This approach gave early benefits but in later generations, without updating marker effects, the estimation of marker effects based on less related reference individuals did not pay off.
The objective of this study was to investigate the impacts of breeding value estimation method, reference population structure, and trait genetic architecture, on long-term response to genomic selection without updating marker effects.
Nevertheless, the effect of updating marker effects is expected to be small, because the additional data becoming available for retraining each generation will usually be smaller than the already existing reference population, and the change in accuracy due to adding records to the reference population shows a diminishing return relationship.
For j in {1, … , p} update marker effects with a draw from a normal density with mean [ ∑ i = 1 n x i j 2 + σ 2 σ β j − 2 ] − 1 ∑ i = 1 n x i j y ˜ ˜ i and variance σ 2 [ ∑ i = 1 n x i j 2 + σ 2 σ β j − 2 ] − 1, where y ˜ ˜ i = y i − μ − ∑ k ≠ j x i k β k and σ β j 2 is the prior conditional variance of the jth marker effect.
(4) Marker j was then updated in marker matrix M by using the Y ^ values as the estimate of the missing values.
This result is contrary to that of Muir [ 10], but he did not update the marker effects every generation, which made the accuracy decay rapidly.
If the BIC score decreased by removing a QTL marker, we updated the model by removing that marker.
Use of a shallow reference population, most closely related to the selection candidates, gave early benefits while in later generations, when marker effects were not updated, the estimation of marker effects based on a deeper reference population did not pay off.
MaizeGDB map pages include views of mapped markers and genotyping scores, updated files for sequence-based markers with assigned map locations, GenBank (17) accession numbers and FASTA-formatted sequences as bulk downloads (see, e.g. the UMC 98 'sequence view' for chromosome 1 at http://www.maizegdb.org/cgi-bin/displaymapwithaccessions.cgi id = 143431).
The QTL mapping study conducted using the Puget × 90 2079 RIL population [ 31, 45] was updated by adding 53 markers (41 SSRs and 12 RAPDs) to the genetic map.
