TELP, a sensitive and versatile library construction method for next-generation sequencing.
Bottom Line: Next-generation sequencing has been widely used for the genome-wide profiling of histone modifications, transcription factor binding and gene expression through chromatin immunoprecipitated DNA sequencing (ChIP-seq) and cDNA sequencing (RNA-seq).By minimizing the DNA purification steps that cause major sample loss, our method achieved a high sensitivity in ChIP-seq library preparation.Using this method, we achieved the following: (i) generated high-quality epigenomic and transcription factor-binding maps using ChIP-seq for murine adipocytes; (ii) successfully prepared a ChIP-seq library from as little as 25 pg of starting DNA; (iii) achieved paired-end sequencing of the ChIP-seq libraries; (iv) systematically profiled gene expression dynamics during murine adipogenesis using RNA-seq and (v) preserved the strand specificity of the transcripts in RNA-seq.
Affiliation: Singapore Institute for Clinical Sciences, Agency for Science, Technology and Research (A*STAR), Singapore 117609, Singapore.Show MeSH
Related in: MedlinePlus
Mentions: Strand-specific RNA sequencing can add substantial value to the RNA-seq data sets, especially in applications such as the identification of regulatory antisense transcripts (21). However, many RNA-seq library construction methods do not preserve information regarding which strand was transcribed. This is mainly due to the synthesis of double-stranded cDNA using random primers. Existing strand-specific RNA-seq protocols involve extra steps such as selective labeling of the cDNA strand with dUTP (22) or bisulfite conversion of cytosines to uracils in RNA (23). As TELP is compatible with library preparation from both ssDNA and dsDNA, we tested whether it could maintain the strand specificity in RNA-seq. As an initial test, we constructed mimic sequencing libraries from both dsDNA and ssDNA using TELP (Figure 6A). After addition of the anchor primer and adapter sequences, the resulting mimic libraries were 92 bp longer than the original DNA fragments (Figure 6B). Both ssDNA and dsDNA mimic libraries were cloned and subsequently sequenced by Sanger sequencing. The sequencing results showed that all 13 clones from the ssDNA mimic library contain the same positive strand sequence. In contrast, nine clones from the dsDNA mimic library contain either the positive or the negative strand sequence at a random rate (four versus five). We conclude that the strand-specific information was faithfully maintained by TELP in these mimic libraries (Figure 6C).
Affiliation: Singapore Institute for Clinical Sciences, Agency for Science, Technology and Research (A*STAR), Singapore 117609, Singapore.