Drug important target evaluation method based on weighted network

Abstract

The invention relates to a drug important target evaluation method based on a weighted network. The method comprises the following steps: collecting a compound, a target and a pathway where target protein is located in a drug; scoring the chemical structure similarity of each compound in the drug, the protein interaction combination and the influence degree of the target protein in the pathway; representing the degree of the interaction relationship of the compound-target, the target-target and the target-pathway by using the scores, and constructing a compound-target-pathway weighted network by using the scores as weight values; calculating an evaluation parameter pi value of each target through the weighted network; sorting the evaluation parameter values of all the targets in a descending order, drawing a broken line graph of the names of the targets and the evaluation parameter values, finding out elbow points, and regarding all points larger than the evaluation parameter values corresponding to the elbow points as important targets. According to the method, the scale and the aggregation degree of the neighbors of targets are combined, the attributes of targets are described more meticulously, and the importance of targets in the network can be judged more accurately.

Description

Technical field [0001] The invention relates to a technology for evaluating drug targets, in particular to a method for evaluating important drug targets based on a weighted network. Background technique [0002] Because of its small-world and scale-free features, the complex network method has something in common with the complexity of TCM theory. Therefore, it is intuitive to apply the complex network method to TCM research. The complex network method can intuitively see the relationship through the establishment of a network model. In recent years, it has been gradually used in the fields of syndrome-treatment rules, prescriptions and medicine compatibility. Hopkins proposed the concept of network pharmacology in 2007, trying to design drugs with multiple pharmacological effects by analyzing the relationship between drugs, targets and diseases through network methods. At present, network pharmacology has become a hotspot in pharmaceutical research, and has been highly valued ...

AI Technical Summary

Problems solved by technology

Contrary to the degree index, although the aggregation coefficient can reflect the edge connection of the neighbor node to a certain extent, it cannot reflect the scale of the neighbor node of the node.

[0004] Therefore, the traditional unweighted network analysis method cannot take into account the strength of the connection between nodes, and there is also the problem of biased judgments on important nodes such as compounds and targets in the network.

Images