In the post genome era, considerable effort has been put into genetic association study with single nucleotide polymorphisms (SNPs) to investigate genes affecting traits, for example diseases and response to drugs. Although various software tools for SNP association study read plain text files as input data, their formats is not standardized. Manual data conversion may cause incorrect input. In addition, validity of analysis may be lost by experimental fault and by data not under assumption of analysis method. To detect various errors in input data, we implemented 19 rules as SNP data consulting program. The program can also infer first cause of error and then suggest how user should correct it. We demonstrate the program is effective not only for in-house data but also for published data where errors are expected to have been removed. With this program, biologist would be able to perform intended and valid analysis