Pluto Bioinformatics

GSE63285: Kdm5b depletion in embryonic stem cells does not induce intragenic promoters or affect transcription [RNA-Seq]

Bulk RNA sequencing

The application of massively-parallel sequencing (MPS) technology to understanding how the genome is regulated is of major importance, as we now appreciate sequence variants associated with common heritable diseases to be enriched at regulatory elements in the genome. There is, however, an enormous challenge inherent to these studies, which are much more varied as assays and in terms of analytical approaches than the approaches used to define the sequence variants themselves. For example, there areis a plethora of assays that test 5-methylcytosine genome-wide, numerous different analytical approaches to defineing peaks from chromatin immunoprecipitation sequencing (ChIP-seq), and increasingly diverse aspects of genomic regulation being studied. The extremely large data sets generated do not lend themselves to distribution without some degree of pre-processing, and the hardware requirements for running these analyses are often substantial. We have highlighted these concerns previously, but the central concern for our field is that genomics research is especially prone to poor reproducibility, and is a prime candidate for implementation of measures that could help to address this general problem.To illustrate the practical issues involved in genomics reproducibility, we sought to analyze our and three previoulsy published studies of the histone demethylase KDM5B using the same analytical approaches. SOURCE: Andrew Johnston (Andrew.Johnston@med.einstein.yu.edu) - Price 314 Albert Einstein College of Medicine