Researchers at the University of Washington and FHCRC have developed a new method to measure and correct for protocol-specific sequence bias for quantification of sequence abundance in RNA-Seq experiments using a simple graphical model. Their model does not rely on existing gene annotations, and model selection is performed automatically making it applicable with few assumptions. They have evaluated the method on several data sets, and by multiple criteria, and demonstrate that it effectively decreases bias and increases uniformity.
Availability: The method is freely available under the LGPL license at: http://bioconductor.org/packages/release/bioc/html/seqbias.html
- Jones DC, Ruzzo WL, Peng X, Katze MG. (2012) A new approach to bias correction in RNA-Seq. Bioinformatics [Epub ahead of print]. [abstract]