A total of 312 boars (201 Landrace and 111 Large White) were genotyped with a custom-made low throughput genotyping microarray (called SNiPORK) based on array primer extension (APEX) technology. The results were used to association studies between genotyped SNP markers and daily gains, meat content and selection index. Among the 60 SNP markers analyzed, 14 of them showed statistically significant associations between the genotype and the level of at least one trait. In order to find extremely beneficial or unfavorable intergenic diplotype combinations, 5 SNP markers were selected: CASTA499C, MYF6 T255C, PKLR C384T, SFRSI C1146T and TNNT3T 153C, which showed statistically significant associations at P<or=0.01 within one of the traits and the frequency of homozygotes with a minor allele of at least 0.1. Among 10 possible permutations, statistically significant associations were found only for a combination of SNP markers in TNNT3 x SFRSI genes and, interestingly, for all combinations with SNP located within the calpastatin (CAST) gene commonly known as a gene influencing pork quality traits. This study also found that CAST allele A (which is beneficial for pork tenderness) is also favorable for growth rate. This effect is clearly increased with additive alleles C of myogenic factor MYF6 and troponin T3 (TNNT3) and is decreased with each allele Tfrom serine rich splicing factor 1 (SFRS1) gene. For meat content, the most favorable genotype of calpastatin gene was AC, the effect of which was generally increased with each C allele of MYF6 and TNNT3 and decreased with each allele T from SFRSl and PKLR (puruvate kinase) genes, respectively. The calpastatin AC genotype seemed to be beneficial for selection index, although in combination with the CC genotype of troponin T3, calpastatin genotypes AA and CC reached higher values. In the case of the combination of genotypes TNNT3 x SFRS1, the most preferable for all analyzed traits is the CC genotype of TNNT3, especially in combination with CTor CC genotypes in the SFRS1 gene. We conclude that searching for interaction effects between candidate SNPs (even of moderate influence) may lead to interesting and valuable findings enabled better understanding and applications of SNPs in pork yield and quality improvement programmes.