RNA-Seq: Modeling non-uniformity in short-read rates in RNA-Seq data.

[Newsbot!]

Syndicated from PubMed RSS Feeds

Modeling non-uniformity in short-read rates in RNA-Seq data.

Genome Biol. 2010 May 11;11(5):R50

Authors: Li J, Jiang H, Wong WH

ABSTRACT: After mapping, RNA-Seq data can be summarized by a sequence of read counts commonly modeled as Poisson variables with constant rates along each transcript, which actually fit data poorly. We suggest using variable rates for different positions, and propose two models to predict these rates based on local sequences. These models explain more than 50% of the variations and can lead to improved estimates of gene and isoform expressions for both Illumina and Applied Biosystems (ABI) data.

PMID: 20459815 [PubMed - as supplied by publisher]



More...

[pzumbo]

As reported in, 'Biases in Illumina transcriptome sequencing caused by random hexamer priming' (Hansen et al.), the FRT-seq protocol (Mamanova et al) does not suffer from uniformity issues, 'Mamanova et al. recently described an alternative protocol for sequencing RNA using the Illumina Genome Analyzer, in which reverse transcription takes place directly on the flow cell and which yields stranded reads and avoids polymerase chain reaction amplification. RNA is not converted to dscDNA using random priming, instead sequencing adapters are ligated directly onto RNA fragments. The ligated RNA library is then reverse transcribed on the flow cell. Data from their study does not show the nucleotide biases reported here.'

Although intuitively Illumina's directional RNA-seq protocol should not have nucleotide biases neither (particularly if the bias is indeed a result of random primers), I checked against a directional and nonDirectional dataset from the same sample (results are attached ; the plots are of frequency against position).

Perhaps Illumina's directional mRNA-seq protocol, then, without the need for any of the various adjustments as suggested in either Hansen et al or Li et al, is a better prep (especially in light of the fact that it preserves strand information) -- ?

--Paul

[combiochem]

Wow it's amazing. It seems that the new protocol blows away the problems with random priming and PCR amplification.