OBJECTIVE： To explore the difference of genotypes in patients with coronary artery disease （CAD） and different responses to clopidogrel. METHODS： Totally 159 CAD patients were selected from cardiology department of our hospital during Mar. 2013-Nov. 2015. They were given clopidogrel+aspirin for dual antiplatelet therapy for at least 1 year. Turbidimetry method was used to detect the percentage of platelet aggregation induced by adenosine diphosphate （ADP） and arachidonic acid （AA） before and after treatment. The polymorphism of cytochrome P450 （CYP）2C19， CYP3A5， wild type leucine 33 allele （PLA1）/proline 33 allele （PLA2） were detected by PCR-RFLP. RESULTS： There were 3 kinds of CYP2C19 genotypes （*2/*2， *2/*1， *1/*1）， 3 kinds of CYP3A5 genotypes （*3/*3， *3/*1， *1/*1） and 3 kinds of PLA1/PLA2 genotypes （A1/A2， A2/A2， A1/A1）； the frequency of each genotype was in line with Hardy-Weinberg balance （P＞0.05）. Among 159 cases， there were 81 cases of clopidogrel “semi-response”， accounting for 50.9％； 78 cases of clopidogrel “response”， accounting for 49.1％. The frequencies of CYP2C19 gene deletion （*2/*2 or *2/*1 genotype） and *2 allele in clopidogrel “semi-response” patients were significantly higher than clopidogrel “response” patients， with statistical significance （P＜0.05）. The frequencies of PLA1/PLA2 gene deletion （A2/A2 or A1/A2 genotype） and A2 allele in clopidogrel “semi-response” patients were significantly higher than clopidogrel “response” patients， with statistical significance （P＜0.05）. The frequencies of CYP3A5 gene deletion （*3/*3 or *3/*1 genotype） and *3 allele in clopidogrel “semi-response” patients were slightly higher than clopidogrel “response” patients， without statistical significance （P＞0.05）. After treatment， the percentage of ADP or AA-induced platelet aggregation in different genotypes patients were significantly lowered， compared to before treatment； but the percentage of platelet aggregation in CYP2C19 gene deletion and CYP3A5 gene deletion patients were significantly higher than gene expression patients （*1/*1 genotype）， with statistical significance （P＜0.05）. There was no statistical significance in the percentage of platelet aggregation between PLA1/PLA2 gene deletion patients and gene expression patients （P＞0.05）. CONCLUSIONS： The incidence of clopidogrel “semi-response” in CAD patients is high. CYP2C19 and PLA1/PLA2 gene polymorphism may be related to clopidogrel “semi-response”， while CYP3A5 gene polymorphism has no relationship with it. CYP2C19 and CYP3A5 gene deletion may weaken inhibitory effects of clopidogrel on platelet aggregation of CAD patients.