and can be introduced into a cell in linear or circular form. A transgene sequence(s) can be contained within a DNA minicircle, which may be introduced into the cell in circular or linear form. If introduced in linear form, the ends of a transgene sequence can be protected (e.g, from exonucleolytic degradation) by any method. For example, one or more dideoxynucleotide residues can be added to the 3' terminus of a linear molecule and / or self-complementary oligonucleotides can be ligated to one or both ends. Additional methods for protecting exogenous polynucleotides from degradation include, but are not limited to, addition of terminal amino group(s) and the use of modified internucleotide linkages such as, for example, phosphorothioates, phosphorami dates, and O-methyl ribose or deoxyribose residues.

[0175] A transgene can be flanked by recombination arms. In some instances, recombination arms can comprise complementary regions that target a transgene to a desired integration site. A transgene can also be integrated into a genomic region such that the insertion disrupts an endogenous gene. transgene can be integrated by any method, e.g, non-recombination end joining and / or recombination directed repair. A transgene can also be integrated during a recombination event where a double strand break is repaired. A transgene can also be integrated with the use of a homologous recombination enhancer. For example, an enhancer can block non¬ homolog ous end joining so that homology directed repair is performed to repair a double strand break.

[0176] A transgene can be flanked by recombination arms where the degree of homology between the arm and its complementary sequence is sufficient to allow homologous recombination between the two. For example, the degree of homology between the arm and its complementary sequence can be 50% or greater. Two homologous non-identical sequences can be any length and their degree of non- homology can be as small as a single nucleotide (e.g, for correction of a genomic point mutation by targeted homologous recombination) or as large as 10 or moreDocket No. 229303-701601 / PCTkilobases (e.g., for insertion of a gene at a predetermined ectopic site in a chromosome). Two polynucleotides comprising the homologous non-identical sequences need not be the same length.

[0177] A polynucleotide can be introduced into a cell as part of a vector molecule having additional sequences such as, for example, replication origins, promoters and genes encoding antibiotic resistance. Moreover, transgene polynucleotides can be introduced as naked nucleic acid, as nucleic acid complexed with an agent such as a liposome or poloxamer, or can be delivered by viruses (e.g., adenovirus, AAV, herpesvirus, retrovirus, lentivirus and integrase defective lentivirus (IDLV)) A virus that can deliver a transgene can be an AAV virus,

[0178] A transgene is generally inserted so that its expression is driven by the endogenous promoter at the integration site, namely the promoter that drives expression of the endogenous gene into which a transgene is inserted. A transgene may comprise a promoter and / or enhancer, for example a constitutive promoter or an inducible or tissue / cell specific promoter. A minicircle vector can encode a transgene.

[0179] A transgene can be inserted into an endogenous gene such that all, some or none of the endogenous gene is expressed. For example, a transgene as described herein can be inserted into an endogenous locus such that some (N-terminal and / or C -terminal to a transgene) or none of the endogenous sequences are expressed, for example as a fusion with a transgene. In other cases, a transgene (e.g, with or without additional coding sequences such as for the endogenous gene) is integrated into any endogenous locus, for example a safe-harbor locus.

[0180] When endogenous sequences (endogenous or part of a transgene) are expressed with a transgene, the endogenous sequences can be full-length sequences (wild-type or mutant) or partial sequences. The endogenous sequences can be functional. Non-limiting examples of the function of these full length or partial sequences include increasing the serum half-life of the polypeptide expressed by a transgene (e.g, therapeutic gene) and / or acting as a carrier.

[0181] Furthermore, although not required for expression, exogenous sequences may also include transcriptional or translational regulatory sequences, for example, promoters, enhancers, insulators, internal ribosome entry’ sites, sequences encoding 2A peptides and / or polyadenylation signals.Exemplary ModificationsDocket No. 229303-701601 / PCT

[0182] Provided herein are systems comprising a synthetic biological circuit, wherein the synthetic biological circuit comprises a genetically modified population of cells. In some embodiments, the genetically modified population of cells comprise a genomic disruption in one or more genes. In some embodiments, the genetically modified population of cells comprise a transgene encoding for a protein. Non-limiting examples of target genes for genomic disruption or for transgene delivery' that can be used for the modification of the synthetic biological circuit are provided in Table 1, Table 2, or Table 3. In some embodiments, the synthetic biological circuit comprises a population of neurons. In some embodiments, the modified population of neurons have an increased mean firing rate of field potentials (also referred to herein as spike potentials) relative to a population of neurons that have not been modified. In some embodiments, the modified population of neurons have an increased field potential amplitude relative to a population of neurons that have not been modified. In some embodiments, the modified population of neurons have an increased mean field potential amplitude relative to a population of neurons that have not been modified. In some embodiments, the modified population of neurons increase the number of active electrodes in the system relative to a population of neurons that have not been modified.Table 1. Exemplary Targets for Gene Editing.Target(s) Exemplary Gene and Exemplary Target Sequence(s)NCBI Gene ID(s)Glutamate receptors (GRM1) GRM1 Gene ID: 2911 NCBI Reference Sequence: NG 012839.2TTTAGCATAGTTTCTTTAGTGCTTGCTTAAATGAAACATCCTGAAAGAGCAAGTTAAAATGTGT TAAATATCATCATTGGATAGATCCAAAGATGATGGCAAAGAGAATGGCAAAATCTAGATTGAAT GTGATTTATTTAAATATTTAATAATATTTCCCCCATTGAGCTAGATAATTGATTATTAATATTG GAAGTATAAATTCTTTCTTCAGGAAGAAGTAATGCTTCATTATTTTTAAACAGTGGTTTTATTT TCCATATTTCATTATAAGATTGTTTTATTTGGTGAGAGGTGGCTAATAGTGGGTTGCTTTCTGA GGAGTAGAATTATGTTTTTGGAAAATGAATTGAACATGTTTTCACTAATTTTGCTGCGAAAGGA CTAATATTATAATTTTGCCCATGTGTTAGTTTGGATTACTATGTTTCCTTTCTGCCTTCTGATT GGTGTTGTATGGCAAAGACAATGTCTCCGTACTATTATAGTAATTGATTCCAGACTACATTGTG TTCACAGAACAGCATCATAGAAATATGTGGGGTTTATTTTATTACCAAGCAGTTACACCA'i'TCA CCTTGCCATTCATATATAATACCTTGTTTCTTCTCAAAGGGAATTAGTTACTTCTCTTATTAAA CGTAACATCTGTAAAAACTAAATTGCATTTCTTTCCTTCTATCTGGAAAGGAATTTTGTG(SEQ ID NO: 1)NMDA receptors (glutamate GRIN2B Gene ID: 2904 > NG 031854.2:5411 -449676 Homo sapiens glutamate ionotropic receptor NMDA type subunit 2B (GRIN2B), RefSeqGene on chromosome 12ionotropic receptor NMDA typesubunit 2B) TCTGAGCACAGCTTCATCCCTGAGCCCAAAAGCAGTTGTTACAACACCCACGAGAAGAGAATCT ACCAGTCCAATATGCTAAATAGGTGAGTGTGGAACAGAGGGCACATTGATGGTGGAGATGAGAG GGGAGCTTCAAAGACTGGGGTTGGAAGTTCAGAAACTGCAGGAAGAACAGTTCAAGACGGTAGC AGGTTGAAGCCTGGCCTTGGGCATGGCAATTGGTCATGATTGGGGGAGTGTTTTGGATCCAACC TTGCCAAAACAACTTTCATTTATTGAGCACTTATTATGTGTTAGATGTTACTCTAAGGGTTTTG CATGTTTGGTTTCATTTGATCCTCAAAATAACTCTATAAGTCAGGTTTTAACAATTGTTTCTAT TTTACAATACGAAAGAATAGGCTTGGAGGTGTAAATTATACAATGTCATGTTGCAAGTGAGTGG TCAGTTTGGAGTTGAACCTAGGCCTCTCTGATCTGTAAGTACTACTTGATCGCAAAGAAAGATG CAGGAGACTTAGAAATAAGGGAGGACCAGTCCGCTTACCTTTGGAATG (SEQ ID NO: 2 ) GABA receptors (Gamma- GABARAP Gene ID: > NC_000017.ri:c7242449-7240008 Homo sapiens chromosome 17, GRCh38.pl4 Primary aminobutyric acid A receptors 11337 Assembly[GABA(A) receptor) GCAAATTCGTGGATCGCTCCGCTGAATCCGCCCGCGCGTCGCCGCCGTCGTCGCCGCCCCCCGT CCCGGCCCCCCTGGGTTCCCTCAGCCCAGCCCTGTCCAGCCCGGTTCCCGGGAGGATGAAGTTC GTGTACAAAGAAGAGCATCCGTTCGAGAAGCGCCGCTCTGAGGGCGAGAAGATCCGAAAGAAAT ACCCGGACCGGGTGCCGGTGAGGACAGTAGCCAGGGCCGAGGGCGCGGGACGCTTACGCTGACGCCACAGCCTGGCCTTGGATCGGGACAGGCAGCGCAGGGATCAGGCTGCTGGCTGTGGTGGGGTTGGCAGCGAGCCTGAAACTGCTGTCCTTGGGACCCAGGATGAAAGGAGTGGGGTAGTGGTGGATG TGGGGTTGAGAGGGATGTGGACAGGAGAGAGAAACC TTAGCATTTGAGGCCTTAGAGAGTTTGACAGCCTGAGCTTCAAATGGTCATCCCCCCAGCTAGC TAAGGTTGCGAGATTGAATCACTAGTCGACCCAAACAAGTCACCTCCGTCTGATTAAGATAGTT GTTTATGGTGACGCCCTATTGGGGGATGGATCCTGGATATTTAAATGAGCTCCAGTCAGTGAGG TCAGGATCTCTGGGACTCAGTGCTGCACCGCGTGTACAAAGACAAAGAGATGAATCTGGGGTAG ATTTAGGTGCTAGGAATATTGTTTCTTGAGTTCTTGACAGGTGCAGCTTCGGGGCTGCATTCCT TCTGTCTCCTAATTTTCATCTCTTTATCAGGTGATAGTAGAAAAGGCTCCCAAAGCTCGGATAG GAGACCTGGACAAAAAGAAATACCTGGTGCCTTCTG ATCTCACAGGTGGGAACCTGATCCAAGACTCAGGCTTGCTTCCTGTGGGTGGGAGTGCAGTCTC TGCGGGAGGAGGCTCCACCTAGCAGCTGTTCTTTTGAGGGCATTTTTGACCTCTGTGACTTTCT TGCATCTTGTATCTTTTGCAGTTGGTCAGTTCTACTTCTTGATCCGGAAGCGAATTCATCTCCG AGCTGAGGATGCCTTGTTTTTCTTTGTCAACAATGTCATTCCACCCACCAGTGCCACAATGGGT CAGCTGTACCAGGTATGGTGACTGGGAAGTGGTGGAGGCTTTGGAGAGGAATCTTGGAAGAGCG TGAGGGGAAGAAAGTGTTGTTACATCAGTAGTTTTGGACTAGCTTCCTGGCTTGCCTTAAAGTT TCATAATAGCCGTGGGTTGGGGGTTTGGACAGAACA GTAGGAGTAGAGAGAAGGAAAGAGAACAGTGAGCTAGAGAAACTATCTGTTCGGGACTCTTCCC AGTCCCCCTCCCCATTCCAAAGCAGCAGAAAACAGTTTGAGGCCTTT'i'TGCCCAGGACTGCAGA GATTATACTGCTCGTGTGGTAGGCCTTGGAGTCAGTGAAGGAGTCAGGCAATGGTTTGTGGTTC TAGAAGAACGTGTCTGGGTCGTGGTTGGTTTAAGTTACTGGAGCCCAGTTGCTGAACAGTTTTC TGCTTTCATCCCAGGAACACCATGAAGAAGACTTCTTTCTCTACATTGCCTACAGTGACGAAAG TGTCTACGGTCTGTGAAGCTGCTGCCCCTGAGCTGGAGGGGGGTCTCATTCTACAAAGAGAGAG GTGGCCCCCCTTTCTTGACCTCCTCCTCCTTCAAGC TCAAACACCACCTCCCTTATTCAGGACCGGCACTTCTTAATGTTTGTGGCTTTCTCTCCAGCCT CTCTTAGGAGGGGTAATGGTGGAGTTGGCATCTTGTAACTCTCCTTTCTCCTTTCTTCCCCTTT CTCTGCCCGCCTTTCCCATCCTGCTGTAGACTTCTTGATTGTCAGTCTGTGTCACATCCAGTGA TTGTTTTGGTTTCTGTTCCCTTTCTGACTGCCCAAGGGGCTCAGAACCCCAGCAATCCCTTCCT TTCACTACCTTCTTTTTTGGGGGTAGTTGGAAGGGACTGAAATTGTGGGGGGAAGGTAGGAGGC ACATCAATAAAGAGGAAACCACCAAGCTGAACTGAATTTTGCCTTGTGTTGCTCCCCTCGTCCC GCTGATTTTAAGTCTTTCCAAGGTGTCAGTGGGTTT CAGTGGTGGGGAAAGAAGAGTACTGGGTACAAGCTGGAGGGATAGAAGTATATTTTGGTTTATT CTGTTCATGTTGGGCTTTTCCCTGTCTGCAAAAAGAGGGTGCTTTTGTTGTGATGGAATGGAAT ACTGAGGATTATTTCTTGAAACTTTAGTTTTATAACACGCATGTGAAACTAAATGTTAAAAATG CTCATGTAAAAAAAAATTTTTTTTTACTGTGGGTTCCTGTGGAGAAAGTTCCGAAGTACCTGCT TTAGGTGAACATCCACATTTGCTAGAACATTCTAACTAAGATATTTTCATGTGTGCAAGCTAGT AAAACGGCTGTTCTCAGTTGCA (SEQ ID NO: 3 )Serotonin transporters (solute carrier SLC6A4 Gene ID: 6532 > NG_011747.2:5240-46618 Homo sapiens solute carrier family 6 member 4 (SLC6A4), RefSeqGene on chromosome 17family 6 member 4, sodiumdependent serotonin transporter) ATAAAA. TATTATTTATTTTGATTTTGGCTCCTGCTGAAATGTCGTGCCTGAGGTGAGTGCCTCCATGTGCCCTGGGGCTTGGGGCTGTTCGTTTCTGTCAATTTCCTGCTGTCAGACTGGGTACACTCTTGATGCACAGTCATTGAGTTTCTTTGCCCAGGAAAACCCACGTGCAGCTTTGGATGTGGGCGC ACTACTCATAATTCCCATGATGTTTGCCATACTCACCCCTCTACAGCCAAGTGCCCTAGTGCCT GAGAGAGGCAAATCTGTCTTCCACTTCTACGTTCACTTAACAGATCCCCTGCTCTGGGCTTCTT AAGGCTGCTTTAGGTAAACAACAAGTAGCTTACTAAAAATATTCTTTTTCACTGCAGCCAAGTT AAAAAAAAGAAAATATTGTGTACGAGTAGAGGCCAG GATGTGTGAGGGGGTTTCTGAAGTAGCACTTGCCCCAACCGGAGGTTGAGTCCCCCAATGACTG AGTGGGATTTCACACACCTGACATTAAGAGACAAGCAGCCGGGATGGCTCCTAACCAGATCTTC TTTCCCTGTTCCCAAACCATCTTTCTTCATAGCCGGCTCTGGGGATGAGGAGCCTGGGTTAGGA GGAAGGTTTGCAATTGACCAGGTTCCTGTTTTGAAGGCTTCCACCTAGACTTAAGATAGCACCG CTCAGAAGATGGATGTGTGTTTAG (SEQ ID NO: 4)Dopamine receptor D2 DRD2 Gene ID: 1813 > NG_008841.1:4882-70675 Homo sapiens dopamine receptor D2 (DRD2), RefSeqGene on chromosome 11 AGTACTTGTCACCTAATTATCTCCTCTCTTGATAAGCTAGATGGTCCCTTCCAGGGCAGC'i'TAG TAGAGAGCATGGGATGTGATGTTTCAGATTCCAGCTCTGCTGCACACCTGCCAGGTGAACTTGG CCACGTTACATGGCCTCTCTGGGCTTCAGTTCCCTCACCTATGAGTGGGATAAGCAAGCCCTTC TTGTAAAAGTTTTAAGAACGATACATGAGATAAAGTGCAATGCCCACAGTAGATGTCTTATGGA CAGTGGCCACCAATGCATTTTCTT (SEQ ID NO: 5)Sodium ion channels - SCN1A SCN1A Gene ID: 6323 > NG 011906.1 Homo sapiens sodium voltage-gated channel alpha subunit 1 (SCN1A),RefSeqGene (LRG_8) on chromosome 2TCATTCAGGGACAGAATCCCTCAATGAATCAGAGAAAAAGAGGGCTAAGTATTGTTGGCATAGG GGATTGTTCTTCTAGTAAGCACAGGTGAAATTCTCCCTTCCTTCAAATGCGCTTTGTGGTAAAG CTCTATCTGTTTTCTGGACTCATACTCCTTCAAAGATGATAGAGGGAATGCAGGCAGAGGTGAG ACTTGGAGCAGTGCTTCTCAAACTCTGTTGAGGATGTGAGGTCTCTATTCTTCTATAAAATGAA GTAAAAATTAGTTTCTAGAAAACTGAAATGAAAAACAGGCCCACAACAACATCCCCCTACTTTT TTATTATTAGAATAGACAGACACAACATTAATTGGCCAAATTGCCTTACAAATTTCTAAACATG GACTCTCCATATCTATACTTAACCAGTACTGGTTGG GGGACCATACTTTCACAGCCAGTTTAGGACACTTCTACCCATACCAGATAGAAAGTGCTTAATT TTCAAAAACCAACTCCATTAGCAGGTAGGGGTGTGGGAGGGTGGGTTTGCTTTACTTGAATTTA GAAAATCCTGCAAGTATAGGGGAAATCTGTTAATGAGCTAAGACTATACAAAGATTCTTAGTAG AAAATAAACATAACATTTATGCTTGCTGTATCATAACATGATCTGTGCAGATATGCATCCTAAG GCTTAGAAAAGAAAGCTTTCGAAGAATGGGTGAATTTAAAATCAAGAAGCTTAAAATTTCCTTCCCCAAATATGCTCTTCACAGAGATAAAGGC (SEQ ID NO: 6)Docket No. 229303-701601 / PCTEXAMPLESExample 1: Wetware 3- Valued Adder

[0183] The techniques described herein can be used to configure a set of biological computational units (e.g., a set of neurons) to function as a biological computational system that implements a 2-input adder with 3-valued inputs and a 5-valued output. In some examples, the biological computational system is trained to function as a 2-input adder by providing sample inputs (e.g, using an MIA 130 and / or the techniques described above) to the system, decoding the system’s output (e.g., using an MIA 130 and / or the techniques described above), and providing a reinforcing stimulus signal to the system. Each time the value of the decoded output is the correct value (the value representing the sum of the sample inputs), the same stimulus signal can be applied to the system, as the biological computational units tend to respond to expected stimuli by maintaining their behavior. When the value of the decoded output is an incorrect value (not the value representing the sum of the sample inputs), an unexpected (e.g, random) stimulus signal can be applied to the system, as the biological computational units tend to respond to unexpected stimuli by changing their behavior.

[0184] FIG.2A shows a truth table for a two-input adder with 3-valued inputs and a 5-valued output. In the example of FIG.2A, the three digital values of the 3 -valued inputs are 0, 1, and 2. These values can be represented, for example, by three different ranges of voltages that can be applied as stimulus to an SBC by the MIA 130. In the example of FIG. 2A, the five digital values of the 5-valued inputs are 0, 1, 2, 3, and 4. These values can be detected as output signals of the SBC by the MIA 130. As can be seen, the output of the adder is the sum of the adder’s two input values.Example 2: Wetware 4-Valued Logic Gates

[0185] The techniques described herein can be used to configure a set of biological computational units (e.g., a set of neurons) to function as a biological computational system that implements a 2-input logic gate with 4-valued inputs and a 4-valued output. In some examples, the biological computational system is trained to function as a 4-valued logic gate by providing sample inputs (e.g, using an MIA 130 and / or the techniques described above) to the system, decoding the system’s output (e.g., using an MIA 130 and / or the techniques described above), and providing a reinforcing stimulus signal to the system. Each time the value of the decoded outputDocket No. 229303-701601 / PCTis the correct value (the value corresponding to the sample input), the same stimulus signal can be applied to the system, as the biological computational units tend to respond to expected stimuli by maintaining their behavior. When the value of the decoded output is an incorrect value (not the value corresponding to the sample input), an unexpected (e.g., random) stimulus signal can be applied to the system, as the biological computational units tend to respond to unexpected stimuli by changing their behavior.

[0186] FIG. 2B shows truth tables for four types of 4-valued logic gates (4-valued AND, OR, NAND, and NOR gates). As discussed above, these 4- valued logic gates can be used to construct many wetware 4-valued processing systems (e.g, 4-valued ALUs). In the example of FIG. 2B, the four digital values of the 4-valued logic are 0, 1, 2, and 3. These values can be represented, for example, by four different ranges of voltages that can be applied as stimulus to an SBC by the MIA 130 and detected as output signals of the SBC by the MIA 130.

[0187] As can be seen, the output of the 4-valued AND gate is the minimum of the gate’s two input values (i.e., 4AND(A, B) = minimum(A, B)). The output of the 4- valued NAND gate is the complement of the output of 4-valued AND gate for the two input values (i.e., 4NAND(A, B) = complement(4AND(A, B))). The output of the 4-valued OR gate is the maximum of the gate’s two input values (i.e., 4OR(A, B) = maximum(A, B)). The output of the 4-valued NOR gate is the complement of the output of 4-valued OR gate for the two input values (i.e., 4NOR(A, B) = complement(4OR(A, B))).Example 3: A Stimulatory Active Human Neuron Network and Electrical Imaging System

[0188] MaxOne chips were sterilized with tergazyme and cell culture medium was added to the chips for 3 days. Chips were coated with PDL and GelTrex. Neurons were thawed and plated on the MaxOne chips. Cells were allowed approximately one hour to adhere to MEA surface before the well was flooded with media. Cultures were maintained in a low O2 incubator kept at 5% CO2, 5% O2, 36°C and 80% relative humidity. Every two days, half the media from each well was removed and replaced according to the protocol outlined in FIG. 3.

[0189] The MaxOne system is based on complementary meta-oxide-semiconductor (CMOS) technology and allowed recording from up to 1024 channels. A MaxOne electrical activity scan was performed on days 15 and 29 (FIG. 4 and FIG. 5) using a MaxOne High-density microelectrode array (HD-MEA) system (MaxWell Biosystems, Zurich, Switzerland). LicensedDocket No. 229303-701601 / PCTMaxLab Live Scope V20.1 software was used to run activity scans. Gain was set to 512x with a 300 Hz high pass filter. Spike threshold was set to be a signal six sigma greater than background noise as per recommended software settings. Mean, max and variance of both amplitudes and firing rates was extracted from these assays and mapped using custom software: the first nine components of discrete cosine transform basis functions of space were used to summarize the spatial profile of spiking activity. The ensuing coefficients were then used in subsequent correlation analyses. Functioning glutamatergic neurons were evaluated 18 days post-thaw and shown in FIG. 3. Monoculture activity was approximately 0.3% across the duration of the scans.Example 4: Genetically Modified Human Neuron Network and Electrical Imaging System

[0190] Neuro2a cells and human NT2 neuronal cells were thawed and cultured. The cells were modified using CRISPR-Cas gene editing. Exemplary guide RNAs are provided in Table 2 and Table 3.Table 2. GABA Receptor KO CRISPR gRNA designs.SAM gRNA name SAM gRNA sequence Target Ref Seq GABRG3 SAM guide RNA 1 AGCGCACACAGCCCCGGCCT NM_033223 GABRG3 SAM guide RNA 2 GTCCCCTGGCAGCGCCTCCG NM_033223 GABRG3 SAM guide RNA 3 TGGGGAAACTCCTCCCCGCC NM_033223Table 3. Glutamate Receptor KO CRISPR gRNA designs.SAM gRNA name SAM gRNA sequence Target RefSeq GRIA1 SAM guide RNA 1 ACTGTGGGGTTGCCCCTTTC NM_001258020 GRIA1 SAM guide RNA 2 AGGTGGGGGTGTGTGACACG NM_001258020 GRIA1 SAM guide RNA 3 TCCTTTCTGTGTGTGCAGAA NM_001258020

[0191] Cells were seeded on MaxOne chips for evaluation according to the protocols shown in Example 2.Example 5: Glutamatergic Neuron and Astrocyte Co-Culture Produces a Robust Neuronal NetworkDocket No. 229303-701601 / PCT

[0192] Six MaxOne chips were prepared. Glutamatergic neurons and rat astrocytes were seeded on the chips in co-culture. Electrical activity was evaluated 9 days post-seeding.

[0193] Chips that were seeded with both neurons and astrocytes exhibited a higher active electrode % relative to the chips seeded only with neurons without astrocytes (FIG. 6). Therefore, synthetic biological circuits comprising two cell types enhanced the overall electrical output of the circuit.Example 6: Genetically Modified Human Neuron Network and Electrical Imaging System for Computing Applications

[0194] Brain organoids are artificially grown 3D aggregates that resemble the human brain from cells described in Examples 2-4,

[0195] Three different sets of neurons are modified using CRISPR-Cas9 to knock out the glutamate receptor using the guides provided in Table 3 by transfecting each set of neurons with Cas9 and gRNA complexes targeting GRIA1. A control, unmodified set of neurons are also cultured in parallel. Astrocytes are added to increase electrical activity of the neurons.

[0196] A multi-well high resolution microelectrode array (MEA) system is used to record every active cell across multiple areas of biological samples with readouts at different scales such as, the network level (population spike time, population spike rates, bursts), cell level (individual spike times, individual spike rates, waveform), and the sub-cellular level (spatially resolved waveforms).

[0197] Neuronal spikes are recorded in 4-logic gate files according to spike frequency and amplitude. When the firing rate of the neurons is constant over a selected time period (e.g., 500 milliseconds to 1 second) across the majority of electrodes (e.g., greater than 2-5 electrodes), the neurons receive predictable, stimulatory pulses that are in the form of a short square bi-phasic pulse delivered at 75 millivolts (mV) at 100 Hz over approximately 100 ms. A Digital to Analog Converter (or DAC) on the MEA reads and applies this pulse sequence to a given electrode for the “m sync” neurons.

[0198] When the firing rate of neurons is erratic over the selected time period for any electrode, the erratic firing neurons can receive an unpredictable external stimulus into the system, this feedback stimulus can be set at 150 mV voltage and 5 Hz. This stimulation can be applied at random sites at a random timescale over the 8 predefined input electrodes, for a period of four seconds, followed by a configurable rest period of four seconds where stimulation is paused. UponDocket No. 229303-701601 / PCTregulation of the firing rate of the neurons, the predictable pulse is given. As the neurons are receiving positive and negative feedback, the SBCs provided herein can be used to complete simple tasks using logic and mathematical algorithms. Genetic engineering of the cells used in the SBC can selectively enhance electrical signals, such as firing rate and current amplitudes, that are decoded and provide output signals to the in silico computer. The genetically engineered cells can also increase the computational processing speed and efficiency of the systems provided herein.Example 7: Biological Qubits

[0199] Neuronal cells of various types, (e.g., human iPSC-derived neurons and genetically-modified neurons) were plated on microelectrode arrays (MEAs) in order replicate continuous analogue logic gates and demonstrate the ability to perform simple mathematical computations to establish a new class of computer, a bioquantum computer. The hardware for the system is shown in FIG. 7. The timeline and protocols for neuronal cell culture, preparing the MEA system, and performing the assays are provided in further detail below and in FIG. 8, FIG. 9, FIG. 10, FIG.11, and FIG. 12.Neuronal Cell Culture

[0200] Human induced pluripotent stem cell (iPSC)-denved neurons enable the study of neuronal cell biology in both healthy and pathological contexts. In vitro phenotypic assays of sensory neuron activity are important tools for characterizing hyperexcitable and / or abnormally spontaneously active cells. The following materials were used in the neuronal culture for bioquantum computing.Table 4.Item Supplier6-well plate (with Lid, autoclave Lid MaxWell Biosystems AGbefore use)MaxTwo Mainframe MaxWell Biosystems AGPrimary Neurons (cortical, Transnetyx*hippocampal, etc. ) In-house generated cellsHuman iPSC-derived Neurons Fujifilm Cellular Dynamics* bit.bio*Elixirgen Scientific* BrainXell*In-house generated cellsBorate Buffer 20X ThermoFisher ScientificCell Culture Media Multiple VendorsEthanol 70% Multiple VendorsDocket No. 229303-701601 / PCTSterile Deionized Water Multiple VendorsSurface Coating Material Multiple VendorsSterile Syringe Filter (0.22 pm pore) Multiple VendorsTerg-a-zyme Sigma- AldrichBreathe-Easy® Sealing Membrane Sigma-AldrichNuncTMSquare BioAssay Dishes ThermoFisher Scientific(optional)Low Retention Tip (optional) Multiple VendorsIncuboxTM(optional)** InspheroCell culture reagents:1% BSA:0.5g BSA in 50ml PBS - dilute to 0.1% for use each timeNT3:Stock: 50ug / ml (5000x) - reconstitute 5ug in lOOul PBS + 0.1% BSA25x 4ul aliquot - -80°CBDNFStock: lOug / ml (2000x) - reconstitute 5ug in 500ul PBS + 0.1% BSA50x lOul aliquot - -80°CDAPT:Stock: 20mM (2000x) - reconstitute 1 Omg in 1156ul DMSO12x lOOul aliquot - -80°CDoxycycline:Stock: 20mg / ml (20000x) - reconstitute 1g in 50ml H2O50x 1ml aliquots - -80°CPoly-D-Lysine:lx borate buffer: 2.5ml of 20X borate buffer stock + 47.5ml H2OResuspend 5mg vial in 50ml lx borate buffer (lOOug / ml)50x 1ml aliquots - -20°C storageGeltrex:Thaw on ice in fridge overnight - KEEP COLD50x lOOul aliquots - -80°C storageMedia prep:Basal medium (b: GN):98ml Neurobasal1ml Pen / strep1ml Glutamax50ul B-mercaptoethanolStored at 4°C 3 weeks - aliquot off some working stockComplete medium (comp: GN):19.5ml basal mediumDocket No. 229303-701601 / PCT- 400ul B274ul NTS (lOng / ml final)lOul BDNF (5ng / ml final)Stored at 4°C 4 daysComplete medium + Doxycycline (comp: GN+D):5ml complete medium[1:10 20mg / ml doxy in water] → 2.5ul (1ug / ml final)Add fresh each timeComplete medium + Doxycycline + DAPT (comp: GN+D+DAPT):5ml complete medium[1:10 20mg / ml doxy in water] - 2.5ul ( 1 ug / ml final)- 2.5ul DAPT ( 1 OuM final)sgRNA prep:1. Reconstitute 2nmol vial in 20ul sterile H2O - 100uM stock2. Prepare 10uM working solution by diluting 1:10 in sterile H2OGAB ARAP:GGATCTTCTCGCCCTCAG GAAGATCCGAAAGAAATA GTAGACACTTTCGTCACT AG CC GT6-Well Plate / Chip Preparation

[0201] A 1%-Terg-a-zyme solution (10 g / L) in deionized water was prepared. 2 mL of 1%-Terg-a-zyme solution was added into each well of a 6- well plate and incubated at room temperature for 2 hours, The 1%-Terg-a-zyme solution was removed, and each well of the 6-well plate was washed three times with deionized water. The solution was completely washed out,6-well plate Sterilization and Pre-conditioning

[0202] The 6-well plate front and back were sprayed thoroughly with 70% ethanol. Each well and compartment of the 6-well plate was filled with 70% ethanol. The 6-well plate was transferred to the biological safety cabinet. The 70% ethanol was removed from the wells and compartments after 30 mins. Each well of the 6-well plate was washed three times with sterile deionized water. The water was aspirated with a vacuum pump, and the bottom of the 6-well plate was dried. Each well of the 6-well plate was filled with 1.2 mL of complete cell culture media. The 6-well plate was covered with a Breathe-Easy® sealing membrane, and the autoclaved lid was placed on top. The 6-well plate was kept inside the 5% CO? incubator at 37°C, RH >95%, for 2 days. Before cellDocket No. 229303-701601 / PCTplating, the complete cell culture medium was aspirated from the 6-well plate, and each well was washed once with sterile deionized water. The water was completely aspirated with a vacuum pump. Surface coating was then performed.6-Well Plate Surface Coating

[0203] 100 mL of 1X borate buffer was prepared by diluting 5 mL of 20X borate buffer in 95 mL sterile deionized water. A 7% PEI stock solution was prepared by weighing 1.4 g of 50% PDL solution into a sterile beaker and adding 8.6 mL of sterile deionized water. The solution was mixed for approximately 30 mins using a magnetic stirrer until fully dissolved. A final 0.07% PDL solution was prepared by diluting 0.5 mL of intermediate 7% PEI solution in 49.5 mL of 1X borate buffer fresh before use. The solution was sterilized before use by filtering through a 0.22 µm filter unit. 100 mL of IX borate buffer was prepared by diluting 5 mL of 20X borate buffer in 95 mL sterile deionized water. A 100 µg / mL suspension was prepared by resuspending 5 mg of PDL in IX borate buffer. The solution was sterilized before use by filtering through a 0.22 µm filter unit.50 µL of the coating solution was added to the center of each well in the 6-well plate, covering the entire electrode array. The 6-well plate was covered with a Breathe-Easy® sealing membrane, and the autoclaved lid was placed on top. The 6-well plate was incubated in a 5% CO? incubator at 37°C, RH >95%.

[0204] The coating solution was aspirated completely. Each well of the 6-well plate was washed three times with sterile deionized water. The sterile deionized water was aspirated with a vacuum pump, and the 6-well plate was dried inside the biological safety cabinet for 1 hour.Secondary) Coating

[0205] Laminin stock was thawed on ice, aliquots were prepared, and diluted in chilled solution as specified. Geltrex stock was thawed on ice m a 4°C fridge overnight, aliquots were prepared, and diluted 1:100 in chilled medium. For Cell Plating, 5 to 50 µL of secondary coating solution was added to the center of each well in the 6-well plate. The 6-well plate was covered with a Breathe-Easy® sealing membrane, and the autoclaved lid was placed on top. The 6-well plate was incubated according to the specified conditions. The secondary coating solution was aspirated, leaving a thin film behind (no washing), and cell plating was immediately performed.Docket No. 229303-701601 / PCTCell Thawing and Co-Culture Preparation (Day 0)Media Preparation

[0206] Three media types were prepared prior to cell thawing: basal medium (b: GN), complete medium (comp: GN), and complete medium supplemented with doxycycline (comp: GN+D).Thawing and Processing Rat Astrocytes

[0207] Rat astrocytes (Gibco; lot #2989011) were rapidly thawed in a 37 °C water bath. The cell suspension was transferred into an empty 15 mL Falcon tube. The cryovial was rinsed with 1 mL b: GN, which was added dropwise to the Falcon tube. An additional 8 mL b: GN was added dropwise, and the tube was gently inverted. Cells were centrifuged at 250 x g for 5 min.Thawing and Processing Glutamatergic Neurons

[0208] ioGlutamatergic neurons (BitBio; lot #PRD000139) were thawed in a 37 °C water bath and transferred into an empty 15 mL Falcon tube. The cryovial was rinsed with 1 mL b: GN and added dropwise to the Falcon tube, followed by 4 mL b: GN added dropwise. The suspension was gently inverted and centrifuged at 200 × g for 3 min. The supernatant was discarded.Cell Resuspension and Counting

[0209] Both cell types were resuspended in 1 mL comp: GN. Cell counts were as follows: -Glutamatergic neurons: 1.14 x 106(77% viability)-Astrocytes: 1.1 x 106(78.5% viability)For each chip / well, 1.5 × 105glutamatergic neurons and 3 × 104astrocytes were used. For 10x conditions, 1.5 × 106glutamatergic neurons and 3 × 105astrocytes were required (corresponding to 1 mL and 272 pL, respectively).Co-Culture Assembly and Seeding

[0210] The required volumes of glutamatergic neurons and astrocytes were combined in a 15 mL Falcon tube, centrifuged at 200 × g for 3 min, and the supernatant was discarded. The pellet was resuspended in 60 pL comp: GN+D. A 7 pL aliquot of the resuspended cells was embedded directly into each Geltr ex- coated drop on the chip / well.Docket No. 229303-701601 / PCTIncubation

[0211] Chips were incubated at 37 °C with 5% CO2 for 1 h, after which 600 pL comp: GN+D was added to each chip / well. Cultures were maintained at 37 °C with 5% CO2 for 2 days.Genetic Engineering

[0212] The neuro-excitatory and inhibitory state of the neurons were, in some assays, manipulated by CRISPR genetic engineering such that the action potentials of these neurons and their analogue non- binary states were manipulated to create changes in the speed of these neural networks and thus the response to input / output stimuli creating novel computing paradigms.

[0213] Guide RNAs (gRNAs) were designed to the desired region of a gene using the CRISPR Design Program (Zhang Lab, MIT 2015). Multiple primers to generate gRNAs were chosen based on the highest ranked values determined by off-target locations,

[0214] The gRNAs were cloned together using the target sequence cloning protocol (Zhang Lab, MIT). Briefly, the oligonucleotide pairs were phosphorylated and annealed together using T4 PNK (NEB) and 10X T4 Ligation Buffer (NEB) in a thermocycler with the following protocol: 37°C 30 minutes, 95°C 5 minutes and then ramped down to 25°C at 5°C / minute. pENTRl-U6-Stuffer-gRN / X vector (made in house) was digested with FastDigest BbsI (Fermentas), FastAP (Fermentas) and 10X Fast Digest Buffer were used for the ligation reaction. The digested pENTRl vector was ligated together with the phosphorylated and annealed oligo duplex (dilution 1:200) from the previous step using T4 DNA Ligase and Buffer (NEB). The ligation was incubated at room temperature for 1 hour and then transformed and subsequently mini-prepped using GeneJET Plasmid Miniprep Kit (Thermo Scientific). The plasmids were sequenced to confirm the proper insertion.

[0215] Neuronal receptors involved in excitatory synaptic transmission were targeted by engineered gRN As, and include target sequences of the genes provided in the table below. Gene sequences were obtained from the National Center for Biotechnology Information (NCBI), which is available on the world- wide web at chttps: / / www. -cbi.r-l.mh.gov / gene>.Table 5.Docket No. 229303-701601 / PCTGene Name Target Name NCBI Gene ID and (SEQID NO)GABRA1 Gamma-aminobutyric acid type A 2554 (SEQ ID NO: 7)receptor subunit alpha- 1GABRA2 Gamma-aminobutyric acid type A 2555 (SEQ ID NO: 8)receptor subunit alpha-2GABRB3 Gamma-aminobutyric acid type A 2562 (SEQ ID NO: 9)receptor subunit beta- 3GABBR1 Gamma-aminobutyric acid type B 2550 (SEQ ID NO: 10)receptor subunit 1GABBR2 Gamma-aminobutyric acid type B 9568 (SEQ ID NO: 11)receptor subunit 2GABRR1 Gamma-aminobutyric acid type A 2569 (SEQ ID NO: 12)receptor subunit rho-1GABRR2 Gamma-aminobutyric acid type A 2570 (SEQ ID NO: 13)receptor subunit rho-2GABRR3 Gamma-aminobutyric acid type A 200959 (SEQ ID NO: 14)receptor subunit rho-3GRIA1 Glutamate receptor ionotropic, 2890 (SEQ ID NO: 15)AMP A type subunit 1GRIA2 Glutamate receptor ionotropic, 2891 (SEQ ID NO: 16)AMP A type subunit 2GRIA3 Glutamate receptor ionotropic, 2892 (SEQ ID NO: 17)AMPA type subunit 3

[0216] Cell culture medium was combined with the gRNA plasmid, and a Cas9 plasmid. Another tube of cell culture medium was combined with approximately I ul of Lipofectamine 2000 Transfection reagent (Invitrogen). The solutions were combined together and incubated for 15 minutes at room temperature. The DNA-lipid complex was added dropwise to wells of the 24 well plate. Cells were incubated for 3 days at 37°C and genomic DNA was collected using the GeneJET Genomic DNA Purification Kit (Thermo Scientific). Activity of the gRNAs was quantified by a Surveyor Digest, gel electrophoresis, and densitometry.

[0217] Sequences of target integration sites were acquired from an ensemble database. PCR primers were designed based on these sequences using Primer3 software to generate targeting vectors of varying lengths, Ikb, 2kb, and 4kb in size. Targeting vector arms were then PCR amplified using Accuprime Taq HiFi (Invitrogen), following manufacturer’s instructions. The resultant PCR products were then sub cloned using the TOPO-PCR-Blunt II cloning kit (Invitrogen) and sequence verified. Expression of the targeted genes were evaluated in neurons. Double strand breaks were created in the target gene sites to modulate neuronal activity.Docket No. 229303-701601 / PCT

[0218] Knockout of the ionotropic glutamate receptors (also referred to as -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors, AMP A receptors, or by their protein acronyms-GluAl, GluA2, and GluA3) improves neuron longevity by decreasing stimulation. Knockout of GABA receptors increased neuron signaling speed and activity.CRISP R-Mediated Knockout

[0219] To perform GAB ARAP knockout in glutamatergic neurons, the transfection was executed using sgRNA-lipid complexes as follows:

[0220] Preparation of sgRNA complexes: For each condition, 75 L of Opti-MEM 'as combined with 3 pL of sgRNA (10 pM; prepared by diluting a 100 pM stock 1:10 in sterile water).

[0221] Preparation of lipofectamine solution: For each condition, 75 pL of Opti-MEM was mixed with 4,5 pL of Lipofectamine RNAiMAX

[0222] Complex formation: 75 pL of the sgRNA solution was combined with 75 pL of the lipofectamine solution, and the mixture was incubated at room temperature for 5 minutes.

[0223] Cell preparation: 500 pL of medium was removed from each well.

[0224] Transfection: 150 L of the sgRNA-lipid complex was added to the corresponding well, and 500 pL of plain Neurobasal medium was supplemented.

[0225] Incubation: The cells were maintained at 37 °C in a humidified 5% CO? incubator for 5 hours.

[0226] Post-transfection medium change: 500 pL was aspirated from each well and replaced with fresh, pre- armed complete medium supplemented with doxycycline and DAPT (comp: GN + D + DAPT).

[0227] Expression of the targeted genes were evaluated in neurons. Double strand breaks were created in the target gene sites to modulate neuronal activity. Knockout of the ionotropic glutamate receptors (also referred to as -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors, AMI’ A receptors, or by their protein acronyms- GluAl, GluA2, and GluA3) improves neuron longevity by decreasing stimulation. Knockout of GABA receptors increased neuron signaling speed and activity.Docket No. 229303-701601 / PCTCell Plating

[0228] The cell suspension was added at the center of the electrode array according to manufacturer’s instructions. For example, a droplet of 7 pL contained approximately 250,000 neurons per MEA.

[0229] The 6- well plate was incubated m a 5% CO2 incubator at 37°C, RH >95%, for 1 hour. After incubation, each well in the 6- well plate was filled with 1.2 mL of pre- warmed complete culture medium. The liquid was gently pipetted on the side of each well.Main tenance and Recording

[0230] 50% of the complete cell culture media was changed on day 2 after plating (or as recommended by the cell supplier). The cell cultures were maintained in a 5% CO2 incubator at 37°C, RH >95%, and 50% of the media was replaced three times a week. Cells formed a confluent array that was viable for over 3 months and also maintained synaptic density (FIG. 13).

[0231] Neural activity was recorded 4 to 24 hours after medium change using the MaxLab Live software by MaxWell Biosystems. The MaxTwo Mainframe and gas supply were turned on 30 mins prior to inserting the 6-well plate. After plugging the 6-well plate into the MaxTwo Mainframe, a 10-min wait was observed before starting the recording session.Example 8: Biological Neural Networks and BioQuantum Computing

[0232] An evoked membrane potential measurement technique was used to convert the analogue signal of the continuous neural network into a digital signal, the cellular arrangement is used as a biological qubit. The central processing unit was a confluent culture of neurons tied to an analogue to digital converter with a standard UI / UX in order to send and receive input and output. A closed loop experiment architecture is shown in FIG. 14. The biological neural network is shown in FIG. 15A-FIG. 15B

[0233] The assays for training the biological qubit included an activity scan assay, a network activity assay, and stimulation assay. An exemplary training protocol is provided below. Cells were trained 5x per week to maintain better-than-probability training accuracies e.g. 50% for 2-digit. The MEAs were regularly subjected to an Activity Scan, which monitors the health and functional electrical activity of the neurons on the chip.Docket No. 229303-701601 / PCTTable 6.Trial Session 1 Rest 6 hours Trial Session 2 Rest 15 hours 1.5 hr Addition of cell 1.5 hr Addition of Activity’ scan 5-10 min culture medium Activity scan 5-10 min cell culture Training trials 1 hr Incubation period Training trials 1 hr medium Inference trials 10-20 min* at 37 ° C, 5% CO2, Inference trials 10-20 min* Incubation Activity’ scan 5-10 min humidified for 6 Activity scan 5-10 min period at 37 °C, 5% CO2, humidified for15 hours

[0234] The neuronal biological qubit cultures were split into one of two groups - a training group, and an interference group. Each group of neurons were given input stimuli that was representative of a computational problem in which the neurons would actively respond, by generating field potentials that correspond with a prediction. The interference group would receive a random stimulus whenever a prediction was incorrect. The training group did not receive any feedback for correct or incorrect predictions but were given a constant stimulus during training. Once the analogue to digital signals were captured, a coding language with a non-binary base X, was devised to take advantage of the continuous analogue states embedded within the neural network / CPU which allows for computation beyond a base-2 binary system (FIG. 15C).

[0235] An increase in activity during the session, returns activity to the original % activity during the rest period (FIG. 16A). The neurons were the most viable when % activity remained stable between each trial session. The neurons lost viability when the % activity did not return to the original % activity and the % of neuronal firing would predictably decrease decreases during the training sessions (FIG. 16B). As training and interference testing were performed, the prediction accuracy of the neurons in the interference group, significantly improved prediction accuracy by the third session relative to the prediction accuracy of the training group of neurons (FIG. 16C). Therefore, the biological qubit could be trained to predict the outcome of computational problems.Example 9: CRISPR / Cas Gene Editing of Glutamate and GABA

[0236] GABAergic neurons inhibit neural signaling by releasing GABA that binds the inhibitory GABA-A / B Receptors. Glutamatergic neurons release glutamate, which is essential for increased neural signaling and network excitation. Intrinsic neuronal activity and processing speedDocket No. 229303-701601 / PCTcan be increased or decreased by genetic modification of neurons in the biological circuit (FIG.17). The neurons in the biological circuit were engineered to remove GABAergic or glutamatergic activity via CRISPR / Cas engineering.

[0237] Neurons were genetically engineered by the methods provided in Example 8. sgRNAs are provided in Table 7.Table 7.Target NCBI SAM gRNA name SAM gRNA sequenceRefSeq GABRG3 SAM AGCGCACACAGCCCCGGCCT (SEQ ID NO: NM_033223 guide RNA 1 18) (SEQ ID NO: 25) GABRG3 SAM NM 033223GTCCCCTGGCAGCGCCTCCG (SEQ ID NO: 19)guide RNA 2 (SEQ ID NO: 26) GABRG3 SAM NM_033223TGGGGAAACTCCTCCCCGCC (SEQ ID NO: 20)guide RNA 3 (SEQ ID NO: 27) GRIA1 SAM guide NM 001258020ACTGTGGGGTTGCCCCTTTC (SEQ ID NO: 21)RNA 1 (SEQ ID NO: 28) GRIA1 SAM guide AGGTGGGGGTGTGTGACACG (SEQ ID NO: NM_001258020 RNA 2 22) (SEQ ID NO: 29) GRIA1 SAM guide NM 001258020 TCCTTTCTGTGTGTGCAGAA (SEQ ID NO: 23)RNA 3 (SEQ ID NO: 30) GRIA1 SAM guide NM_001258020ACTGTGGGGTTGCCCCTTTC (SEQ ID NO: 24)RNA 1 (SEQ ID NO: 31)

[0238] CRISPR-mediated knock out of GABA receptors increased neuron signaling speed. CRISPR-mediated knock out of glutamate receptors increased neuronal longevity and decreased overstimulation of the neurons (FIG. 18). Targeted genetic editing of these receptors enabled precise control of neural behavior and electrical activity.

[0239] Synaptic connections were fully established by day 21 and were maintained over 100+ days In CRISPR-gene edited neurons. Neurons exhibited elevated electrical signaling showing active network formation (FIG. 19). By Day 21, the neuronal networks show synchronous bursting behavior, reflecting coordinated activity. Neurons respond to external stimulation indicating early network excitability (FIG. 20), These traits correspond with the development of mature, healthy neural circuits capable of coordinated activity.

[0240] An exemplary training session is provided in FIG. 15D. Typically when the active electrode percentage was reduced by approximately 50% of its own peak, the training accuracy became impacted. At this point, the accuracy was not recoverable, so the MEA was replaced. AnDocket No. 229303-701601 / PCTincrease in activity during the session, returns activity to the original % activity during the rest period (FIG. 16A) The neurons were the most viable when % activity remained stable between each trial session. The neurons lost viability' when the % activity did not return to the original % activity and the % of neuronal firing would predictably decrease decreases during the training sessions (FIG. 16B). As training and interference testing were performed, the prediction accuracy of the neurons in the interference group, significantly improved prediction accuracy by the third session relative to the prediction accuracy of the training group of neurons (FIG. 16C). Therefore, the biological qubit could be trained to predict the outcome of computational problems.Example 10: Training synthetic biological circuits (SBCs)

[0241] Artificial intelligence has revolutionized the nature of computing, ushering in a new era of independent, autonomous work. However, this comes at an enormous energy expenditure. While quantum computing promises exponential leaps through qubit-based logic and nonbinary computation, it also comes with immense energy requirements.

[0242] In some examples, the asynchronous, multistate nature of human neurons is leveraged to enable massively parallel processing at ultra-low energy consumption. This innovation involves a new, foundational logic unit: the biologic qubit.

[0243] One technical challenge is how to capture the analog biologic signal of human neurons and how to guide them to produce useful digital signals that can interact with the current state of modern computing. One component of our hardware is a microelectrode array, or MEA (FIG.21A), that is able to measure electrical activity in human cells, including neurons. The MEA contains a chip (FIG. 21B). Neurons are cultured on the chip (FIG. 21 C), which maps spatially, temporally and electrically to our neurons via software.

[0244] In some examples, the software interfaces with the microelectrode array to create a command and control architecture with the neurons. This architecture demonstrates the use of a digital switch or field programmable analog array. In this case, the microelectrode array creates coherent logical stimulus response commands to the analog neurons. We send stimulation sequences with unique voltages and frequencies to train the neurons and create a biological neural network, or BNN.

[0245] The biologic qubit (FIG. 21D) is a logic unit that is mapped onto the neurons embedded in the MEA. This biologic qubit serves as a configurable logic block which can scale to createDocket No. 229303-701601 / PCTparallel computing structures. A core workflow in machine learning is the tokenization of data to allow for training and statistical inference. Described below are training and inference paradigms using neurons in the biologic qubit. We run both training and inference trials, in which the training trials include a full closed loop with feedback stimulation and the inference trials do not receive feedback.

[0246] In some examples, the training and inference process is based upon the free energy principle, which states that neurons and indeed all biologic systems organize themselves to reduce unpredictability and lower entropy. Creating the BNN can consist of running trials to tram the neurons as a first proof of principle to recognize digit integer patterns (FIG. 21E).

[0247] A trial can include 4 phases. In the ambient recording phase (FIG. 21F), baseline neuronal activity is captured before the sensory stimulation. The sensory stimulation phase (FIG. 21 G) stimulates the neurons with spatial and frequency coded sequences. The monitoring phase (FIG. 21H) captures the neuronal response to the sensory stimulation. Then the feedback phase (FIG 211) stimulates the neurons with either a deterministic sequence which is short and predictable for a stochastic sequence which is random and unpredictable. The feedback sent depends on whether the neural activity is correct or not.

[0248] In one experiment, results for a 2-digit integer within the first month include a training accuracy of 60% and an inference accuracy of 75%. In some examples, a single biologic qubit is used. In other examples, multiple biologic qubits (FIG. 21J) are integrated on a microelectrode array, and by extension, these biologic neural networks can scale to allow for massive parallel computation.

[0249] The result is a trainable biological neural network driven by a core unit of logic, the biologic qubit. At scale, this architecture has the potential to change the computer-to-energy-consumption paradigm in modern computing.Example 11: Computer System

[0250] While the foregoing disclosure sets forth some examples using specific block diagrams, flowcharts, and examples, each block diagram component, flowchart step, operation, and / or component described and / or illustrated herein can be implemented, individually and / or collectively, using a wide range of hardware, software, or firmware (or any combination thereof) configurations. In addition, any disclosure of components contained within other components is toDocket No. 229303-701601 / PCTbe considered as non-limiting examples since many other architectures can be implemented to achieve the same functionality.

[0251] Included in the discussion above are flowcharts showing steps and acts in accordance with the present disclosure. The processing and decision blocks of the flowcharts above represent steps and acts that can be included in algorithms that carry out these processes. Algorithms derived from these processes (or steps thereof) can be implemented as software integrated with and directing the operation of one or more single- or multi-purpose processors (e.g., central processing units (CPUs), graphics processing units (GPUs), tensor processing units (TPUs), hardware accelerators, etc.) which can include SIMD hardware or be used in combination with SIMD hardware. It is to be appreciated that the flowchart(s) included herein do not depict the syntax or operation of any particular circuit or of any particular programming language or type of programming language. Rather, the flowchart(s) illustrate the functional information one of ordinary skill in the art can use to fabricate circuits or to implement computer software algorithms to perform the processing of a particular apparatus carrying out the types of techniques described herein. It is to also be appreciated that, unless otherwise indicated herein, the particular sequence of steps and / or acts described in each flowchart is merely illustrative of the algorithms that can be implemented and can be varied in implementations and embodiments of the principles described herein.

[0252] Accordingly, in some embodiments, the techniques described herein can be embodied in computer-executable instructions implemented as software, including as application software, system software, firmware, middleware, embedded code, or any other suitable type of software. Such computer-executable instructions can be written using any of a number of suitable programming languages and / or programming or scripting tools, and also can be compiled as executable code.

[0253] When techniques described herein are embodied as computer-executable instructions, these computer-executable instructions can be implemented in any suitable manner, including as a number of functional facilities, each providing one or more operations to complete execution of algorithms operating according to these techniques. A “functional facility,” however instantiated, is a structural component of a computer system that, when integrated with and executed by one or more computers, causes the one or more computers to perform a specific operational role. A functional facility can be a portion of or an entire software element. For example, a functionalDocket No. 229303-701601 / PCTfacility can be implemented as a function of a process, or as a discrete process, or as any other suitable unit of processing. If techniques described herein are implemented as multiple functional facilities, each functional facility can be implemented in its own way; all need not be implemented the same way. Additionally, these functional facilities can be executed in parallel and / or serially, as appropriate, and can pass information between one another using a shared memory on the computer(s) on which they are executing, using a message passing protocol, or in any other suitable way.

[0254] Generally, functional facilities include routines, programs, objects, components, data structures, etc, that perform particular tasks or implement particular abstract data types. Typically, the functionality of the functional facilities can be combined or distributed as desired in the systems in which they operate. In some implementations, one or more functional facilities carrying out techniques herein can together form a complete software package. These functional facilities can, in alternative embodiments, be adapted to interact with other, unrelated functional facilities and / or processes, to implement a software program application,

[0255] Some exemplary functional facilities have been described herein for carrying out one or more tasks. It is to be appreciated, though, that the functional facilities and division of tasks described is merely illustrative of the type of functional facilities that can implement the exemplary techniques described herein, and that embodiments are not limited to being implemented in any specific number, division, or type of functional facilities. In some implementations, all functionality can be implemented in a single functional facility. It is to also be appreciated that, in some implementations, some of the functional facilities described herein can be implemented together with or separately from others (i.e., as a single unit or separate units), or some of these functional facilities can be omitted.

[0256] Computer-executable instructions implementing the techniques described herein (when implemented as one or more functional facilities or in any other manner) can, in some embodiments, be encoded on one or more computer-readable media to provide functionality to the media. Computer-readable media include magnetic media such as a hard disk drive, optical media such as a Compact Disk (CD) or a Digital Versatile Disk (DVD), a persistent or non-persistent solid-state memory (e.g., Flash memory, Magnetic RAM, etc.), or any other suitable storage media. Such a computer-readable medium can be implemented in any suitable manner, including as system memory, accelerator memory, and / or system memory of a computer system (e.g., theDocket No. 229303-701601 / PCTcomputer system 100 of FIG. 1A or the computer system 160 of FIG. 1B) or as a stand-alone, separate storage medium. As used herein, “computer-readable media” (also called “computer- readable storage media”) refers to tangible storage media. Tangible storage media are non-transitory and have at least one physical, structural component. In a “computer-readable medium,” as used herein, at least one physical, structural component has at least one physical property that can be altered in some way during a process of creating the medium with embedded information, a process of recording information thereon, or any other process of encoding the medium with information. For example, a magnetization state of a portion of a physical structure of a computer-readable medium can be altered during a recording process.

[0257] In some, but not all, implementations in which the techniques can be embodied as computer-executable instructions, these instructions can be executed on one or more suitable computing device(s) operating in any suitable computer system, or one or more computing devices (or one or more processors of one or more computing devices) can be programmed to execute the computer-executable instructions. A computing device or processor can be programmed to execute instructions when the instructions are stored in a manner accessible to the computing device / processor, such as in a local memory (e.g., an on-chip cache or instruction register, a computer-readable storage medium accessible via a bus, a computer-readable storage medium accessible via one or more networks and accessible by the device / processor, etc.). Functional facilities that comprise these computer-executable instructions can be integrated with and direct the operation of a single multi-purpose programmable digital computer apparatus, a coordinated system of two or more multi-purpose computer apparatuses sharing processing power and jointly carrying out the techniques described herein, a single computer apparatus or coordinated system of computer apparatuses (co-located or geographically distributed) dedicated to executing the techniques described herein, or any other suitable system.

[0258] Embodiments have been described where the techniques are implemented in circuitry and / or computer-executable instructions. It should be appreciated that some embodiments may be in the form of a method, of which at least one example has been provided. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in w’hich acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.Docket No. 229303-701601 / PCT

[0259] Various aspects of the embodiments described above may be used alone, in combination, or in a variety of arrangements not specifically discussed in the embodiments described in the foregoing and is therefore not limited in its application to the details and arrangement of components set forth in the foregoing description or illustrated in the drawings. For example, aspects described in one embodiment may be combined in any manner with aspects described in other embodiments.

[0260] Having thus described several aspects of at least one embodiment, it is to be appreciated that various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements are intended to be part of this disclosure, and are intended to be within the spirit and scope of the principles described herein. Accordingly, the foregoing description and drawings are by way of example only.Docket No. 229303-701601 / PCTSome Embodiments

[0261] Some embodiments may include any of the following:

[0262] (Al) A biological computer system comprising: an interface 105 including a multi¬ electrode array (MEA) 130 and an interface controller 145 communicatively coupled to the ME A 130; and a synthetic biological circuit 135 disposed on the MEA 130, wherein the interface controller 145 is configured to encode data in base-N electrical signals provided to the synthetic biological circuit by the MEA, wherein N > 2.

[0263] (A2) The biological computer system of clause Al, wherein the interface controller 145 is configured to decode data encoded in base-K electrical signals provided by the synthetic biological circuit and sensed by the MEA 130, wherein K > 2.

[0264] (A3) The biological computer system of clause A l, wherein the synthetic biological circuit is configured to sense the base-N electrical signals, perform an ill-posed inverse task based on the data encoded in the base-N electrical signals, and provide base-K electrical signals encoding a result of the ill-posed inverse task, wherein K > 2.

[0265] (A4) The biological computer system of clause A3, wherein the ill-posed inverse task is an image reconstruction task, an image processing task, an inverse scattering task, or a matrix inversion task.

[0266] (A5) The biological computer system of clause A4, wherein the synthetic biological circuit is configured to perform the ill-posed inverse task using quantum computation.

[0267] (A6) The biological computer system of clause Al, wherein the synthetic biological circuit is configured to sense the base-N electrical signals, deconvolve the data encoded in the base-N electrical signals, and provide base-K electrical signals encoding the deconvolved data, wherein K > 2.

[0268] (A7) The biological computer system of clause Al, wherein the data encoded in the base-N electrical signals comprises a first representation of a digit, and the deconvolved data comprises a second representation of the digit.

[0269] (A8) The biological computer system of clause Al, wherein the synthetic biological circuit is configured to function as a multi-valued logic circuit.

[0270] (A9) The biological computer system of clause Al, wherein the synthetic biological circuit is configured to function as a biologic qubit.Docket No. 229303-701601 / PCT

[0271] (Al 0) The biological computer system of clause A8, wherein the biologic Qubit is configured to perform quantum computation.

[0272] (Al 1) The biological computer system of clause Al, wherein the synthetic biological circuit includes a plurality of neurons.

[0273] (A12) The biological computer system of clause Al 1, wherein the neurons of the plurality' of neurons are integrated in a confluent neuronal array.

[0274] (Al 3) The biological computer system of clause Al 2, wherein the confluent neuronal array is two-dimensional (2D).

[0275] (A 14) The biological computer system of clause A l 1, wherein the neurons of the plurality of neurons are integrated in a three-dimensional (3D) structure.

[0276] (A15) The biological computer system of clause A14, wherein the 3D structure comprises a brain organoid or a brain tissue.

[0277] (A16) The biological computer system of clause A15, wherein the brain organoid or the brain tissue comprises two or more cell types.

[0278] (Al 7) The biological computer system of clause Al 6, wherein the two or more cell types comprise cell types selected from the group consisting of: sensory neurons, motor neurons, interneurons, projection neurons, astrocytes, oligodendrocytes, pyramidal neurons, Purkinje cells, granule cells, microglial cells, radial glial cells, ependymal cells, endothelial cells, pericytes, choroid plexus epithelial cells, glutamatergic neurons, GAB / Xergic neurons, dopaminergic neurons, cholinergic neurons, serotonergic neurons, neural progenitor cells, adult neural stem cells, and any combination thereof.

[0279] (Al 8) The biological computer system of clause Al 5, wherein the brain organoid or the brain tissue comprises in vitro-differentiated human neurons, human induced-pluripotent stem cell derived cells, human embryonic stem cell derived cells, or a combination thereof.

[0280] (Al 9) The biological computer system of clause All, wherein one or more neurons in the plurality of neurons are genetically engineered.

[0281] (A20) The biological computer system of clause All, wherein one or more neurons in the plurality of neurons are genetically engineered to knock out one or more GABA receptors.

[0282] (A21) The biological computer system of clause All, wherein one or more neurons in the plurality of neurons are genetically engineered to knock out one or more glutamate receptors.Docket No. 229303-701601 / PCT

[0283] (A22) The biological computer system of clause A20 and A21, wherein the knock out comprises a genomic disruption in a target sequence of a gene encoding for the GABA receptor or a genomic disruption in a target sequence of a gene encoding for a glutamate receptor.

[0284] (A23) The biological computer system of clause A20 and A21, wherein the knock out comprises a CRISPR / Cas mediated genomic disruption in a target sequence of a gene.

[0285] (A24) The biological computer system of clause All, further comprising a solution, wherein the solution, the plurality of neurons, and a surface of the MEA are disposed in a chamber.

[0286] (A25) The biological computer system of clause A 10, wherein the interface controller 145 is communicatively coupled to an in silico computing device 110,

[0287] (A26) The biological computer system of clause A25, wherein the in silico computing device is configured to train the synthetic biological circuit to perform a computational task.

[0288] (A27) The biological computer system of clause A26, wherein training the synthetic biological circuit to perform the computational task includes: in a first phase, measuring, via the MEA, first neuronal activity of the plurality of neurons; in a second phase, stimulating the plurality of neurons, wherein stimulating the plurality of neurons includes providing, by the MEA, one or more input signals encoding input data for the computational task; in a third phase, measuring, via the MEA, second neuronal activity of the plurality of neurons responsive to the stimulating; and in a fourth phase, determining, based on the second neuronal activity, whether the synthetic biological circuit has performed the computational task satisfactorily, and if so, providing a first type of stimulus to the plurality of neurons via the MEA or providing no stimulus to the plurality of neurons via the MEA, and otherwise, providing a second type of stimulus to the plurality of neurons via the MEA.

[0289] (A28) The method of clause A27, wherein the first type of stimulus is deterministic and / or predictable.

[0290] (A29) The method of clause A28, wherein the second type of stimulus is stochastic, unpredictable, and / or punitive.

[0291] (A30) The biological computer system of clause Al, further including a first set of one or more synthetic biological circuits disposed on the MEA, wherein the first set of one or more synthetic biological circuits includes the synthetic biological circuit.Docket No. 229303-701601 / PCT

[0292] (A31) The biological computer system of clause A30, wherein the ME A is a first MEA, wherein the interface further comprises one or more second MEAs communicatively coupled to the interface controller, and wherein the system further includes one or more second sets of synthetic biological circuits, wherein the one or more second sets of synthetic biological circuits are disposed, respectively, on the one or more second MEAs.

[0293] (A32) The biological computer system of clause Al, wherein the interface controller 145 is further configured to encode data in base-2 electrical signals provided to the synthetic biological circuit by the MEA.

[0294] (Bl) A method comprising: providing, by a multi-electrode array (MEA) 130 to a synthetic biological circuit disposed on the MEA, one or more base-N electrical signals encoding data, where N > 2.

[0295] (B2) The method of clause Bl, further comprising: sensing, by the synthetic biological circuit, the base-N electrical signals; performing, by the synthetic biological circuit, a computational task based on the data encoded in the base-N electrical signals; and providing, by the synthetic biological circuit, one or more electrical signals encoding a result of the computational task.

[0296] (B3) The method of clause Bl, wherein the computational task comprises an ill-posed inverse task.

[0297] (B4) The method of clause B3, wherein the performing the ill-posed inverse task includes reconstructing an image based on the data, processing an image represented by the data, performing an inverse scattering of the data, and / or inverting a matrix represented by the data.

[0298] (B5) The method of clause B3, wherein performing the ill-posed inverse task includes deconvolving the data encoded in the base-N electrical signals, thereby producing deconvolved data.

[0299] (B6) The method of clause B5, wherein the data encoded in the base-N electrical signals comprises a first representation of a digit, and the deconvolved data comprises a second representation of the digit.

[0300] (B7) The method of clause B2, wherein the synthetic biological circuit comprises a biologic Qubit, and wherein the biologic Qubit performs the computational task using quantum computation.Docket No. 229303-701601 / PCT

[0301] (B8) The method of clause B2, further comprising: decoding, by an interface controller 145 communicatively coupled to the MEA, data encoded in the one or more electrical signals provided by the synthetic biological circuit.

[0302] (B9) The method of clause B2, wherein the synthetic biological circuit includes a plurality of neurons.

[0303] (B10) The method of clause B9, further comprising: training the synthetic biological circuit to perform the computational task, wherein the training is performed by a computing device communicatively coupled to an interface controller communicatively coupled to the MEA.

[0304] (Bl 1) The method of clause B10, wherein training the synthetic biological circuit to perform the computational task includes: in a first phase, measuring, via the MEA, first neuronal activity of the plurality of neurons; in a second phase, stimulating the plurality of neurons, wherein stimulating the plurality of neurons includes providing, by the MEA, one or more input signals encoding input data for an instance of the computational task; in a third phase, measuring, via the MEA, second neuronal activity of the plurality of neurons responsive to the stimulating; and in a fourth phase, determining, based on the second neuronal activity, whether the synthetic biological circuit has performed the instance of the computational task satisfactorily, and if so, providing a first type of stimulus to the plurality of neurons via the MEA or providing no stimulus to the plurality of neurons via the MEA, and otherwise, providing a second type of stimulus to the plurality of neurons via the MEA.

[0305] (Bl 2) The method of clause Bl 1, wherein the first type of stimulus is deterministic and / or predictable, and wherein the second type of stimulus is stochastic, unpredictable, and / or punitive.

[0306] (Cl) A biological computer system comprising: an interface 105 including a multi-electrode array (MEA) 130 and an interface controller 145 communicatively coupled to the MEA 130; and a synthetic biological circuit 135 disposed on the MEA 130 and configured to: perform an ill-posed inverse task based on data encoded in one or more signals provided by the MEA, and provide one or more signals encoding a result of the ill-posed inverse task.

[0307] (C2) The biological computer system of clause Cl, wherein performing the ill-posed inverse task includes reconstructing an image based on the data, processing an image representedDocket No. 229303-701601 / PCTby the data, performing an inverse scattering of the data, and / or inverting a matrix represented by the data.

[0308] (C3) The biological computer system of clause Cl, wherein the synthetic biological circuit is configured to perform the ill-posed inverse task using quantum computation.

[0309] (C4) The biological computer system of clause Cl, wherein performing the ill-posed inverse task includes deconvolving the data, thereby producing deconvolved data.

[0310] (C5) The biological computer system of clause C4, wherein the data encoded in the one or more signals comprises a first representation of a digit, and the deconvolved data comprises a second representation of the digit.

[0311] (C6) The biological computer system of clause Cl, wherein the one or more signals encoding the data include base-N electrical signals, wherein N > 2,

[0312] (C7) The biological computer system of clause Cl, wherein the synthetic biological circuit is configured to function as a biologic qubit.

[0313] (C8) The biological computer system of clause C7, wherein the biologic Qubit is configured to perform quantum computation.

[0314] (C9) The biological computer system of clause Cl, wherein the synthetic biological circuit includes a plurality of neurons.

[0315] (CIO) The biological computer system of clause C9, wherein the interface controller 145 is communicatively coupled to a computing device 110 configured to tram the synthetic biological circuit to perform the ill-posed inverse task.

[0316] (Cl 1) The biological computer system of clause CIO, wherein training the synthetic biological circuit to perform the ill-posed inverse task includes: in a first phase, measuring, via the MEA, first neuronal activity of the plurality of neurons; in a second phase, stimulating the plurality of neurons, wherein stimulating the plurality of neurons includes providing, by the MEA, one or more input signals encoding input data for the ill-posed inverse task; in a third phase, measuring, via the MEA, second neuronal activity of the plurality of neurons responsive to the stimulating; and in a fourth phase, determining, based on the second neuronal activity, whether the synthetic biological circuit has performed the ill-posed inverse task satisfactorily, and if so, providing a first type of stimulus to the plurality of neurons via the MEA or providing no stimulus to the plurality of neurons via the MEA, and otherwise, providing a second type of stimulus to the plurality of neurons via the MEA.Docket No. 229303-701601 / PCT

[0317] (C12) The method of clause Cl 1, wherein the first type of stimulus is deterministic and / or predictable, and wherein the second type of stimulus is stochastic, unpredictable, and / or punitive.

[0318] (DI) A method comprising: providing, by a multi-electrode array (MEA) 130 to a synthetic biological circuit disposed on the MEA, one or more signals encoding data; performing, by the synthetic biological circuit, an ill-posed inverse task based on the data encoded in the one or more signals provided by the MEA; and providing, by the synthetic biological circuit, one or more signals encoding a result of the ill-posed inverse task,

[0319] (D2) The method of clause DI, wherein the one or more signals encoding the data include one or more base-N electrical signals, where N > 2.

[0320] (D3) The method of clause DI, wherein performing the ill-posed inverse task includes: sensing, by the synthetic biological circuit, the one or more signals encoding the data,

[0321] (D4) The method of clause DI, wherein the performing the ill-posed inverse task includes reconstructing an image based on the data, processing an image represented by the data, performing an inverse scattering of the data, and / or inverting a matrix represented by the data.

[0322] (D5) The method of clause DI, wherein performing the ill-posed inverse task includes deconvolving the data encoded in the one or more signals, thereby producing deconvolved data.

[0323] (D6) The method of clause DI, wherein the synthetic biological circuit comprises a biologic Qubit, and wherein the biologic Qubit performs the ill-posed inverse task using quantum computation.

[0324] (D7) The method of clause DI, further comprising: sensing, by the MEA, the one or more signals encoding the result of the ill-posed inverse task; and decoding, by an interface controller communicatively coupled to the MEA, data encoded in the one or more signals sensed by the MEA.

[0325] (D8) The method of clause DI, wherein the synthetic biological circuit includes a plurality of neurons.

[0326] (D9) The method of clause D8, further comprising: training the synthetic biological circuit to perform the ill-posed inverse task, wherein the training is performed by a computing device communicatively coupled to an interface controller communicatively coupled to the MEA.Docket No. 229303-701601 / PCT

[0327] (El) A biological computer system comprising: an interface 105 including a multi¬ electrode array (MEA) 130 and an interface controller 145 communicatively coupled to the MEA 130; and a synthetic biological circuit 135 disposed on the MEA 130 and configured to perform a computational task based on data encoded in one or more signals provided by the MEA, wherein the synthetic biological circuit includes a plurality of neurons, and one or more neurons in the plurality of neurons are genetically engineered.

[0328] (E2) The biological computer system of clause El, wherein the one or more neurons are genetically engineered to knock out one or more GABA receptors and / or one or more glutamate receptors.

[0329] (E3) The biological computer system of clause El, wherein the one or more genetically engineered neurons comprise at least one genomic disruption in a target sequence of a gene.

[0330] (E4) The biological computer system of clause E3, wherein the at least one genomic disruption is in a target sequence of a gene selected from the group consisting of: GABRA1, GABR. A2, GABRA3, GABRA4, GABR. A5, GABRA6, GABRB1, GABRB2, GABRB3, GABRG1, GABRG2, GABRG3, GABRD, GABRE, GABRQ, GABRP, GABRRI, GABRR2, GABRR3, GABBRI, GABBR2, GRINI, GRIN2A, GRIN2B, GRIN2C, GRIN2D, GRIN3A, GRIN3B, GRIA1, GRIA2, GRIA3, GRIA4. GRIK1, GRIK2, GRIK3, GRIK4, GRIK5, GRM1, GRM2, GRM3, GRM4, GRM5, GRM6, GRM7, GRM8, DRD1, DRD5, DRD2, DRD3, DRD4, ADRA1A, ADRA1B, ADRA1D, ADRA2A, ADRA2B, ADRA2C, ADRB1, ADRB2, or ADRB3.

[0331] (E5) The biological computer system of clause El, wherein the genetic engineering of the one or more neurons increases an electrical conductivity of the one or more neurons, a firing rate of the one or more neurons, and / or a robustness of the one or more neurons to electrical current.

[0332] (E6) The biological computer system of clause El, wherein the computational task includes an ill-posed inverse task, and wherein performing the ill-posed inverse task includes providing, by the synthetic biological circuit, one or more signals encoding a result of the ill- posed inverse task.

[0333] (E7) The biological computer system of clause E6, wherein performing the ill-posed inverse task includes reconstructing an image based on the data, processing an image representedDocket No. 229303-701601 / PCTby the data, performing an inverse scattering of the data, and / or inverting a matrix represented by the data.

[0334] (E8) The biological computer system of clause E6, wherein performing the ill-posed inverse task includes deconvolving the data, thereby producing deconvolved data.

[0335] (E9) The biological computer system of clause E8, wherein the data encoded in the one or more signals comprises a first representation of a digit, and the deconvolved data comprises a second representation of the digit.

[0336] (El 0) The biological computer system of clause E6, wherein the synthetic biological circuit is configured to perform the ill-posed inverse task using quantum computation.

[0337] (El 1) The biological computer system of clause El, wherein the one or more signals encoding the data include base-N electrical signals, wherein N > 2,

[0338] (El 2) The biological computer system of clause El, wherein the synthetic biological circuit is configured to function as a biologic qubit.

[0339] (El 3) The biological computer system of clause E12, wherein the biologic Qubit is configured to perform quantum computation.

[0340] (E14) The biological computer system of clause El, wherein the synthetic biological circuit includes a plurality of neurons.

[0341] (E15) The biological computer system of clause E14, wherein the interface controller is communicatively coupled to a computing device configured to train the synthetic biological circuit to perform the computational task.

[0342] (El 6) The biological computer system of clause El 5, wherein training the synthetic biological circuit to perform the computational task includes: in a first phase, measuring, via the MEA, first neuronal activity of the plurality of neurons; in a second phase, stimulating the plurality of neurons, wherein stimulating the plurality of neurons includes providing, by the MEA, one or more input signals encoding input data for an instance of the computational task; in a third phase, measuring, via the ME A, second neuronal activity of the plurality of neurons responsive to the stimulating; and in a fourth phase, determining, based on the second neuronal activity, whether the synthetic biological circuit has performed the instance of the computational task satisfactorily, and if so, providing a first type of stimulus to the plurality of neurons via the MEA or providing no stimulus to the plurality of neurons via the MEA, and otherwise, providing a second type of stimulus to the plurality of neurons via the MEA.Docket No. 229303-701601 / PCT

[0343] (El 7) The method of clause El 6, wherein the first type of stimulus is deterministic and / or predictable, and wherein the second type of stimulus is stochastic, unpredictable, and / or punitive.

[0344] (Fl) A method comprising: providing, by a multi-electrode array (MEA) 130 to a synthetic biological circuit disposed on the MEA, one or more signals encoding data; performing, by the synthetic biological circuit, a computational task based on the data encoded in the one or more signals provided by the MEA; and providing, by the synthetic biological circuit, one or more signals encoding a result of the computational task, wherein the synthetic biological circuit includes a plurality of neurons, and one or more neurons in the plurality of neurons are genetically engineered,

[0345] (F2) The method of clause Fl, wherein the one or more signals encoding the data include one or more base-N electrical signals, where N > 2.

[0346] (F3) The method of clause Fl, wherein performing the computational task includes: sensing, by the synthetic biological circuit, the one or more signals encoding the data,

[0347] (F4) The method of clause Fl, wherein the performing the computational task includes performing an ill-posed inverse task based on the data.

[0348] (F5) The method of clause Fl, wherein the synthetic biological circuit comprises a biologic Qubit, and wherein the biologic Qubit performs the computational task using quantum computation.

[0349] (F6) The method of clause Fl, further comprising: sensing, by the MEA, the one or more signals encoding the result of the computational task; and decoding, by an interface controller communicatively coupled to the MEA, data encoded in the one or more signals sensed by the MEA.

[0350] (F7) The method of clause Fl, further comprising: training the synthetic biological circuit to perform the ill-posed inverse task, wherein the training is performed by a computing device communicatively coupled to an interface controller communicatively coupled to the MEA.

[0351] (Gl) A biological processing device comprising: a multi-electrode array (MEA) 130; and a synthetic biological circuit 135 disposed on the MEA 130, wherein the synthetic biological circuit is configured to decode data encoded in base-N electrical signals provided by the MEA,Docket No. 229303-701601 / PCTperform a computational task based on the decoded data, and provide one or more signals encoding a result of the computational task to the MEA, wherein N > 2.

[0352] (G2) The biological processing device of clause Gl, wherein the one or more signals encoding the result of the computational task include base-K electrical signals, wherein K > 2.

[0353] (G3) The biological processing device of clause Gl, wherein the performing the computational task includes performing an ill-posed inverse task.

[0354] (G4) The biological processing device of clause Gl, wherein the performing the computational task includes deconvolving the decoded data, thereby producing deconvolved data.

[0355] (G5) The biological processing device of clause Gl, wherein the synthetic biological circuit is configured to perform the computational task using quantum computation.

[0356] (G6) The biological processing device of clause Gl, wherein the computational task includes a multi-valued logic task.

[0357] (G7) The biological processing device of clause Gl, wherein the synthetic biological circuit is configured to function as a biologic qubit.

[0358] (G8) The biological processing device of clause G7, wherein the biologic Qubit is configured to perform quantum computation.

[0359] (G9) The biological processing device of clause Gl, wherein the synthetic biological circuit includes a plurality of neurons.

[0360] (Hl) A biological processing device comprising: a multi-electrode array (MEA) 130; and a synthetic biological circuit 135 disposed on the MEA 130, wherein the synthetic biological circuit is configured to perform an ill-posed inverse task based on data encoded in one or more signals provided by the MEA, and provide one or more signals encoding a result of the ill-posed inverse task to the MEA.

[0361] (H2) The biological processing device of clause Hl, wherein performing the ill-posed inverse task includes reconstructing an image based on the data, processing an image represented by the data, performing an inverse scattering of the data, and / or inverting a matrix represented by the data.

[0362] (H3) The biological processing device of clause Hl, wherein performing the ill-posed inverse task includes deconvolving the data, thereby producing deconvolved data.Docket No. 229303-701601 / PCT

[0363] (H4) The biological processing device of clause Hl, wherein the synthetic biological circuit is configured to perform the ill-posed inverse task using quantum computation.

[0364] (H5) The biological processing device of clause Hl, wherein the one or more signals encoding the data include base-N electrical signals, wherein N > 2.

[0365] (H6) The biological processing device of clause Hl, wherein the synthetic biological circuit is configured to function as a biologic qubit.

[0366] (H7) The biological processing device of clause H6, wherein the biologic Qubit is configured to perform quantum computation.

[0367] (H8) The biological processing device of clause Hl, wherein the synthetic biological circuit includes a plurality of neurons,

[0368] (II) A biological processing device comprising: a multi-electrode array (MEA) 130; and a synthetic biological circuit 135 disposed on the MEA 130, wherein the synthetic biological circuit includes a plurality of neurons, and one or more neurons in the plurality of neurons are genetically engineered, wherein the synthetic biological circuit is configured to perform a computational task based on data encoded in one or more signals provided by the MEA, and provide one or more signals encoding a result of the computation to the MEA.

[0369] (12) The biological processing device of clause 11, wherein the one or more neurons are genetically engineered to knock out one or more GABA receptors and / or one or more glutamate receptors.

[0370] (13) The biological processing device of clause II, wherein the one or more genetically engineered neurons comprise at least one genomic disruption m a target sequence of a gene.

[0371] (14) The biological processing device of clause 13, wherein the at least one genomic disruption is in a target sequence of a gene selected from the group consisting of: GABRA1, GABRA2, GABRA3, GABRA4, GABRA5, GABRA6, GABRB1, GABRB2, GABRB3, GABRG1, GABRG2, GABRG3, GABRD, GABRE, GABRQ, GABRP, GABRR1, GABRR2, GABRR3, GABBR1, GABBR2, GRIN1, GRIN2A, GRIN2B, GRIN2C, GRIN2D, GRIN3A, GRIN3B, GRIA1, GRIA2, GRIA3, GRIA4. GRIK1, GRIK2, GRIK3, GRIK4, GRIK5, GRM1, GRM2, GRM3, GRM4, GRM5, GRM6, GRM7, GRM8, DRD1, DRD5, DRD2, DRD3, DRD4, ADRA1A, ADRA1B, ADRA1D, ADRA2A, ADRA2B, ADRA2C, ADRB1, ADRB2, or ADRB3.Docket No. 229303-701601 / PCT

[0372] (15) The biological processing device of clause II, wherein the genetic engineering of the one or more neurons increases an electrical conductivity of the one or more neurons, a firing rate of the one or more neurons, and / or a robustness of the one or more neurons to electrical current.

[0373] (16) The biological processing device of clause II, wherein the computational task includes an ill-posed inverse task, and wherein performing the ill-posed inverse task includes providing, by the synthetic biological circuit, one or more signals encoding a result of the ill- posed inverse task.

[0374] (17) The biological processing device of clause 16, wherein performing the ill-posed inverse task includes reconstructing an image based on the data, processing an image represented by the data, performing an inverse scattering of the data, and / or in verting a matrix represented by the data,

[0375] (18) The biological processing device of clause 16, wherein performing the ill-posed inverse task includes deconvolving the data, thereby producing deconvolved data.

[0376] (19) The biological processing device of clause 16, wherein the synthetic biological circuit is configured to perform the ill-posed inverse task using quantum computation.

[0377] (110) The biological processing device of clause II, wherein the synthetic biological circuit is configured to function as a biologic Qubit, and wherein the biologic Qubit is configured to perform quantum computation.

[0378] (111) The biological processing device of clause II, wherein the one or more signals encoding the data include base-N electrical signals, wherein N > 2.

[0379] (112) The biological processing device of clause II, wherein the synthetic biological circuit includes a plurality of neurons.

[0380] JI. A method of fabricating a biological processing device, the method comprising: (a) culturing a population of neuronal cells, wherein the population of neuronal cells comprise at least one neuron cell subtype and at least one glial cell subtype; (b) dispersing and seeding a population of neuronal cells on a surface of a microelectrode array (MEA) comprising a plurality of microelectrodes, wherein the neuronal cell seeding density is at least 100,000 cells / cm2to 250,000 cells / cm2to form a confluent neuronal array, and wherein the population of neuronal cells form one or more synthetic biological circuits on the plurality of microelectrodes; and (c)Docket No. 229303-701601 / PCTcommunicatively coupling the MEA with an interface controller configured to encode data in base-N electrical signals provided to the confluent neuronal array by the MEA, wherein N > 2.

[0381] KI. A method of fabricating a biological processing device, the method comprising: (a) culturing a population of neuronal cells, wherein the population of neuronal cells comprise at least one genomic disruption in a target sequence of a gene; (b) dispersing and seeding a population of neuronal cells on a surface of a microelectrode array comprising a plurality of microelectrodes, wherein the neuronal cell seeding density is at least 100,000 cells / cm2to 250,000 cells / cm2to form a confluent neuronal array, and wherein the population of neuronal cells form one or more synthetic biological circuits on the plurality of microelectrodes; and (c) communicatively coupling the MEA with an interface controller configured to encode data in base-N electrical signals provided to the confluent neuronal array by the MEA, wherein N > 2,

[0382] K2. The method of claim KI, wherein the at least one genomic disruption is in a target sequence of a gene selected from the group consisting of: GABRA1, GABRA2, GABRA3, GABRA4, GABRA5, GABRA6, GABRB1, GABRB2, GABRB3, GABRG1, GABRG2, GABRG3, GABRD, GABRE, GABRQ, GABRP, GABRR1, GABRR2, GABRR3, GABBR1, GABBR2, GRIN1, GRIN2A, GRIN2B, GRIN2C, GRIN2D, GRIN3A, GRIN3B, GRIA1, GRIA2, GRIA3, GRIA4. GRIK1, GRIK2, GRIK3, GRIK4, GRIK5, GRM1, GRM2, GRM3, GRM4, GRM5, GRM6, GRM7, GRM8, DRD1, DRD5, DRD2, DRD3, DRD4, ADRA1A, ADRA1B, ADRA1D, ADRA2A, ADRA2B, ADRA2C, ADRB1, ADRB2, or ADRB3.

[0383] (LI) A biological computer system comprising: an interface 105 including a field- programmable signal transceiver (FPST) and an interface controller 145 communicatively coupled to the FPST 130; and a synthetic biological circuit 135 disposed on the FPST 130 and configured to: perform an inverse computational task based on data encoded in one or more signals provided by the FPST, and provide one or more signals encoding a result of the inverse computational task.

[0384] (L2) The biological computer system of clause LI, wherein performing the inverse computational task includes reconstructing an image based on the data, processing an image represented by the data, performing an inverse scattering of the data, and / or inverting a matrix represented by the data.

[0385] (L3) The biological computer system of clause LI, wherein the synthetic biological circuit is configured to perform the inverse computational task using quantum computation.Docket No. 229303-701601 / PCT

[0386] (L4) The biological computer system of clause LI, wherein performing the inverse computational task includes deconvolving the data, thereby producing deconvolved data.

[0387] (L5) The biological computer system of clause L4, wherein the data encoded in the one or more signals comprises a first representation of a digit, and the deconvolved data comprises a second representation of the digit.

[0388] (L6) The biological computer system of clause LI, wherein the one or more signals encoding the data include base-N electrical signals, wherein N > 2.

[0389] (L7) The biological computer system of clause LI, wherein the synthetic biological circuit is configured to function as a biologic qubit.

[0390] (L8) The biological computer system of clause L7, wherein the biologic qubit is configured to perform quantum computation.

[0391] (L9) The biological computer system of clause LI, wherein the synthetic biological circuit includes a plurality of neurons.

[0392] (L10) The biological computer system of clause L9, wherein the interface controller 145 is communicatively coupled to a computing device 110 configured to train the synthetic biological circuit to perform the inverse computational task.

[0393] (LI 1) The biological computer system of clause L10, wherein training the synthetic biological circuit to perform the inverse computational task includes: in a first phase, measuring, via the FPST, first activity of the plurality of neurons; in a second phase, applying signals to the plurality of neurons, wherein applying signals to the plurality of neurons includes providing, by the FPST, one or more input signals encoding input data for the inverse computational task; in a third phase, measuring, via the FPST, second activity of the plurality of neurons responsive to the input; and in a fourth phase, determining, based on the second activity, whether the synthetic biological circuit has performed the inverse computational task satisfactorily, and if so, providing a first type of input to the plurality of neurons via the FPST or providing no input to the plurality of neurons via the FPST, and otherwise, providing a second type of input to the plurality of neurons via the FPST

[0394] (L12) The biological computer system of clause Lil, wherein the first type of input is deterministic and / or predictable, and wherein the second type of input is stochastic, unpredictable, and / or punitive.Docket No. 229303-701601 / PCT

[0395] (LI 3) The biological computer system of clause LI, wherein the SBC includes a biological neural network.

[0396] (LI 4) A method comprising: providing, by a field- programmable signal transceiver (FPST) 130 to a synthetic biological circuit disposed on the FPST, one or more signals encoding data; performing, by the synthetic biological circuit, an inverse computational task based on the data encoded in the one or more signals provided by the FPST; and providing, by the synthetic biological circuit, one or more signals encoding a result of the inverse computational task.

[0397] (LI 5) The method of clause L14, wherein the one or more signals encoding the data include one or more base-N electrical signals, where N > 2.

[0398] (LI 6) The method of clause L14, wherein performing the inverse computational task includes: sensing, by the synthetic biological circuit, the one or more signals encoding the data.

[0399] (LI 7) The method of clause LI 4, wherein the performing the inverse computational task includes reconstructing an image based on the data, processing an image represented by the data, performing an inverse scattering of the data, and / or inverting a matrix represented by the data.

[0400] (LI 8) The method of clause L14, wherein performing the inverse computational task includes deconvolving the data encoded in the one or more signals, thereby producing deconvolved data.

[0401] (LI 9) The method of clause LI 4, wherein the synthetic biological circuit comprises a biologic Qubit, and wherein the biologic qubit performs the inverse computational task using quantum computation.

[0402] (L20) The method of clause L14, further comprising: sensing, by the FPS T, the one or more signals encoding the result of the inverse computational task; and decoding, by an interface controller communicatively coupled to the FPST, data encoded in the one or more signals sensed by the FPST.

[0403] (L21) The method of clause L14, wherein the synthetic biological circuit includes a plurality' of neurons.

[0404] (L22) The method of clause L21, further comprising: training the synthetic biological circuit to perform the inverse computational task, wherein the training is performed by a computing device communicatively coupled to an interface controller communicatively coupled to the FPST.Docket No. 229303-701601 / PCTReferences

[0405] [1] S B. Jo et al., Recent advance on multivalued logic gates: A materials perspective. Advanced Science 2021, 8, 2004216 (pp. 1-20), February’ 26, 2021.

[0406] [2] S. J. Basha et al., High performance quaternary' logic designs using GNFETs. ePrime - Advances in Electrical Engineering, Electronics and Energy 5 (2023) 100197 (June 17, 2023), 8 pages.

[0407] [3] Z. T. Sandhie et al,, Investigation of Multiple-valued Logic Technologies for Beyond-binary Era. ACM Comput. Surv. 54, 1, Article 16 (January' 2021), 30 pages,

[0408] [4] Adams, B, et al. Entanglement and coherence in pure and doped Posner molecules. Scientific Reports, 15, 12559. https: / / doi.org / 10.1038 / s41598-025-96487-5 (2025),

[0409] [5] Player, T. C, et al., Posner qubits: Spin dynamics of entangled Ca₉(PO₄)₆ molecules and their role in neural processing. Journal of the Royal Society Interface, 15(20180494). https: / / doi.org / 10.1098 / rsif.2018.0494 (2018).

[0410] [6] Straub, J. S., et al., Evidence for a possible quantum effect on the formation of lithi um-doped amorphous calcium phosphate from solution. Proceedings of the National Academy of Sciences, 122(10), e2423211122. https: / / doi.org / 10.1073 / pnas.2423211122 (2025).

[0411] [7] Fisher, M. P. A. Quantum cognition: The possibility of processing with nuclear spins in the brain. Annals of Physics, 362, 593-602. https: / / doi.org / 10.1016 / j.aop.2015.08.020 (2015).

[0412] [8] Adams, B. et al., Spin quantum computing, spin quantum cognition. arXiv preprint, https: / / arxiv. org / abs / 2510.07196 (2025).

[0413] [9] Agarwal, S. et al., The biological qubit: Calcium phosphate dimers, not trimers. The Journal of Physical Chemistry Letters, 14(10), 2518-2525https: / / doi.org / 10.1021 / acs.jpclet.3c00234 (2023).

Claims

Docket No. 229303-701601 / PCTCLAIMSWhat is claimed is:

1. A biological computer system comprising:an interface 105 including a multi-electrode array (MEA) 130 and an interface controller 145 communicatively coupled to the MEA 130; anda synthetic biological circuit 135 disposed on the MEA 130,wherein the interface controller 145 is configured to encode data in base-N electrical signals provided to the synthetic biological circuit by the MEA, wherein N > 2,2. The biological computer system of claim 1, wherein the interface controller 145 is configured to decode data encoded in base-K electrical signals provided by the synthetic biological circuit and sensed by the MEA 130, wherein K > 2.

3. The biological computer system of claim 1, wherein the synthetic biological circuit is configured to sense the base-N electrical signals, perform an ill-posed inverse task based on the data encoded in the base-N electrical signals, and provide base-K electrical signals encoding a result of the ill-posed inverse task, wherein K > 2.

4. The biological computer system of claim 3, wherein the ill-posed inverse task is an image reconstruction task, an image processing task, an inverse scattering task, or a matrix inversion task.

5. The biological computer system of claim 4, wherein the synthetic biological circuit is configured to perform the ill-posed inverse task using quantum computation.

6. The biological computer system of claim 1, wherein the synthetic biological circuit is configured to sense the base-N electrical signals, deconvolve the data encoded in the base-N electrical signals, and provide base-K electrical signals encoding the deconvolved data, wherein K > 2.Docket No. 229303-701601 / PCT7. The biological computer system of claim 1, wherein the data encoded in the base-N electrical signals comprises a first representation of a digit, and the deconvolved data comprises a second representation of the digit.

8. The biological computer system of claim 1, w’herein the synthetic biological circuit is configured to function as a multi-valued logic circuit.

9. The biological computer system of claim 1, wherein the synthetic biological circuit is configured to function as a biologic Qubit.

10. The biological computer system of claim 8, wherein the biologic Qubit is configured to perform quantum computation.

11. The biological computer system of claim 1, wherein the synthetic biological circuit includes a plurality of neurons.

12. The biological computer system of claim 11, wherein the neurons of the plurality of neurons are integrated in a confluent neuronal array.

13. The biological computer system of claim 12, wherein the confluent neuronal array is two-dimensional (2D).

14. The biological computer system of claim 11, wherein the neurons of the plurality of neurons are integrated in a three-dimensional (3D) structure.

15. The biological computer system of claim 14, wherein the 3D structure comprises a brain organoid or a brain tissue.

16. The biological computer system of claim 15, wherein the brain organoid or the brain tissue comprises two or more cell types.Docket No. 229303-701601 / PCT17. The biological computer system of claim 16, wherein the two or more cell types comprise cell types selected from the group consisting of: sensory neurons, motor neurons, interneurons, projection neurons, astrocytes, oligodendrocytes, pyramidal neurons, Purkinje cells, granule cells, microglial cells, radial glial cells, ependymal cells, endothelial cells, pericytes, choroid plexus epithelial cells, glutamatergic neurons, GABAergic neurons, dopaminergic neurons, cholinergic neurons, serotonergic neurons, neural progenitor cells, adult neural stem cells, and any combination thereof.

18. The biological computer system of claim 15, wherein the brain organoid or the brain tissue comprises in vitro-differentiated human neurons, human induced-pluripotent stem cell derived cells, human embryonic stem cell derived cells, or a combination thereof,19. The biological computer system of claim 11, wherein one or more neurons in the plurality of neurons are geneti cally engineered.

20. The biological computer system of claim 11, wherein one or more neurons in the plurality of neurons are genetically engineered to knock out one or more GABA receptors.

21. The biological computer system of claim 11, wherein one or more neurons in the plurality of neurons are genetically engineered to knock out one or more glutamate receptors.

22. The biological computer system of claim 20 and 21, wherein the knock out comprises a genomic disruption in a target sequence of a gene encoding for the GABA receptor or a genomic disruption in a target sequence of a gene encoding for a glutamate receptor.

23. The biological computer system of claim 20 and 21, wherein the knock out comprises a CRISPR / Cas mediated genomic disruption in a target sequence of a gene.

24. The biological computer system of claim 11, further comprising a solution, w’herein the solution, the plurality of neurons, and a surface of the MEA are disposed in a chamber.Docket No. 229303-701601 / PCT25. The biological computer system of claim 10, wherein the interface controller 145 is communicatively coupled to an in silico computing device 110.

26. The biological computer system of claim 25, wherein the in silico computing device is configured to train the synthetic biological circuit to perform a computational task.

27. The biological computer system of claim 26, wherein training the synthetic biological circuit to perform the computational task includes:in a first phase, measuring, via the MEA, first neuronal activity of the plurality of neurons;in a second phase, stimulating the plurality of neurons, wherein stimulating the plurality of neurons includes providing, by the MEA, one or more input signals encoding input data for the computational task;in a third phase, measuring, via the MEA, second neuronal activity of the plurality of neurons responsive to the stimulating; andin a fourth phase,determining, based on the second neuronal activity, whether the synthetic biological circuit has performed the computational task satisfactorily, and if so, providing a first type of stimulus to the plurality of neurons via the MEA or providing no stimulus to the plurality of neurons via the MEA, andotherwise, providing a second type of stimulus to the plurality of neurons via the MEA.

28. The method of claim 27, wherein the first type of stimulus is deterministic and / or predictable.

29. The method of claim 28, wherein the second type of stimulus is stochastic, unpredictable, and / or punitive.Docket No. 229303-701601 / PCT30. The biological computer system of claim 1, further including a first set of one or more synthetic biological circuits disposed on the MEA, wherein the first set of one or more synthetic biological circuits includes the synthetic biological circuit.

31. The biological computer system of claim 30, wherein the MEA is a first MEA, wherein the interface further comprises one or more second MEAs communicatively coupled to the interface controller, and wherein the system further includes one or more second sets of synthetic biological circuits, wherein the one or more second sets of synthetic biological circuits are disposed, respectively, on the one or more second MEAs.

32. The biological computer system of claim 1, wherein the interface controller 145 is further configured to encode data in base-2 electrical signals provided to the synthetic biological circuit by the MEA.

Images