[Objectives] DNA-based research methods have been widely applied in the study of species taxonomy and phylogeny, ecological genetics, and conservation biology, however, the comparative studies on DNA samplings are generally scarce. In this study, we aimed to compare the performance and applicability of different DNA sampling and extraction methods by using the captive-bred Chinese giant salamander (Andrias davidianus) as study objects. [Methods] DNA samples were obtained by four sampling methods including buccal swabbing, skin swabbing, shed skin sampling and tail venous blood sampling (Fig. 2). DNA was extracted by four different methods including Kit method, high salt method, phenol-chloroform method and magnetic bead method based on the most accessible skin swabbing samples. All the DNA qualities and concentrations were detected by gel electrophoresis and nucleic acid protein analyzer, and each of the sampling DNA was identified by PCR and sequencing of the mitochondrial COI fragments. In addition, the daily and feeding behaviors of the experimental animals were observed after samplings. [Results] Our results showed that the daily and feeding behaviors of A. davidianus didn’t change visibly, thus the four sampling methods could belong to the category of non-disruptive DNA sampling that has minimal impact on the fitness, behavior or welfare of the experimental animals. Among the four different sampling methods, the tail venous blood sampling performs optimally, followed by buccal, skin swabbing, and then shed skin sampling in terms of DNA quality and concentration (Fig. 3a, b). However, the DNA obtained by four different extraction methods based on skin swabbing samples were generally similar (Fig. 3c, d). Although different sampling methods performed differences in obtaining DNA, all the extracted DNA could be amplified of the target COI gene fragments (Fig. 3e, f). [Conclusion] By summarizing the advantages, disadvantages and precautions of different DNA sampling and extraction methods (Table 1), we suggest that the most appropriate sampling method should be selected according to the experimental purposes and specific conditions. In terms of extraction methods, however, considering the toxicity of reagents, the complexity of operation, time and economic costs, the kit method is what we recommended. This study could provide some guidance for non-disruptive sampling and relevant molecular ecology studies of A. davidianus and other rare and endangered animals in the future.