Quasiphysical And Anthropopathic Algorithm For Protein Structure Prediction

Protein structure prediction is a fundamental theory problem of molecular biology.Finding the solution of this problem not only have significant theory significance in life cognition science,but also great practical significance in drug research on diseases treatment. Application of available scientific apparatus and technical means to detect certain proteins macromolecules sector is facing certain difficulties.Thus, scholars have brought forward a number of simplified models , and have made use of computer science and the technology to study these models .Preliminary results show that the studying of these simplified models is of important values.For instance, the AB model discussed in this paper has high credibility on the issue of protein structure prediction.This paper introduce the structure method of AB model first,then the corresponding mathematical models are established.However, this model has two constraints, so it can not be the freedomly used to do the optimization. Therefore, the paper uses quasiphysical method to translate the original problem into unconstrained optimization problem.At this time,the optimization can easily be done. Then,a classic optimization algorithm named grad-descent method is introduced detailedly,and it is tested by a simple example on its credibility and excellent degree.Test shows that the credibility of the algorithm is very high, but the excellent degree is not enough .In other words, it can not reach the global minimum but local minimum.Therefore, this paper tries to improve this algorithm from two different aspects. On one hand, we bring forward a fairly good initial pattern structure method; on the other hand,a number of anthropopathic strategies are designed.Practice indicates that these improvements not only ensure the credibility of the algorithm, but also a higher excellent degree .Finally,the paper prospects and reviews the work of this stage and proposes some further improvement suggestions on the work from the two aspects which has discussed above.