|Holland, Jim - Jim|
Submitted to: Maize Genetics Conference Abstracts
Publication Type: Abstract only
Publication Acceptance Date: 1/5/2005
Publication Date: 3/10/2005
Citation: Tarter, J., Holland, J.B. 2005. Detecting epistasis through the use of double-introgression near-isogenic lines of maize. Maize Genetics Conference Abstracts. Interpretive Summary: Epistasis is the interaction between the effects of genes on a common phenotype. The importance of epistasis compared to the direct effects of genes on important agricultural traits in plants is controversial. We developed a novel approach to test the importance of epistasis in maize. We created lines that were 95% or more identical to inbred B73, except that they had different blocks of genes substituted from inbred Tx303. We combined these blocks to make new lines that had both blocks together and compared the effects of the paired chromosome block substitutions to their individual effects. We found that epistasis was often significant, but less frequently and with lower magnitude than direct genetic effects.
Technical Abstract: Epistasis is an interaction between loci that is observed on some phenotype. Detecting epistatic interactions in typical mapping populations is confounded by segregation of the genetic background. This problem can be reduced through the use of near-isogenic lines (NILs). Objectives of this study were to test for evidence of epistatic interactions, their frequencies, and magnitudes in both inbred and hybrid combinations of maize. Using marker-assisted selection, 127 double introgression near-isogenic lines (dNILs) were developing using B73 as the recurrent parent and two introgressions from the donor parent Tx303. Epistasis was detected by comparing dNILs to their specific parental single-introgression NILs. After the first year of testing, 54 dNIL inbred lines (43%) exhibited epistasis for all six agronomic traits measured. Epistasis for only one trait was detected in 32 lines (25%), and epistasis for two traits simultaneously was detected in 17 lines (14%). In the hybrid trials, 37 dNILs (29%) exhibited epistasis for eight of the ten agronomic traits measured. Twenty-five lines (20%) exhibited epistasis for one trait while 10 lines (8%) exhibited epistasis for two traits simultaneously. Only 5 epistatic genetic combinations were in common between the inbred and hybrid lines.