Characterizing rhizome bud dormancy in Polygonatum kingianum: Development of novel chill models and determination of dormancy release mechanisms by weighted correlation network analysis.

This study was conducted to explore specific chill models and the mechanisms underlying rhizome bud dormancy break in Polygonatum kingianum. Rhizome buds were subjected to various chilling temperatures for different duration and then transferred to warm conditions for germination and subsequent evaluation of their response to temperature and chilling requirements. A CUkingianum model was constructed to describe the contribution of low temperature to the chill unit, and it was suggested that 2.97°C was the optimum temperature and that 11.54°C was the upper limit for bud release. The CASkingianum model showed the relationship between chilling accumulation and sprouting percentage; therefore, rhizome bud development could be predicted through the model. Weighted correlation network analysis (WGCNA) of transcriptomic data of endo-, eco- and nondormant rhizome buds generated 33 gene modules, 6 of which were significantly related to bud sprouting percentage. In addition, 7 significantly matched transcription factors (TFs) were identified from the promoters of 17 "real" hub genes, and DAG2 was the best matched TF that bound to AAAG element to regulate gene expression. The current study is valuable for developing a highly efficient strategy for seedling cultivation and provides strong candidates for key genes related to rhizome bud dormancy in P. kingianum.

HighSSR: high-throughput SSR characterization and locus development from next-gen sequencing data.

MOTIVATION: Microsatellites are among the most useful genetic markers in population biology. High-throughput sequencing of microsatellite-enriched libraries dramatically expedites the traditional process of screening recombinant libraries for microsatellite markers. However, sorting through millions of reads to distill high-quality polymorphic markers requires special algorithms tailored to tolerate sequencing errors in locus reconstruction, distinguish paralogous loci, rarify raw reads originating from the same amplicon and sort out various artificial fragments resulting from recombination or concatenation of auxiliary adapters. Existing programs warrant improvement. RESULTS: We describe a microsatellite prediction framework named HighSSR for microsatellite genotyping based on high-throughput sequencing. We demonstrate the utility of HighSSR in comparison to Roche gsAssembler on two Roche 454 GS FLX runs. The majority of the HighSSR-assembled loci were reliably mapped against model organism reference genomes. HighSSR demultiplexes pooled libraries, assesses locus polymorphism and implements Primer3 for the design of PCR primers flanking polymorphic microsatellite loci. As sequencing costs drop and permit the analysis of all project samples on next-generation platforms, this framework can also be used for direct simple sequence repeats genotyping. AVAILABILITY: http://code.google.com/p/highssr/