Akutsu T, Tashimo H
Human Genome Center, University of Tokyo, Japan.
Pac Symp Biocomput. 1998:413-24.
This paper proposes a novel method of deriving a contact potential (a pair score function) for protein threading. In this method, the constraint that the score of the native threading is minimum over all possible threadings is expressed in a form of linear inequalities, and then parameters defining the contact potential are determined by applying a program package of linear programming. The most important advantage of this method over the previous methods is that this method can learn a score function from a small number of training data. The proposed method was evaluated using Lathrop and Smith's algorithm for finding optimal threadings and was shown to be effective for computing nearly correct threadings.
本文提出了一种用于蛋白质穿线法推导接触势（成对评分函数）的新方法。在该方法中，天然穿线法的得分在所有可能的穿线法中为最小值这一约束条件以线性不等式的形式表示，然后通过应用线性规划程序包来确定定义接触势的参数。该方法相对于先前方法最重要的优势在于，它能够从少量训练数据中学习评分函数。使用Lathrop和Smith寻找最优穿线法的算法对所提出的方法进行了评估，结果表明该方法对于计算近乎正确的穿线法是有效的。
