Multiple correlated phenotypes are frequently collected in genome‐wide association studies (GWASs), and a systematic, simultaneous analysis of multiple phenotypes can integrate the signals from single phenotypes, therefore increasing the power of detecting genetic signals. However, fundamental questions remain open, including the conditions and reasons under which the multivariate analysis is beneficial, how a highly significant signal arises in the multivariate analysis. To understand these issues, we propose to decompose the multivariate model into a series of simple univariate models. This transformation offers a clearer quantitative analysis of the circumstances under which a multivariate approach can be beneficial for the bivariate phenotypes case. A real data analysis is employed to illustrate how to interpret how the signals arising from multivariate GWASs. 相似文献
Variations in the potential glycosylation sites were observed in hemagglutinin (HA) sequences of H9N2 avian influenza virus isolated in China, deposited in the Influenza Virus Resource of NCBI before 2017, which showed a deleted glycosylation site at amino acid residue 218, and an introduced glycosylation site at amino acid residue 313. Based on the variations in the glycosylation sites at these amino acids, H9N2 avian influenza viruses could be divided into three categories. Firstly, most of the H9N2 influenza viruses were 218G+ viruses; less 313G+ viruses were isolated between 1997 and 2004. Secondly, the occurrence of the 218G+/313G+ viruses increased, while the 218G+/313G? viruses decreased from 2005 to 2012. Thirdly, from 2013 to 2016, the 218G?/313G+ viruses were predominant compared to the 218G+/313G+ viruses. Here, based on an F/98 virus backbone, a 218G+/313G? virus, two reassortment viruses were generated, and named rF/HA218G+/313G+ and rF/HA 218G+/313G?, respectively. HA protein migration demonstrated that the glycosylation sites at amino acid residues 313 and 218 were both functional. The absence of the glycosylation site at amino acid residue 218 and the presence of the glycosylation site at amino acid residue 313 increased antibody binding and moderately prevented the virus from escaping neutralization with homologous antisera. Additionally, compared to the F/98 virus (218G+/313G?), the viruses rF/HA218G+/313G+ or rF/HA218G?/313G+ showed significantly increased infectivity of MDCK cells, chicken embryo eggs, and trachea and lung tissue of SPF chickens, but did not display differences in airborne spread in chickens or infectivity of mice compared with its parental virus F/98.