Chunk #47 — Materials and Methods — LincRNA Discovery — Overlapping transcripts passing all filters at each expression cutoff were merged and grouped by proximity
After merging these expression filtered, overlapping lincRNAs, FPKMs were recalculated (Dataset S8) for the merged lincRNAs using the modified HTSeq program described above. Due to the incomplete nature of the lincRNA structures resulting from de novo assembly, overlapping and nearby lincRNAs were considered to represent different potential models of the same lincRNA gene (rather than isoforms). Therefore, in the rare instances where two or more lincRNA models partially overlap but do not satisfy our merging criteria (above), the reads mapping to these overlapping portions were fully assigned to each lincRNA.
