Skip to main content
. 2016 Jun 10;25(14):3125–3141. doi: 10.1093/hmg/ddw154

Figure 1.

Figure 1.

Filtering and merging pipeline to create a comprehensive human lincRNA dataset. After pooling the four original datasets, lincRNAs were filtered based on length and overlap with pseudogenes and protein-coding genes. After filtering, they were merged on the basis of exon structure, to produce a final comprehensive set of 54 944 lincRNAs that was used in all subsequent analyses.