Mapping Of QTLs Underlying Panicle Length In Rice(Oryza Sativa L.) Using Deep Sequencing

Mapping of quantitative trait loci (QTLs) is a main research content of modern quantitative genetics. The technology of next generation sequencing (or deep sequencing) has the advantage of high throughput and low cost, and has become an efficient tool for genetic and genomic studies. In this study, according to the principle of bulked segregant analysis (BSA), the method of combining deep sequencing and BSA for QTL mapping was investigated, and the feasibility and efficiency of the method was validated with rice panicle length as a practical example.Data source:An F2population (248plants) was created from a cross between a long-panicle line LPBG08and a short-panicle japonica rice cultivar Nipponbare. Fifteen plants with extremely long panicles and extremely short panicles were selected from the population to make a long-panicle and a short-panicle DNA pools, respectively. The two DNA pools were sequenced with the Solexa method, respectively.Methods for data analysis:With the known genome sequence of Nipponbare as reference, the sequence data were filtered, trimmed, mapped and merged and then SNPs were detected using the software package CLC Genomics Workbench4.0. With the large number of SNPs acquired, the difference of allelic frequency (d) between the two pools within a certain window (genome segment) was estimated. As the window moved with a given step length along the genome, a profile of lvalue changing along the genome was obtained. The threshold corresponding to5%overall significance level was estimated using the method of permutation tests. The peaks of d value higher than the threshold are likely to be the positions of QTLs.Main results:1. After filtering and trimming, the long-panicle pool had29,826,403valid reads, of which27,771,525could be mapped to the reference genome, corresponding to a total length of2,030Mb, with an average sequencing depth of5.40x and an average coverage of0.95; the short-panicle pool had28,658,802valid reads, of which 21,771,525could be mapped to the reference genome, corresponding to a total length of1,960Mb, with an average sequencing depth of5.40x and an average coverage of0.95. A total of525,608SNPs were identified with an average density of1.41/kb.2. The whole genome was scanned with a400-kb sliding window and a10-kb step length. The estimated threshold corresponding to5%overall significance level was0.228.3. A total of3QTLs were detected, located in chromosomes3,8and11. The QTL on chromosome3was the most significant.This study has demonstrated the feasibility and efficiency of the QTL mapping method based on a combination of deep sequencing and BSA. An outstanding merit of this method is that it can directly map QTLs onto the physical map. As the deep sequencing technology develops further and its cost continuously decreases, it can be expected that this method will be widely applied in QTL mapping and will become a new developing direction of QTL mapping methodology.