Genome-wide association study with machine-learning between H. pylori gastric-cancer and duodenal-ulcer strain reveal discriminatory amino acid differences and novel oncoprotein candidates. Helicobacter pylori

Genome-wide association study (GWAS) aims to reveal a genetic variation associated with a phenotype through genome comparison. GWAS must deal with linkage among genetic variations as a cause of false positive. An organism with the highest rate of homologous recombination breaking up linkage would be an ideal target for GWAS bypassing this problem. We tried GWAS in such a bacterial species, Helicobacter pylori, to distinguish between gastric (stomach) cancer and duodenal ulcer strains.