QTL mapping of soybean cyst nematode race 9: a generalized linear modeling approach
The Female Index (FI) is a relative measure of host suitability of a soybean line for a particular nematode population and often shows a non-normal distribution. Moreover, most quantitative trait loci (QTL) mapping methods assume that the phenotype follows a normal distribution such as composite int...
| Autores: | , , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2015 |
| País: | Argentina |
| Institución: | Consejo Nacional de Investigaciones Científicas y Técnicas |
| Repositorio: | CONICET Digital (CONICET) |
| Idioma: | inglés |
| OAI Identifier: | oai:ri.conicet.gov.ar:11336/7843 |
| Acceso en línea: | http://hdl.handle.net/11336/7843 |
| Access Level: | acceso abierto |
| Palabra clave: | Female index Generalized linear model Glycine max Heterodera glycines https://purl.org/becyt/ford/4.1 https://purl.org/becyt/ford/4 |
| Sumario: | The Female Index (FI) is a relative measure of host suitability of a soybean line for a particular nematode population and often shows a non-normal distribution. Moreover, most quantitative trait loci (QTL) mapping methods assume that the phenotype follows a normal distribution such as composite interval mapping (CIM). Therefore, a generalized linear modeling (GLM) approach was employed to map QTL for resistance to race 9 of the soybean cyst nematode (SCN) using a total of 83 simple sequence repeat markers (SSR). Two GLM models were tested: model 1, where the FI was treated as a continuous variable, assuming a Gamma distribution with a logarithmic link function; and model 2, where the FI was treated as a categorical trait in a five-item hierarchy, assuming a multinomial distribution with a cumulative logit link function. The FI data of 108 recombinant inbred lines (RIL) confirmed the non-normal distribution for race 9 of the SCN (Shapiro-Wilk?s w=0.86, P<0.0001, skewness=1.52 and kurtosis=2.93). Eight RIL were confirmed to be resistant (FI≤10), and 23 to be highly susceptible (FI≥100). Both GLM models identified one QTL for SCN on the molecular linkage group G, between the markers Satt275 and Satt038 at 48.4 centiMorgans (P=0.017 and 0.033, for models 1 and 2, respectively). Additionally, these results were also compared with the CIM and Bayesian interval mapping (BIM) methods, assuming experimental data with a non-normal response, to determine the robustness and statistical power of these two methods for mapping QTLs. The results make clear that generalized linear modeling approach can be used as an efficient method to map QTLs in a continuous trait with a non-Gaussian distribution. CIM and BIM were robust enough for a reliable mapping of QTLs underlying nonnormally distributed data. |
|---|