This article addresses a currently relevant topic related to detecting associations of copy number polymorphism with traits and will be of interest to readers of Genetics Research. The simulation study showed that additional CNP information could increase the accuracy of the predicted genotypic value, compared to using SNP information alone in an association study. The accuracy was strongly dependent on the heritability of CNP phenotypes (correlation between genotype and CNP phenotype) (Table 3). The increased prediction accuracy with CNP information could also result in smaller mean square prediction errors (Table 4). On the other hand, the authors found that linkage disequilibrium (LD) between SNPs and CNPs was not much different from that between two SNPs. The authors demonstrated that an increased mutation rate and a larger number of segregating alleles are unlikely to influence LD with a nearby SNP. I have the following important comments.1) One possible reason for high accuracy in using CNP information was that CNP was the causal gene itself in the simulation. If CNP had negligible effects and was simply linked to a causal gene, the performance might not be much different from that of SNPs, because as stated above the LD structure was not much different between SNPs and CNPs. In this situation, success in using CNP information to predict genotype values ​​would be conditional on 1) effects due to the causal CNP explaining a significant proportion of the genetic variation, 2) causal CNPs needing to be genotyped, and 3) l The accuracy of genotyping for CNP should be reasonably high. Overall, how easily and accurately are causal CNPs identified? What is the underlying distribution of effects due to CNPs? We can expect the corr...... in the center of the paper...... the same effect of the same direction and magnitude. Would this be realistic?10) Line 158, in case the number of copies is the same (c=b), the equation would give zero. Is this right?11) Line 171, SNP phenotypes -> CNP phenotypes?12) Line 242, 2 or 3 alleles > 2 or 3 copies?13) Table 2. It might be useful to have a likelihood ratio for each model.14) Row 267, phenotype -> phenotype?15) The introduction is human/livestock oriented, but the rest of the document is presented in zootechnical terminology. A discussion of the relevance of the findings to human populations would be helpful. 16) Line 343 "Assuming that x copies at a CNP locus lead to the effect of x times the effect of 1 copy", is this a realistic assumption? The 4-copy CNP genotype is likely to have the same functional impact for individuals with 2/2 vs. 4/0.