Does anyone know why it is important in CLIPseq/RNAseq to exclude reads mapped to repetitive elements from further analysis downstream (such as enrichment)? Would retaining reads mapped to repetitive elements affect normalization and downstream enrichment analysis of a gene with DEseq2?
An example: the Repbase database is used by the eCLIP paper (van Nostrand, 2016) for mapping repetitive elements against their libraries and the mapped reads are removed there (picture).
Also, does anyone know a good alternative to REPbase which is free or less expensive?
Answers on these questions would be greatly appreciated!