G protein-coupled receptors (GPCRs) are important drug targets, and more than 30% of commercial drugs are known to act on various GPCRs. Structure-based drug design has been shown to be effective to improve drugs with fewer side effects and higher efficacy. However, human GPCRs generally have low thermal stability and are often denatured during purification, making bulk production a difficult task. This has lagged research on GPCR structural analysis. We have developed a method for predicting the thermostable mutants of GPCRs using statistical mechanics theory calculations. Using this technique, we succeeded in the purification of many human GPCRs. In this talk, I would like to introduce the current status of production method of human GPCRs in the world and discuss the effectiveness and limits of the developed technique.