Abstract: The complex network is an effective method to study the heat resistant mechanism of xylanase. The residue interaction networks of thermostable xylanase from Thermoascus aurantiacus xylanase and mesophilicxylanase from Streptomyces lividans xylanase were constructed, and frequent path mining was carried out using Dijkstra algorithm and graph-based frequent pattern mining algorithm.The results showed that at the same temperature,the communication of Xyna＿Strliwasmore active than that of Xyna＿Theau, and was more affected by the increased temperature. The frequent pathways of Xyna＿Strliare concentrated in the β1, β2 and loop8, β7 regions, while the frequent pathways of Xyna＿Theauare closely related to the β1, β2, β3 regions, and the two proteins were found to have the same pathways in some regions. In addition, comparison showed that Xyna＿Strli communication was greatly affected by C terminal, turn and heat-sensitive region when the temperature rised.