Simulating the idea of electronic engineering, from the basic parameterized components to continuously construct more complex results, such as from logic gates to various modules to integrated circuits and finally to computers to networks, our biological level is also from the expression of genes and proteins to their interactions to specific biological pathways to specific cells, tissues, organs, systems, and organisms. The disadvantage is that biological systems are nonlinear, and we can only approximate them with linearity and cannot be equal;

We can only make certain abstractions of biological systems, such as central laws, and then explore various related relationships on this basis. We believe that various corresponding elements can be constructed in the underlying biological reaction to correspond to certain functions, such as memory, we can explore the memory of organisms, we can build a certain model through the measurement of allosteric effects, methylation, phosphorylation, ubiquitination, etc. of multiple genes and proteins, and then perform pattern recognition.

Of course, the construction of biological elements is not a problem (the selection of objects with a certain correlation is a must for the modeling of objects, and the construction of feedback structures is a must), the problem is how to build a system of elements with high internal aggregation and low external coupling, so as to be able to interact with each other with more certainty (the complexity of biological systems makes their relationships uncertain). Therefore, we may need to use a new approach, no longer based on the deterministic steps of linear systems, but on massively parallel operations (high redundancy) of nonlinear systems, and finally only part of the data we need.

At present, we can only hope to find a relatively certain pathway, that is, a pathway with a relatively low degree of coupling with the surrounding environment, so as to use relevant theories to guide our modeling. Otherwise, the expression of the polygenic plasmid we hope to construct may be unsatisfactory after being transferred into the engineered bacteria, that is, we also need to consider the design of the final gene expression circuit, such as the transfer of the four Yamanaka Shinya factors to make the reverse differentiation of fibroblasts into iPS cells is the result of large-scale screening, we can reduce the scope of screening through advance design, and even achieve the formation of a regulatory network, as long as we ensure its high dependence on the external environment, we can ensure biosecurity to a certain extent, that is, we need to externally regulate through the addition of external substances at the same time。

1. Inspiration brought by engineering standards such as iGEM

1 Acting against specific diseases through the activation/inhibition of relevant signaling pathways, such as the combination of the NO pathway and other pathways for diseases such as bacteremia, which requires specific programming to maintain its expression level at a higher level 2 The role of specific proteins, possible immobility points, such as receptors, It can play a more significant role in its effects3 at the cellular level through the action of different signaling pathways to guide its possible biological processes, such as proliferation, differentiation, apoptosis, etc., 4. Use the known functional modules to combine to form a system with certain complex functions5. For the resistance of the virus, it can also take the resistance of corresponding proteins, such as CD4 to the construction of HIV6 logic circuits7, corresponding functions, such as the integration of immune-related pathways, 7. the development of therapies for related diseases, 8. Ensuring biological safety, 9. Based on the functional realization of specific elements, such as programming

2. DNA assembly technology and genome synthesis

The question is how to maintain its consistency with the original system, and how to eliminate the error of high-throughput sequencing? It seems that the idea of modularity can only be adopted, and each module should have a little overlapping sequence to facilitate the identification of its order, and then it can be screened through large-scale synthesis.

3. Gene regulation circuits

The construction of various engineered components is the low-level work.

1Repressilator Oscillator Elowitz & LeiblerNature (2000) 403:335-338 Select lacI-lite; cI-lite; tetR-lite: to build an oscillator. The use of tmRNA to terminate indicates that the central law of DNA, RNA, and protein can be used in the realization of a specific function, just like Newtonian mechanics.

2Toggleswitch (bistable switch) Gardner & Collins Nature (2000) 403:339-342 Modeling of differential equations:

The implementation of a feature also requires a certain output to determine the success of the build. For example, the correlation can be revealed through the change in the concentration of the relevant chemical substance, or even the change in the rate of its change.

These suggests that RNA can play an important role in the modeling and decoupling of systems.

4. Simulation calculation

StochasticmodelThe stochastic process can describe the possible biological responses at a macroscopic level, and its original assumption is that there is a certain distribution: 1. The time to generate the next generation Pτ(t) = ∣w(x) ∣1*e^∣w(x) ∣1t; τ = 1/∣w∣*log1/r1.2 The occurrence of different reactions depends on the allocation of weights3 update status

ODE(ordinary differential equation)dP(I,j)/dt=[w]P(I,j)

PDE (Partial Differential Equations)

The Biophysicalmodel applies the correlation between signals and systems to understand the modeling between the various components, the logical order of which is the product of the transfer function.

BooleanModel (Boolean logic)

FluxBalance****ysis (FBA)

How do we develop the modeling of the network on these foundations, i.e., simulations at the overall level? We don't want to adopt a model that traverses the bottom of an electronic engineering and that organisms may adopt a model that may be less certain but has some effect, but the point is that organisms are imperfect, but the task of each generation of organisms is to survive, so that there are infinite possible recombinations within them and finally a better model emerges that can thrive and survive under the pressure of natural selection. This model of genetic algorithm may better describe the network system: we know that it will evolve and can make short-term predictions about where it will go. This is equivalent to understanding at a sufficiently high-dimensional level, just as the Laplace transform converts differential equations into algebraic equations to understand changes between states at the level of distribution.

Correlation analysis and modeling of known objects through high-throughput measurements is in line with the complexity of biological networks, and only highly correlated ones have good application value.

5. Protein signaling pathway

It is designed through a combination of relevant signaling pathways, which allows the organism to deterministically express a specific substance. This is an external form of gene therapy.

Selecting these objects for a linear combination is simulating a Fourier transform. This protein circuit is consistent with the construction of gene circuits.

Standardized research ideas require us to find sufficiently high-dimensional modules and then reassemble them, but the high degree of interconnection of network systems makes it impossible to have such an ideal structure. Of course, it's still okay to get close.

At present, the effects of plasmids can be observed by transferring into engineered bacteria (restriction enzymes and ligases), but the combined effects of various gene sequences are considered in advance, because there is a complex network relationship that makes its expression a probabilistic event, so that it cannot meet our needs. And the stability of its expression is also a big problem. Perhaps the current idea is to conduct large-scale screening to obtain subjects whose expression is more stable and whose passage is relatively stable?

6. Primary metabolism

The metabolome is a more complex high-dimensional level. We use the combination of genes and proteins to influence their possible biochemical reactions, so that the overall balance shifts in a certain direction, such as obtaining the target product.

7. Can gene editing tools have a more certain impact on the expression of the network?

The previous Cre‐loxp system used enzymes to recognize specific sequences loxP (a specific DNA sequence from phage P1), while the CRISPR-cas9 system uses a system similar to the mechanism of acquired immunity in bacteria to work by recognizing CRISPR special DNA repeats, precisely targeting through crRNA, and using cas9 enzymes to cleave DNA sequences.

The hybridization of this sequence inspires us to use the gene expression of different species to play a deterministic role. At the cellular level, it can be fused with mouse lymphocytes and tumor cells as in the production of monoclonal antibodies.

Orthogonal biology, like the Fourier transform, selects a specific object to a specific degree to simulate all aperiodic signals, i.e., selective expression to treat diseases (we tentatively consider signaling pathways as linearly independent substrates)?