BACKGROUND: Malaria reduction and future elimination in China is made more difficult by the importation of cases from neighboring endemic countries, particularly Myanmar, Laos, and Vietnam, and increased travel to Africa by Chinese nationals. The increasing prevalence of artemisinin resistant parasites across Southeast Asia highlights the importance of monitoring the parasite importation into China. Artemisinin resistance in the Mekong region is associated with variants of genes encoding the K13 kelch domain protein (pf13k), found in specific genetic backgrounds, including certain alleles of genes encoding the chloroquine resistance transporter (pfcrt) and multidrug resistance transporter PgH1 (pfmdr1). METHODS: In this study we investigated the prevalence of drug resistance markers in 72 P. falciparum samples from uncomplicated malaria infections in Tengchong and Yingjiang, counties on the Yunnan-Myanmar border. Variants of pf13k, pfcrt and pfmdr1 are described. RESULTS: Almost all parasites harboured chloroquine-resistant alleles of pfcrt, whereas pfmdr1 was more diverse. Major mutations in the K13 propeller domain associated with artemisinin resistance in the Mekong region (C580Y, R539T and Y493H) were absent, but F446I and two previously undescribed mutations (V603E and V454I) were identified. Protein structural modelling was carried out in silico on each of these K13 variants, based on recently published crystal structures for the K13 propeller domain. Whereas F446I was predicted to elicit a moderate destabilisation of the propeller structure, the V603E substitution is likely to lead to relatively high protein instability. We plotted these stability estimates, and those for all previously described variants, against published values for in vivo parasitaemia half-life, and found that quadratic regression generates a useful predictive algorithm. CONCLUSION: This study provides a baseline of P. falciparum resistance-associated mutations prevalent at the China-Myanmar border. We also show that protein modelling can be used to generate testable predictions as to the impact of pfk13 mutations on in vivo (and potentially in vitro) artemisinin susceptibility.