Influence Of QTL Mapping Methods And Correlative Factors On Mapping Effect

The identification and precise mapping of quantitative trait loci (QTLs) are the preconditions for understanding the genetic mechanism of quantitative traits and establishing feasible breeding program. The effect of QTL mapping was influenced by many factors including population structure and size, number of QTLs, density of genetic markers, heritability of traits etc., especially the genetic and statistic models, so it is essential to select appropriate method according to actual situations before mapping QTLs. In this study, backcross population for QTL mapping was constructed by computer simulation in order to study power of detection and mapping precision of QTLs at different levels of factors. The considered factors included population size (200, 500, 1000), heritability (0.2, 0.4, 0.5, 0.8), marker density (2cM, 5cM, 7cM, 10cM) and number of QTL (1, 4, 6, 9). Interval mapping (IM), composite interval mapping (CIM) and multiple interval mapping (MIM) were used to detect QTLs and estimate parameters with the same simulative result. Simultaniously, comparison of results from different methods was carried out. The results showed that population size influenced effects of QTL mapping by IM and CIM, but no apparent influence was found for MIM. When population size was increased, the accuracy of estimates for QTL position and effects and the rate of identified QTL were improved. Heritability had a greater influence on QTL mapping by CIM and MIM, but a little by IM, especially the estimates of QTL effects. The estimates of QTL effects and LR value were almost not influenced by marker density, but it had a greater influence on the estimation of QTL position. When the distance between markers was reduced, the precision of estimation for QTL position was increased and the results from the three methods became similar. When only one QTL exists on chromosome, interval mapping would obtain a better result; but when many linked QTLs exist, poor results were obtained by IM. MIM expressed superiority when more QTLs were considered, and CIM method was appropriate when QTL number is moderate. With the epistasis existed, MIM would obtain better estimates of QTL additive effects.