Open AccessJoint Analysis of Near‐Isogenic and Recombinant Inbred Line Populations Yields Precise Positional Estimates for Quantitative Trait Loci
Author(s) -
Kump Kristen L.,
Holland James B.,
Jung Mark T.,
Wolters Petra,
BalintKurti Peter J.
Publication year - 2010
Publication title -
the plant genome
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.403
H-Index - 41
ISSN - 1940-3372
DOI - 10.3835/plantgenome2010.05.0011
Subject(s) - quantitative trait locus , biology , family based qtl mapping , genetics , inbred strain , trait , allele , inclusive composite interval mapping , gene mapping , gene , chromosome , computer science , programming language
Data generated for initial quantitative trait loci (QTL) mapping using recombinant inbred line (RIL) populations are usually ignored during subsequent fine‐mapping using near‐isogenic lines (NILs). Combining both datasets would increase the number of recombination events sampled and generate better position and effect estimates. Previously, several QTL for resistance to southern leaf blight of maize were mapped in two RIL populations, each independently derived from a cross between the lines B73 and Mo17. In each case the largest QTL was in bin 3.04. Here, two NIL pairs differing for this QTL were derived and used to create two distinct F 2:3 family populations that were assessed for southern leaf blight (SLB) resistance. By accounting for segregation of the other QTL in the original RIL data, we were able to combine these data with the new genotypic and phenotypic data from the F 2:3 families. Joint analysis yielded a narrower QTL support interval compared to that derived from analysis of any one of the data sets alone, resulting in the localization of the QTL to a less than 0.5 cM interval. Candidate genes identified within this interval are discussed. This methodology allows combined QTL analysis in which data from RIL populations is combined with data derived from NIL populations segregating for the same pair of alleles. It improves mapping resolution over the conventional approach with virtually no additional resources. Because data sets of this type are commonly produced, this approach is likely to prove widely applicable.