Friday, April 13, 2012

A novel transposase-based, low-input methyl-seq method

The role of DNA methylation in gene regulation and epigenetics is of ever-increasing interest. Until now, investigating DNA methylation has been crippled either by: i) the necessity for microgram quantities of input DNA for whole-genome bisulfite sequencing (WGBS); or ii) incomplete  interrogation of the methylome in the case of reduced-representation bisulfite sequencing (RRBS).

In a recent publication, Adey and Shendure report a transposon-based in vitro library construction method to demonstrate that a highly complex bisulfite sequencing library may be obtained from as little as 10 ng of genomic DNA. The authors describe the use of a custom-loaded transposase for "tagmentation" (tagging and fragmentation of the DNA) that contained an unmethylated transposon sequence. After an oligonucleotide replacement procedure, in which (methylated) adaptors were ligated to each end of the tagmented genomic DNA, standard bisulfite treatment was used. The library was amplified by PCR and sequenced on an Illumina® MiSeq using a single-end, 100-bp run. Over 100 million aligned reads were obtained at high complexity with >96% of CpG and >98% of non-CpG cytosines covered for the genome. The necessity to have part of the transposon DNA unmethylated served as an internal control and demonstrated that the conversion rate of the bisulfite treatment was >99%. The method described will allow for whole-methylome interrogation where DNA sample is limiting (biopsy samples, microdissected tissues, etc.) and may also allow for the utilization of poor-quality or degraded DNA samples.

ResearchBlogging.orgAdey, A., & Shendure, J. (2012). Ultra-low-input, tagmentation-based whole genome bisulfite sequencing Genome Research DOI: 10.1101/gr.136242.111

Friday, April 6, 2012

Evaluation of rRNA removal methods for complex microbial samples

Although much of RNA-Seq research has focused on the human transcriptome, there is considerable interest in applications of RNA-Seq to environmental metatranscriptomics samples. Giannoukos et al. at the Broad Institute evaluated several methods for removal of ribosomal RNA (rRNA) from total RNA preparations of complex microbial communities in order to maximize RNA-Seq reads. They also examined bacterial populations from clinical human stool samples.

The researchers used total RNA from well-characterized microbial culture samples individually, as well as a mixed pool, for benchmarking rRNA removal methods:
  • Epicentre Ribo-Zero™ Meta-Bacteria Kit (solution-based hybridization capture);
  • NuGEN Ovation Prokaryotic RNA-Seq System (NuGEN) (proprietary set of "not so random" primers to avoid rRNA as template during first and second strand cDNA synthesis);
  • Duplex-Specific Nuclease (DSN; degradation of fast re-annealing abundant cDNAs to preferentially deplete ribosomal cDNA);
  • Life Technologies MicrobEXPRESS Kit (solution-based hybridization capture); and
  • Epicentre Terminator Exonuclease (degrades RNAs with a 5'-monophosphate group).
The authors constructed RNA-Seq libraries after using each of the methods above, sequenced them on the Illumina platform, and mapped the reads to the corresponding reference genomes. For the clinical samples, they mapped reads to 649 bacterial reference genomes.

The results demonstrated the superiority of the Ribo-Zero Kit in reducing rRNA in the microbial samples. In artificially degraded samples and in the clinical samples, the Ribo-Zero Kit performed well, while some of the other methods performed poorly or failed completely at removing rRNA. The authors further note the potential for simple automation of the Ribo-Zero procedure, and that the Ribo-Zero process can be applied to profile gene expression in both simple and complex, naturally occurring bacterial communities.

ResearchBlogging.orgGiannoukos, G. et al. (2012). Efficient and robust RNA-seq process for cultured bacteria and complex community transcriptomes Genome Biology, 13 (3) DOI: 10.1186/gb-2012-13-3-r23