Cargando…

Combined QTL mapping on bi-parental immortalized heterozygous populations to detect the genetic architecture on heterosis

From bi-parental pure-inbred lines (PIL), immortalized backcross (i.e., IB(1) and IB(2), representing the two directions of backcrossing) and F(2) (i.e., IF(2)) populations can be developed. These populations are suitable for genetic studies on heterosis, due to the present of both homozygous and he...

Descripción completa

Detalles Bibliográficos
Autores principales: Huo, Xuexue, Wang, Jiankang, Zhang, Luyan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10112513/
https://www.ncbi.nlm.nih.gov/pubmed/37082336
http://dx.doi.org/10.3389/fpls.2023.1157778
Descripción
Sumario:From bi-parental pure-inbred lines (PIL), immortalized backcross (i.e., IB(1) and IB(2), representing the two directions of backcrossing) and F(2) (i.e., IF(2)) populations can be developed. These populations are suitable for genetic studies on heterosis, due to the present of both homozygous and heterozygous genotypes, and in the meantime allow repeated phenotyping trials across multiple locations and years. In this study, we developed a combined approach of quantitative trait locus (QTL) mapping, when some or all of the four immortalized populations (i.e., PIL, IB(1), IB(2), and IF(2)) are available. To estimate the additive and dominant effects simultaneously and accurately, suitable transformations are made on phenotypic values from different populations. When IB(1) and IB(2) are present, summation and subtraction are used. When IF(2) and PIL are available, mid-parental values and mid-parental heterosis are used. One-dimensional genomic scanning is performed to detect the additive and dominant QTLs, based on the algorithm of inclusive composite interval mapping (ICIM). The proposed approach was applied to one IF(2) population together with PIL in maize, and identified ten QTLs on ear length, showing varied degrees of dominance. Simulation studies indicated the proposed approach is similar to or better than individual population mapping by QTL detection power, false discovery rate (FDR), and estimated QTL position and effects.