Top PDF Programming Chemical Kinetics: Engineering Dynamic Reaction Networks with DNA Strand Displacement

Programming Chemical Kinetics: Engineering Dynamic Reaction Networks with DNA Strand Displacement

Programming Chemical Kinetics: Engineering Dynamic Reaction Networks with DNA Strand Displacement

Chapter 2 presents our work on the “device physics” of toehold-mediated DNA strand dis- placement. Even though DNA strand displacement has been a major workhorse of dynamic DNA nanotechnology, the biophysics and molecular mechanisms underlying strand displacement had not been studied in detail with a view to explaining strand displacement kinetics. In particular, state of the art models of strand displacement biophysics predicted a blunt-end strand displace- ment rate about 3 orders of magnitude faster than experimental measurements. We resolve this discrepancy and present a unified view of strand displacement biophysics and kinetics by study- ing the process at multiple levels of detail, using an intuitive model of a random walk on a 1- dimensional energy landscape, a secondary structure kinetics model with single base-pair steps, and a coarse-grained molecular model that incorporates three-dimensional geometric and steric effects. Further, we experimentally investigate the thermodynamics of 3-way branch migration. Two factors explain the dependence of strand displacement kinetics on toehold length: (i) the physical process by which a single step of branch migration occurs is significantly slower than the fraying of a single base pair, and (ii) initiating branch migration incurs a thermodynamic penalty, not captured by state-of-the-art nearest neighbor models of DNA, due to the additional overhang it engenders at the junction. Our findings are consistent with previously measured or inferred rates for hybridization, fraying, and branch migration, and provide a biophysical explanation of strand displacement kinetics. Our work paves the way for accurate modeling of strand displace- ment cascades, which would facilitate the simulation and construction of more complex molecular systems. Chapter 2 is based on our published research manuscript: Srinivas et al. [148]. My own personal contributions to this work are listed in Chapter 2.
Show more

245 Read more

Programming dynamic nonlinear biomolecular devices using DNA strand displacement reactions

Programming dynamic nonlinear biomolecular devices using DNA strand displacement reactions

A precise overview on the background theory of CRNs, the computational methodology and DSD mechanism is given in Chapter 2. The underlying DSD mechanism of four chemical reactions, viz., catalysis, bimolecular, degradation, and annihilation, that are considered in this thesis, is illustrated using the software tool Visual DSD. Chapter 3 shows results on how chemical reactions can be used to design and implement a number of nonlinear system theoretic operators, thus signif- icantly extending the results obtained in [31] that considered only linear systems. It is also shown how polynomial functions, rational functions and power components can be implemented by using a combination of the aforementioned four types of chemical reactions. Based on this, the new results are highlighted through three ap- plications, namely, computation of (1) fractional exponent, (2) absolute value, and (3) logarithm of arbitrary base. In Chapter 4, an important class of nonlinear con- trollers is realised and implemented in a closed-loop feedback system as a reference tracking problem. The design exploits bimolecular as well as unimolecular chem- ical reactions, allowing the implementation of highly nonlinear synthetic control circuits based on sliding mode control theory. Simulation results for the closed- loop response indicate that, compared to a traditional proportional+integrator (PI) controller, the implemented quasi sliding mode (QSM) controller results in dra- matically faster performance with more accurate tracking of reference signals, as well as providing a more modular approach that is less affected by the presence of retroactivity. The proposed design fully exploits the inherently nonlinear nature of biomolecular reaction kinetics, and makes for the first time a direct link between the biological concept of ultrasensitivity and the engineering theory of sliding mode control.
Show more

125 Read more

Implementing nonlinear feedback controllers using DNA strand displacement reactions

Implementing nonlinear feedback controllers using DNA strand displacement reactions

Abstract—We show how an important class of nonlinear feedback controllers can be designed using idealized abstract chemical reactions and implemented via DNA strand displace- ment (DSD) reactions. Exploiting chemical reaction networks (CRNs) as a programming language for the design of complex circuits and networks, we show how a set of unimolecular and bimolecular reactions can be used to realize input-output dynam- ics that produce a nonlinear quasi sliding mode (QSM) feedback controller. The kinetics of the required chemical reactions can then be implemented as enzyme-free, enthalpy/entropy driven DNA reactions using a toehold mediated strand displacement mechanism via Watson-Crick base pairing and branch migration. We demonstrate that the closed loop response of the nonlinear QSM controller outperforms a traditional linear controller by facilitating much faster tracking response dynamics without introducing overshoots in the transient response. The resulting controller is highly modular and is less affected by retroactivity effects than standard linear designs.
Show more

14 Read more

Implementing nonlinear feedback controllers using DNA strand displacement reactions

Implementing nonlinear feedback controllers using DNA strand displacement reactions

Abstract—We show how an important class of nonlinear feedback controllers can be designed using idealized abstract chemical reactions and implemented via DNA strand displace- ment (DSD) reactions. Exploiting chemical reaction networks (CRNs) as a programming language for the design of complex circuits and networks, we show how a set of unimolecular and bimolecular reactions can be used to realize input-output dynam- ics that produce a nonlinear quasi sliding mode (QSM) feedback controller. The kinetics of the required chemical reactions can then be implemented as enzyme-free, enthalpy/entropy driven DNA reactions using a toehold mediated strand displacement mechanism via Watson-Crick base pairing and branch migration. We demonstrate that the closed loop response of the nonlinear QSM controller outperforms a traditional linear controller by facilitating much faster tracking response dynamics without introducing overshoots in the transient response. The resulting controller is highly modular and is less affected by retroactivity effects than standard linear designs.
Show more

14 Read more

Improved computation of natural logarithm using chemical reaction networks

Improved computation of natural logarithm using chemical reaction networks

An objective of synthetic biology is to design biomolecular circuits for in situ monitoring and control. Recently, nucleic acid reactions have been proposed as a potential solution for these purposes [1 – 4]. A key advantage of nucleic acid reactions consists in the ease and precision with which these can be implemented, as their design relies essentially on the well-known Watson-Crick base-pairing mechanism (i.e. adenine-thymine and guanine-cytosine pairing), which enables precise programming and timing of molecular interactions simply by the choice of relevant sequences. This approach has allowed the implementation of a number of complex circuits based on DNA strand displacement [5], DNA enzyme [6] and RNA enzyme [7] reactions, and has been used for the modelling and implementation of various nucleic-acids-based circuits such as feedback controllers [8] and predator-prey systems [9]. Recently, it has been shown that any chemical reaction network can be closely approximated by a set of suitably designed DNA strand displacement reactions [10]. This logic can be extended to approximate a set of linear ordinary differential equations (ODEs) by a set of idealised abstract chemical reaction networks (ACRNs) which can then be approximated by a set of suitably designed DNA strand displacement reactions [4].
Show more

7 Read more

KINETIC CONTROL OF NUCLEIC ACID STRAND DISPLACEMENT REACTIONS

KINETIC CONTROL OF NUCLEIC ACID STRAND DISPLACEMENT REACTIONS

The kinetics of incorporating LNA nucleotides into a DNA strand displacement system has been studied. LNA substitutions affect the kinetics of strand displacement in three ways. First, LNAs in the Substrates stablize the duplex probably by reducing the fraying frequency at the terminus of the duplex regions, which lowers the probabilty of successful nucleation between the Substrates and zero toehold invaders. Second, LNAs in the Substrates induce B-form to A-form structural changes, which may hinder the branch migration process. Third, LNAs in the Substrate or the invaders bias random walks during branch migration – which alters the probability of strand displacement to proceed. When incorporating LNA substitutions into a DNA strand displacement system, the leakage rate was reduced up to 50 fold and the invasion rate was maintained elevated. In comparison, kinetics for hybrid DNA/LNA systems, the performance enhancement ratio can be improved by a factor of 70 – providing insights for how to design future high performance chemical reaction networks made from DNA and LNA. By site-specifically substituting LNA nucleotides for DNA nucleotides, while maintaining the original
Show more

134 Read more

Biologically inspired design of feedback control systems implemented using DNA strand displacement reactions

Biologically inspired design of feedback control systems implemented using DNA strand displacement reactions

An emerging design framework that uses abstract chemical reaction networks (CRNs) in the implementation of enzyme- free, entropy driven DNA reactions has recently attracted much attention in the Synthetic Biology community follow- ing a number of successful studies [1]-[3]. The basic idea un- derlying this approach is to design biomolecular circuitry us- ing abstract chemical reactions as a programming language. The designed biomolecular circuity can then be implemented directly in DNA utilising the DNA strand displacement (DSD) method [4]. Through the well-known Watson-Crick base-pairing mechanism (i.e. adenine-thymine and guanine- cytosine), the selection of appropriate DNA sequences allows precise control over the dynamics of the implemented DNA reactions, thus facilitating a precise molecular programming of the desired function, operator or circuit. Sophisticated CAD tools are also now becoming available to facilitate the design of synthetic circuits using this approach [5]. Examples of complex biomolecular circuits successfully designed and implemented through this approach include predator-prey dynamics [6], oscillators [7], and both linear and nonlinear feedback controllers [3], [8], [9].
Show more

5 Read more

Dynamic DNA strand displacement circuits

Dynamic DNA strand displacement circuits

Engineered nucleic acid circuits could also be used to modulate gene expression for synthetic biological applications. While the reaction networks built so far operate in a cell-free environment, it is likely that strand-displacement based reactions networks can be adapted to work inside cells, detecting, analyzing and changing the levels of various cellular RNAs. This research will also benefit from complementary approaches currently taken in RNA synthetic biology [93–95]. Smart therapeutics applications such as those suggested in Refs [33, 36] are a particularly promising area. For this, sensors and logic circuits similar to those already available need to be integrated with molecular actuators based on antisense oligonucleotides, siRNA or ribozymes. First steps in this direction have been taken [96, 97]. In vivo operation may benefit from the use of chemically modified nucleic acids such as LNA, PNA, or 2’-O-Methyl RNA to increase nuclease stability or to modify thermodynamic properties. Such modified nucleic acids may also be of interest for cell free application as sensors or simply as an additional degree of freedom for controlling thermodynamic stability and reaction kinetics of various systems. Recent research on expanded nucleic acid alpha- bets [98] could likewise form the basis for bio-orthogonal embedded computation systems. The strand displacement reactions and systems described in this review are expected to function qualitatively similarly for these other nucleic acids [99].
Show more

290 Read more

Accurate Ratio Computation using Abstract Chemical Reaction Networks

Accurate Ratio Computation using Abstract Chemical Reaction Networks

An objective of synthetic biology is to design biomolecular circuits for in situ monitoring and control. Recently, nucleic acid reactions have been proposed as a potential solution for these purposes [1], [2], [3], [4]. A key advantage of nucleic acid reactions consists in the ease and precision with which these can be implemented, as their design relies essentially on the well-known Watson-Crick base-pairing mechanism (i.e. adenine-thymine and guanine-cytosine pairing), which enables precise programming and timing of molecular in- teractions simply by the choice of relevant sequences. This approach has allowed the implementation of a number of complex circuits based on DNA strand displacement [5], DNA enzyme [6] and RNA enzyme [7] reactions, and has been used for the modelling and implementation of various nucleic-acids-based circuits such as feedback controllers [8] and predator-prey systems [9]. Recently, it has been shown that any chemical reaction network can be closely approxi- mated by a set of suitably designed DNA strand displacement reactions [10]. This logic can be extended to approximate a set of linear ordinary differential equations (ODEs) by a set of idealised abstract chemical reaction networks (ACRNs) which can then be approximated by a set of suitably designed DNA strand displacement reactions [4]. In [11], it was shown that ACRNs for catalysis using two reactants can be used to realise a set of nonlinear operators and, in particular, was used to implement a system S D that computes the ratio
Show more

6 Read more

Efficient initiation and strand transfer of polypurine tract-primed plus-strand DNA prevent strand transfer of internally initiated plus-strand DNA.

Efficient initiation and strand transfer of polypurine tract-primed plus-strand DNA prevent strand transfer of internally initiated plus-strand DNA.

FIG. 5. Analysis of EH3 proviral structures. (A) Two proviral structures can potentially form after infection with EH3. Strand transfer of 39 PPT-primed plus-strand DNA results in the normal proviral structure (39 PPT). Strand trans- fer of the 59 PPT-primed plus-strand DNA results in an aberrant proviral struc- ture (59 PPT). In the aberrant proviral structure, DNA sequences between the 59 PPT and the 39 LTR are duplicated upstream of the 59 LTR (bracket labeled 2.56 kb). NcoI-plus-NotI double digestion of both proviruses results in a 1.85-kb fragment and a 2.8-kb fragment. NcoI restriction digestion of both proviruses results in the 2.8-kb fragment. In addition, plus-strand transfer initiating at the 59 PPT results in a 2.0-kb NcoI-to-NcoI fragment. Abbreviations are the same as in Fig. 2. (B) Southern hybridization analysis of EH3 proviral structures. Viruses were harvested from an EH3-infected helper cell clone and used to infect D17 cells. Genomic DNA was isolated from a pool of more than 10,000 EH3-infected G418-resistant D17 colonies. The absence of the 2.0-kb NcoI-to-NcoI band indicates that plus-strand transfer of DNA initiated at the 5 9 PPT is not detect- able. Each lane contains 30 m g of genomic DNA digested by the indicated enzyme(s), except for the lanes labeled 1/5 amt. and 1/10 amt., which contain 6 and 3 m g of genomic DNA, respectively.
Show more

8 Read more

Automated Feature Engineering for Deep Neural Networks with Genetic Programming

Automated Feature Engineering for Deep Neural Networks with Genetic Programming

There is currently no automated means of engineering features specifically designed for deep neural networks that are a combination of multiple named features from the original vector. Previous automated feature engineering (AFE) work focused primarily upon the transformation of a single feature or upon models other than deep learning (Box & Cox, 1964; Breiman & Friedman, 1985; Freeman & Tukey, 1950). Although model- agnostic genetic programming-based feature extraction algorithms have been proposed (Guo, Jack, & Nandi, 2005; Neshatian, 2010), they do not tailor engineered features to deep neural networks. Feature engineering research for deep learning has primarily dealt with high-dimensional image and audio data (Blei, Ng, & Jordan, 2003; M. Brown & Lowe, 2003; Coates, Lee, & Ng, 2011; Coates & Ng, 2012; Le, 2013; Lowe, 1999; Scott & Matwin, 1999).
Show more

202 Read more

Visual detection of glial cell line-derived neurotrophic factor based on a molecular translator and isothermal strand-displacement polymerization reaction

Visual detection of glial cell line-derived neurotrophic factor based on a molecular translator and isothermal strand-displacement polymerization reaction

reaction between preimmobilized streptavidin and biotin. The accumulation of AuNPs on the test zone is then visual- ized as a characteristic red band. Excess antidigoxin–AuNP conjugates continue to migrate and are captured at the control zone by immunoreactions between antidigoxin antibody and the preimmobilized secondary antibody on the AuNP surface, thus forming a second red band. In the absence of GDNF, no ISDPR duplex DNA product is produced; therefore, no red band is observed at the test zone. In this case, the red band at the control zone indicates that the LFB is working properly. Visual detection is performed by observing the color change caused by the accumulation of AuNPs on the test zone, and quantitative detection can be realized by recording the color intensity of the red band on the test zone with a portable LFB reader (Figure 1C).
Show more

8 Read more

Dynamic Project and Workflow Management for Design Processes in Chemical Engineering

Dynamic Project and Workflow Management for Design Processes in Chemical Engineering

At the start of the design process, the manager creates an initial task net comprising only the task Preparation and its refinements. In the preparation phase, the requirements to the chemical plant are defined, a literature research is carried out, and an initial abstract flow diagram is cre- ated. Based on this information, the alternatives batch and continuous operation are compared, and a decision is performed towards continuous operation. The result of the preparation phase is an abstract flow diagram, according to which the design process may be detailed further. To this end, the manager extends the task net with tasks for designing the reaction, the separation, and the compounding (using an extruder), which are the major steps of the respective chemical process. To design the reaction, a process flow diagram is created which defines alternatives for this part of the chemical process. Only then may the task net be extended with design tasks for elaborating these alternatives (a CSTR reactor, a PFR reactor, or two reactor cascades). Sim- ulations are carried out to evaluate these alternatives; if necessary, laboratory experiments are performed to validate the simulation models. Finally, a decision is performed with respect to the selected alternative (task Compare). The subnet for designing the separation is structured in a similar way; here, the alternatives extraction and evaporation are considered. To start the inves- tigation of the separation step as soon as possible, an initial estimate of the output of the reaction step is made (simultaneous engineering). The subnet for designing the extruder is structured in a different way; special-purpose simulation tools are used to this end. Please note the feedback flow from DesignExtruder to DesignSeparation: it is not clear from the beginning to what extent the extruder can be used for the separation of remaining input substances. This requires negotiation between the designer of the separation step and the designer of the extruder. After having designed all parts of the chemical process, the overall concept is jointly discussed in a final decision step.
Show more

6 Read more

A review of the deterministic and diffusion approximations for stochastic chemical reaction networks

A review of the deterministic and diffusion approximations for stochastic chemical reaction networks

kinetics should have included randomness (´ Erdi and Lente, 2014) for historical remarks). The most popular way to describe the stochastic models of reaction networks is in terms of Con- tinuous Time Markov Chain (CTMC). The reactions are considered as happening at random events that modify the state of the network according to the stoichiometric equations. For some time these two descriptions have been developed in parallel and using different tools: the deterministic models were investigated in the theoretical and mathematical aspects, while the stochastic models were mainly studied from a computational point, e.g. in the search of suitable simulation algorithms.
Show more

23 Read more

Chemical and enzyme kinetics

Chemical and enzyme kinetics

One of the first thing to realize about enzymes reaction is that they do not follow the law of mass action directly. As the concentration of substrate is increased, the rate of the reaction increases only to a certain extent, reaching a maximal reaction velocity at high substrate concentration. This is in contrast with the mass action law, which, when applied directly to the reaction with the enzyme predicts that the reaction velocity increase linearly as the substrate increases.

68 Read more

Kinetics of a Reaction of 3 Chloroacetylacetone with Thioureas

Kinetics of a Reaction of 3 Chloroacetylacetone with Thioureas

The kinetic study of 3-chloroacetylacetone and thioureas has been carried out in ethanol. Thioureas used for study are thiourea, phenyl thiourea, p-methyl phenyl thiourea, p-ethoxy phenyl thiourea and p-chlorophenyl thiourea. The second order rate constants for these reaction were reported. The rate of reaction is first order with respect to 3-chloroacetylacetone and first order with respect to thioureas. The effect of substituents on the rate of reaction was also studied. Thermodynamic parameters are used to explain the nature of the reaction. The reaction products are isolated and characterized. Possible reaction mechanism for the reaction will be proposed and details of the kinetics will be discussed.
Show more

6 Read more

Numerical Simulation of Viscous Dissipation and Chemical Reaction in MHD of Nanofluid | Journal of Engineering Sciences

Numerical Simulation of Viscous Dissipation and Chemical Reaction in MHD of Nanofluid | Journal of Engineering Sciences

heat transfer of the thermophoretic fluid flow past an exponentially stretched surface inserted in porous media in the presence of internal heat generation/absorption, infusion, and viscous dissemination. Afify [8] examined the MHD free convective heat and fluid flow passing over the stretched surface with chemical reaction. A nu- merical analysis of insecure MHD boundary layer flow of a nanofluid past a stretched surface in a porous media was carried out by Anwar et al. [9]. Nadeem and Haq [10] studied the magnetohydrodynamic boundary layer flow with the effect of thermal radiation over a stretching surface with the convective boundary conditions.
Show more

9 Read more

Molecules computing : self-assembled nanostructures, molecular automata, and chemical reaction networks

Molecules computing : self-assembled nanostructures, molecular automata, and chemical reaction networks

We formally showed that predicting the behavior of robust processes does not require simulation of ev- ery reaction event. Specifically, we described a new approximate simulation algorithm called bounded tau- leaping (BTL) that simulates a certain ρ-perturbation as opposed to the exact SSA process. The accuracy of the algorithm in making predictions about the state of the system at given times is guaranteed for (ρ, δ)-robust processes. We further proved an upper bound on the number of leaps of BTL that helps explain the savings over SSA. The bound is a function of the desired length of simulated time t, volume V , and maximum molec- ular count encountered m. This bound scales linearly with t and C = m/V , but only logarithmically with m, while the total number of reactions (and therefore SSA steps) may scale linearly with t, C, and m. When total concentration is limited, but the total molecular count is large, this represents a profound improvement over SSA. We also argue that asymptotically as a function of t and C our algorithm is optimal in as far as no algorithm can achieve sublinear dependence of the number of leaps on t or C. This result is proven based on a widely believed assumption in computational complexity theory. Unlike Gillespie’s tau-leaping [7], our algorithm seems better suited to theoretical analysis. Thus while we believe other versions of tau-leaping have similar worst-case running times, the results analogous to those we obtain for BTL remain to be proved. Our results can also be seen to address the following question. If concerned solely with a particular outcome rather than with the entire process trajectory, can one always find certain shortcuts to determine the probability of the outcome without performing a full simulation? Since our lower bound on computation time scales linearly with t, it could be interpreted to mean that, except in problem-specific cases, there is no shorter route to predicting the outcomes of stochastic chemical processes than via simulation. This negative result holds even restricting to the class of robust SSA processes.
Show more

133 Read more

OPTIMIZATION OF REACTION KINETICS IN GYPSUM CALCINATION

OPTIMIZATION OF REACTION KINETICS IN GYPSUM CALCINATION

Abstract: Gypsum (calcium sulphate dihydrate) is of great industrial importance with over 95,000 ktonnes being used in the world per annum. The greatest use of gypsum is in the production of plaster (calcium sulphate hemihydrate) for use as an interior finisher. Plaster is produced by the calcination (thermal decomposition) of gypsum. There are a number of different phenomena occurring within a calciner, including heat transfer, mass transfer, particle and gas mixing, elutriation and the dehydration reaction itself. All these processes interact with each other. Although a lot of research has been carried out in these areas already, the literature has been found to contain significant discrepancies. This study contains experimental work which has been carried out in order to better understand the physical processes occurring within a gypsum calciner. The rate of dehydration of gypsum (35-67μm in diameter) has been studied in a fluidized bed reactor. Experiments were carried out at bed temperatures of 100 to 170°C. The fluidizing gas was air with water vapour pressures of 0.001 to 0.30 atm. The dehydrations were under differential conditions. A study of the fluidisation and elutriation properties of gypsum in batch vessels (cylindrical and conical) has been carried out. The mechanics of elutriation has been investigated and modeled for various freeboards, superficial gas velocities and air humidities. Residence time distributions were elucidated. Finally, the above experimental data and component models have been investigated for their applicability to producing a model of the laboratory scale gypsum calciner.
Show more

10 Read more

Chapter 12 - Chemical Kinetics

Chapter 12 - Chemical Kinetics

Instantaneous rate can be determined by finding the slope of a line tangent to a point representing a particular timec. C..[r]

6 Read more

Show all 10000 documents...

Related subjects