Until recently, only protein-coding gene sets were used in characterizing general promoter features. Therefore, some widely accepted promoter features and chromatin state signatures may be biased as a consequence of having been inferred from protein-coding genes. In this context, it is hardly surprising that certain sequence and epigenetic features, more specific for protein-coding genes, are less pronounced at lncRNA promoters, while the chromatin states associated with lncRNA promoters are predominantly labeled as inactive promoters. However, these labels were based on manual annotation by biologists, predominantly of protein-coding and intergenic regions [68]. Hence, while it is true that genomic regions with these state-labels tend to be transcriptionally less active than protein coding regions on average in cell lines and tissues explored to date, this does not exclude the possibility that there may exist novel chromatin states associated with lncRNA promoters that have yet to be identified by genome-wide Hidden Markov Model (HMM) based chromatin studies. One of the reasons for this may be a low level of all histone modification signals in lncRNA gene promoters corresponding to low expression of lncRNAs,
