From the research direction of biology, it is impossible to master only a single biological knowledge, whether macro or micro. Judging from the development trend of biology, today's cutting-edge technology may become the basis for the development of biotechnology tomorrow. This requires our students to constantly master new knowledge and understand new achievements. Only in this way can we stand in front of our predecessors ... See /mainweb/sw/kctz/smfz.htm for relevant information.
Secondly, a series of problems in the field of genetic engineering need computer-related problems (such as gene recognizer and biometric pattern recognition). Use a biological computer.
biocomputer
Biological computer is a computer based on the way of solving problems in the biological world. At present, there are mainly biomolecule or supramolecular chips, automata models, bionic algorithms, biochemical reaction algorithms and so on.
The computer industry has developed rapidly in recent decades. However, the current transistor density is close to the theoretical limit of the technology currently used. Can transistor computers continue to develop? Therefore, people are constantly looking for new computer structures. On the other hand, while studying artificial intelligence, people draw lessons from various ways of solving problems in the biological world, that is, the so-called biological algorithm, and put forward some models of biological computers, some of which solve some problems that are difficult to be solved by classical computers.
At present, biological computers mainly include the following categories:
1. Bio-molecular or supramolecular biochip: Based on the traditional computer model, a lot of research and development have been made on the structure and function of small, large and supramolecular biochips in vivo, starting with the search for efficient and miniaturized electronic information carriers and information transmitters. Biochemical circuit belongs to this.
2. Automata model: Based on automation theory, we are committed to finding new computer models, especially non-numerical computer models for special purposes. At present, the research mainly focuses on the analogy of basic biological phenomena, such as neural network, immune network, cellular automata and so on. The difference between different automata is mainly the difference between the internal connections of the network. Its basic feature is collective computing, also known as collectivism, which has great potential in non-numerical computing, simulation and identification.
3. Bionic algorithm: Based on biological intelligence, we are committed to finding a new algorithm mode with bionic concept. Although it is similar to the idea of automata, it does not pursue hardware changes based on algorithms. 4. Biochemical reaction algorithm: Based on the controllable biochemical reaction or reaction system, the high copy number of similar molecules in a small volume is used to pursue high parallelism of operation, thus improving the operation efficiency. DNA computers fall into this category. The following will focus on the computational neural network model in the automaton model and the DNA computer in the biochemical reaction algorithm.
Computational neural network
As early as 1943, psychologist W. mcculloch and mathematician W. Pitts jointly proposed the binary logic model of neurons. 1949 D. Hebb proposed a learning rule to change the connection strength of neurons, which has played an important role in various network models so far. 1962 F. Rosenblatt proposed the perceptron model. 1982, American physicist J.Hopfield put forward a brand-new neural network model, which embodies the basic spirit of D. Marr's computational neural theory, dissipative structure and chaos theory, replaces binary logic with S-shaped curve, and introduces "energy" function, which makes the stability of the network have strict judgment basis, and the model has the intelligence of ideal memory, classification and automatic correction. The analysis of dynamic characteristics of Hopfield model provides a powerful research method.
Neural network system simulates the working mode of the brain, which is formed by a large number of simple neurons widely connected with each other, forming a topological structure. Compared with the traditional computer model, the brain has the following characteristics: first, large-scale parallel processing ability, second, the brain has strong "fault tolerance" and association function, and third, the brain has strong adaptability and self-organization ability. In these aspects, the current traditional computer model is difficult to realize.
The specific neuron model is mainly about how to better reflect the essence of neurons' release potential under stimulation. Most models regard the connection between neurons as a linear connection, and the input layer is directly connected with the output layer, with no so-called hidden unit layer in the middle. Each neuron can only be excited or inhibited, and the input of any neuron is the sum of the outputs of other neurons through synaptic action. If the transition between excited state and inhibited state is considered, the nonlinear input-output characteristics of neurons can be characterized by S-shaped curves, such as J. Hopfield model. According to the concept and method of statistical physics, the input of neurons is determined by the probability of neuron state update, such as Boltzmann machine model; You can also add an intermediate conversion layer, such as perceptron model, to the input and output layers of neurons. Plus the backward propagation model of backward error correction channel and so on. By expressing the morphology and function of neurons differently, different models can be produced.
biocomputer
1994, Dr. L. Adleman of the University of California published the theory of DNA computer in the journal Science, and successfully carried out the operation experiment in the test tube of DNA solution. Dr. L. Adleman's DNA computer is a brand-new concept. The basic idea is that DNA base sequence is used as the carrier of information coding, and modern molecular biology technology is used to control the DNA sequence reaction under the action of enzymes in test tubes as the process of operation; That is, the DNA sequence before the reaction is used as the input data, and the DNA sequence after the reaction is used as the operation result. DNA computer is a chemical reaction computer. So far, some basic NP problems have been solved by experiments with DNA computer model. For example, Dr. L. Adleman's traveling salesman problem (Hamilton diagram problem, HPP) and Princeton University's Chacopton's satisfiability problem (SAT problem). The so-called NP problem is divided according to the algorithm complexity of the problem class, and people are opposite to P problem. P problem refers to an algorithm whose complexity increases with the increase of the scale of the problem, and it can be calculated. NP problem means that the complexity of an algorithm increases exponentially with the scale of the problem, which is actually uncountable. The idea of DNA computer is a potential innovation. DNA computer is fast, and the amount of calculation in a few days is equivalent to the total amount of calculation of all computers in the world since the advent of computers. Its storage capacity is huge, but its energy consumption is only one billionth of that of ordinary computers. Of course, DNA computer is only a theoretical hypothesis, and it is still quite imperfect in many aspects. Mainly manifested in:
The reality and computational potential of 1. structure. The DNA computer takes the encoded DNA sequence as input and reacts in the test tube to complete the calculation. The reaction products and solutions give all solution spaces, but how to separate the optimal solution from other solutions and how to output it is a highly technical problem. At present, there is no satisfactory means of output. With the expansion of problem solving scale, the output will become the bottleneck of DNA computer.
2. Errors in operation. In the process of amplifying DNA, there is a high mismatch rate, and a large amount of DNA will also produce some hundreds of side reactions. Errors will produce false solutions, which will increase the difficulty of outputting the optimal solution.
3. Man-machine interface. How to make the input and output of DNA computer acceptable to ordinary people, otherwise it will not be widely used.
In any case, the introduction of DNA computer broadens people's horizons, inspires people to study life with the concept of algorithm, and challenges many fields. (/question/7358400.html is the original source) Related "Biocomputer problems can go to /s? ie = GB 23 12 & amp； England =% ce% B4% BD% E2% be% F6% b5% C4% C9% fa% ce% ef% ce% ca% cc% E2&; sr = & ampz = & ampcl = 3 & ampf = 8 & ampwd = % C9 % FA % CE % EF+% BC % C6 % CB % E3 % BB % FA & amp； Ct=0 Watch (I hope you can find what you want)
Let's talk about chemistry and calculation first. It is necessary to use c++ to program and edit nonlinear elements and experimental data. The simulation test in the program needs neither test space nor many instruments, which saves a lot of unnecessary resources. Moreover, we can communicate with experts from all over the world on the Internet and conduct experiments together. These are all components that can be developed and utilized by chemical and biological computing software. ...
Chemical computer
Imagine what computers will look like in the future. If someone says, let things like jelly think and express sympathy, do you think it is possible? For modern people who have long been accustomed to and familiar with angular display screens, hosts and mice, it is indeed a bit whimsical to imagine a computer as a soft and slippery jelly with no fixed shape. However, Andrew, a computer expert at the University of Bristol in England, is dreaming of replacing electrons with ions and silicon chips and circuit boards with colloidal substances. When most people are tired, they usually have a cup of coffee, or go for a walk outside to breathe some fresh air. Andrew is different. When he felt that his brain was a little dull and needed some extra stimulation, he asked his robot to pull a plate full of chemical liquid with his metal fingers. This plate of chemical liquid is the "brain" prototype of the liquid computer designed by Andrew. The formation and diffusion of ion wave is a "thinking" process of chemical computer. When the running speed slows down, the "brain" will give instructions to the manipulator to immerse the metal fingers in the dish and shake those magical chemical liquids.
The chemical computer designed by Andrew now simply imitates the feedback process between human arm and brain. His ambition is to design a chemical processor and put computer hardware into a bottle. After 10 years of research, Andrew has now developed liquid logic gates, and thinks that the array he designed has unlimited self-reorganization and repair ability. Computer giant IBM also believes that it is possible to design powerful new computer chips by using this array technology. In addition, Andrew has another ambitious goal, which is to further strengthen the ability of "drum wave" and make it worthy of the title of liquid brain. In order to prove that the concept of liquid brain has unlimited potential and bright prospects. Andrew specially designed the jelly robot, the carrier of liquid brain. It has artificial eyes and synthetic hormones. Perhaps one day, the jelly robot can feel the surrounding environment, and may even feel human emotions. Chemical computer has a very complicated and fascinating feature, which is called Belosov-Chabo Tynsky reaction (BZ reaction). It is a chemical oscillating reaction composed of three different reactions. Each reaction has different molecules and ions. When specific chemical components are added, the first reaction is triggered, and the products produced can trigger the second reaction, then the products of the second reaction can trigger the third reaction, and the products of the third reaction can trigger the first reaction, and so on. Even more amazing is that different reactions will produce different colors, so red and blue waves can be formed. The importance of BZ reaction lies in that it can solve some mathematical problems, especially some problems that are difficult to be solved by computers now. For example, the shortest path problem in a maze. It takes a lot of time to solve this problem with traditional computers, and all the paths should be exhausted and then compared. But the BZ reaction is different. Because waves always take the shortest path when they spread and spread. This problem can be solved as long as the wave trajectory is recorded by the camera.
In the mid-1990s, Andrew realized that BZ reaction has more important applications, that is, it can be used in chemical processors. To this end, he organized a special team and developed two conceptual models of chemical processors. A model can imitate the feedback activities of human arms and brains. The other is composed of two BZ reactions, which can automatically move to the destination in a room full of furniture. Although these two conceptual models performed well, Andrew realized that if the chemical processor is to handle more complex operations, it must have logic gates. A theoretical study by Boston University caught Andrew's attention. The research thinks that we can imitate snooker and make a simple processor. That is to say, each ball can represent 1 or 0, and the collision process of balls is the calculation process. How the ball collides and the ejection direction after collision can be accurately expressed as a logical process. In other words, the collision result is equivalent to a logic gate. In this way, Andrew's task becomes how to make BZ waves collide. Last year, Andrew made a major breakthrough in his research. He put the BZ mixture on the thin glue layer of silver halide, because halide can act as a chemical retarder, and the glue layer can delay the propagation speed of waves. In this way, the BZ reaction did not form a complete circular wave, but formed a small arc, which spread along a straight line. Andrew called it BZ bomb. BZ bullet shows more quasi-particle characteristics than wave characteristics, and its performance is similar to billiards. Andrew found in the experiment that when two BZ bombs collide at a specific angle, they will only produce a unique output in a specific direction. If there is only one input, there is no output in this direction. So Andrew came up with the logic. Since then, he has developed logical OR, logical negation and logical mutual exclusion, which laid a solid foundation for Andrew's chemical processor. Although Andrew's chemical processor is still in its infancy, he has turned his attention to parallel chemical processors. Whether the chemical processor can be successful is still at an unknown stage, but scientists believe that if humans can have the ability to control nano-scale waves, the chemical processor is likely to be realized. As some experts say, whether Andrew's ambition can be realized or not, his research work is of great significance for revealing the mystery of human brain and making better processors. After all, chemical processors are the bridge between biological tissues and organs and electronic equipment.
(/s？ ie = GB 23 12 & amp； bs = % C9 % FA % CE % EF+% BC % C6 % CB % E3 % BB % FA & amp； sr = & ampz = & ampcl = 3 & ampf = 8 & ampwd = % BB % AF % d 1% A7+% BC % C6 % CB % E3 % BB % FA & amp； Ct=0).
That's all I can say. I hope it helps you.