A comparison of interval mapping procedures including the QTL-cluster mapping
スポンサーリンク
概要
- 論文の詳細を見る
Using the deterministic sampling, patterns of the log-likelihood surfaces expected in some interval mapping procedures for estimating the position of, and the effect for, QTL(s) were investigated for the situations where a single QTL or closely linked QTLs are contained in a chromosome segment bracketed with two markers. The mapping procedures compared were the conventional, likelihood-based interval mapping (IM), the regression interval mapping (RIM), and the QTL-cluster mapping (CM) in which the conditional probabilities of transmission of the whole segment marked by the flanking markers were taken into consideration. The half-sib design was used here, and several cases of the true genetic model were considered, differing in the number of QTLs contained in the marker interval, the linkage phase for the sire, and the magnitude of the QTL(s) effect. For the true genetic models where a single QTL or closely linked QTLs being in coupling phase are contained in the interval, with (R)IM, clear global maxima of the log-likelihood were observed within the range of the marker interval. It was shown that the estimates of the QTL(s) effect at the marked segment level are expected to be unbiased. On the other hand, in a setting where the linkage phase for the linked QTLs located in the interval was different from coupling and repulsion, there was found a ridge along the interval for the log-likelihood surface with (R)IM, indicating the dependency between the estimates of the position of, and the effect for, the putative QTL. For this case, it was found that the position of the putative QTL could be estimated as that of one of the flanking markers, and the estimate of the QTL effect be biased. In contrast, it was revealed that CM is expected to provide the unbiased estimate of the QTL(s)-effect at the segment level for any case of the true genetic models considered here. If the aim is for marker-assisted selection rather than mapping closely linked QTLs individually, then the CM approach is considered to be useful.
- 2005-02-01
著者
-
Iwaisaki Hiroaki
Course Of Environmental Management Science Graduate School Of Science And Technology Niigata Univers
-
Matsuda Hirokazu
Course Of Environmental Management Science Graduate School Of Science And Technology Niigata Univers
-
Matsuda Hirokazu
Course Of Environmental Management Science Graduate School Of Science And Technology Niigata Univers
-
Matsuda Hirokazu
Course Of Environmental Management Science Graduate School Of Science And Technology
-
Iwaisaki Hiroaki
Course Of Environmental Management Science Graduate School Of Science And Technology
関連論文
- A comparison of interval mapping procedures including the QTL-cluster mapping
- Best Linear Unbiased Prediction of QTLs-Cluster Effects Using Flanking and Upstream Marker Information in Outbred Populations
- Relative accuracy expected in marker-assisted combined selection for quantitative characters in outbred populations
- Best linear unbiased prediction of additive genetic merit using a combined-merit sire and dam model for marker-assisted selection
- Calculating Elements of Gametic Relationship Matrix in the Model Containing Effects of Marked QTL-Cluster
- A recursive procedure to compute the gametic relationship matrix and its inverse for marked QTL clusters