Cytomorphic Electronic Systems: A review and perspective

Beahm, Douglas Raymond; Deng, Yijie; Riley, Tanner G.; Sarpeshkar, Rahul

doi:10.1109/mnano.2021.3113192

Cited by 5 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The computation time of the cytomorphic chips is independent of the reaction network’s size and complexity even for stochastic simulations, which cannot be parallelized to represent Poisson noise in digital simulations ( Kim et al, 2018 ). For example, models of computation time can be significantly reduced by several orders of magnitude for circuit models including oscillatory repressilators and more complex networks in cancer ( Woo et al, 2015 ; Teo and Sarpeshkar, 2020 ) or in SARS-COV2 infection ( Beahm et al, 2021 ).…”

Section: Discussionmentioning

confidence: 99%

“…The overall mechanism of the simulation is that, given some preset parameters of a circuit, the voltage and current at any node of the circuit at any time are readily available upon simulation; these voltages and currents exactly represent the corresponding molecular concentrations and molecular reaction flux rates, respectively. For example, we have used electronic circuits to model and simulate complex biological processes including genetic circuits in synthetic biology ( Daniel et al, 2013 ; Teo et al, 2015 ; Zeng et al, 2018 ); kinetics of microbial growth and energetics ( Deng et al, 2021 ); tissue homeostasis ( Teo et al, 2019b ); and virus-host interactions ( Beahm et al, 2021 ). However, there are gaps in biologists’ understanding of electronic circuits and the underlying mathematics; and, in their understanding of the analogy of circuit variables to reaction kinetic parameters.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Rapid modeling of experimental molecular kinetics with simple electronic circuits instead of with complex differential equations

Deng

Beahm²,

Ran³

et al. 2022

Front. Bioeng. Biotechnol.

Self Cite

View full text Add to dashboard Cite

Kinetic modeling has relied on using a tedious number of mathematical equations to describe molecular kinetics in interacting reactions. The long list of differential equations with associated abstract variables and parameters inevitably hinders readers’ easy understanding of the models. However, the mathematical equations describing the kinetics of biochemical reactions can be exactly mapped to the dynamics of voltages and currents in simple electronic circuits wherein voltages represent molecular concentrations and currents represent molecular fluxes. For example, we theoretically derive and experimentally verify accurate circuit models for Michaelis-Menten kinetics. Then, we show that such circuit models can be scaled via simple wiring among circuit motifs to represent more and arbitrarily complex reactions. Hence, we can directly map reaction networks to equivalent circuit schematics in a rapid, quantitatively accurate, and intuitive fashion without needing mathematical equations. We verify experimentally that these circuit models are quantitatively accurate. Examples include 1) different mechanisms of competitive, noncompetitive, uncompetitive, and mixed enzyme inhibition, important for understanding pharmacokinetics; 2) product-feedback inhibition, common in biochemistry; 3) reversible reactions; 4) multi-substrate enzymatic reactions, both important in many metabolic pathways; and 5) translation and transcription dynamics in a cell-free system, which brings insight into the functioning of all gene-protein networks. We envision that circuit modeling and simulation could become a powerful scientific communication language and tool for quantitative studies of kinetics in biology and related fields.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Rapid modeling of experimental molecular kinetics with simple electronic circuits instead of with complex differential equations

Deng

Beahm²,

Ran³

et al. 2022

Front. Bioeng. Biotechnol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Prior work has developed mechanistically detailed models of COVID-19 viral replication with specific viral protein and molecule species [53] and host lung and immune system responses with several specific cellular subprocesses [49] and signaling pathways [50]. However, especially with a novel disease, creating realistic dynamic behavior using many cellular, molecular, and protein species-specific parameters can be prohibitive.…”

Section: Circuit Model Rationale and Developmentmentioning

confidence: 99%

“…For example, the transcriptional and translational energy costs to a host cell to produce one virion could be calculated and drugs could be selected that preferentially target metabolically stressed cells. Furthermore, the software-designed-and-run model presented here could be ported to cytomorphic computing chips [1], [4], [53], [88]- [92], allowing for high-speed iterations of drug cocktail formulations, which could be optimized by machine learning. Such computational techniques may prove faster than empirical methods at identifying promising drug cocktails for clinical applications.…”

Section: Cocktails Against Cytokine Storm and Autoimmunitymentioning

confidence: 99%

Drug Cocktail Formulation via Circuit Design

Beahm

Deng

DeAngelo

et al. 2023

IEEE Trans. Mol. Biol. Multi-Scale Commun.

View full text Add to dashboard Cite

Electronic circuits intuitively visualize and quantitatively simulate biological systems with nonlinear differential equations that exhibit complicated dynamics. Drug cocktail therapies are a powerful tool against diseases that exhibit such dynamics. We show that just six key states, which are represented in a feedback circuit, enable drug-cocktail formulation: 1) healthy cell number; 2) infected cell number; 3) extracellular pathogen number; 4) intracellular pathogenic molecule number; 5) innate immune system strength; and 6) adaptive immune system strength. To enable drug cocktail formulation, the model represents the effects of the drugs in the circuit. For example, a nonlinear feedback circuit model fits measured clinical data, represents cytokine storm and adaptive autoimmune behavior, and accounts for age, sex, and variant effects for SARS-CoV-2 with few free parameters. The latter circuit model provided three quantitative insights on the optimal timing and dosage of drug components in a cocktail: 1) antipathogenic drugs should be given early in the infection, but immunosuppressant timing involves a tradeoff between controlling pathogen load and mitigating inflammation; 2) both within and across-class combinations of drugs have synergistic effects; 3) if they are administered sufficiently early in the infection, anti-pathogenic drugs are more effective at mitigating autoimmune behavior than immunosuppressant drugs.

show abstract

“…There have been several attempts to implement hardware specialized for parallel simulation of stochastic biochemical reactions in order to solve the problem of long simulation times in software [14][15][16][17][18][19][20][21][22][23][24]. Ref.…”

Section: Introductionmentioning

confidence: 99%

A Poisson Process Generator Based on Multiple Thermal Noise Amplifiers for Parallel Stochastic Simulation of Biochemical Reactions

Mun

Jeong

et al. 2022

Electronics

View full text Add to dashboard Cite

In this paper, we propose a novel Poisson process generator that uses multiple thermal noise amplifiers (TNAs) as a source of randomness and controls its event rate via a frequency-locked loop (FLL). The increase in the number of TNAs extends the effective bandwidth of amplified thermal noise and hence enhances the maximum event rate the proposed architecture can generate. Verilog-A simulation of the proposed Poisson process generator shows that its maximum event rate can be increased by a factor of 26.5 when the number of TNAs increases from 1 to 10. In order to realize parallel stochastic simulations of the biochemical reaction network, we present a fundamental reaction building block with continuous-time multiplication and addition using an AND gate and a 1-bit current-steering digital-to-analog converter, respectively. Stochastic biochemical reactions consisting of the fundamental reaction building blocks are simulated in Verilog-A, demonstrating that the simulation results are consistent with those of conventional Gillespie algorithm. An increase in the number of TNAs to accelerate the Poisson events and the use of digital AND gates for robust reaction rate calculations allow for faster and more accurate stochastic simulations of biochemical reactions than previous parallel stochastic simulators.

show abstract

Cytomorphic Electronic Systems: A review and perspective

Cited by 5 publications

References 20 publications

Rapid modeling of experimental molecular kinetics with simple electronic circuits instead of with complex differential equations

Rapid modeling of experimental molecular kinetics with simple electronic circuits instead of with complex differential equations

Drug Cocktail Formulation via Circuit Design

A Poisson Process Generator Based on Multiple Thermal Noise Amplifiers for Parallel Stochastic Simulation of Biochemical Reactions

Contact Info

Product

Resources

About