IntroductionFollowing the completion of the human genome a decade ago, we came across a surprisingly low number of proteincoding genes in the human genome compared to the initial predictions. The low number of protein-coding genes, corresponding to only 2% of the human genome, was surprising, because it meant that we were undervaluing the potential roles of alternative promoter usage, alternative splicing, alternative polyadenylation, and so on to explain the vast diversity of proteins in cells. It was also a realization that the rest of the genome must also have some role, either structural or functional. Indeed, over the years, studies have documented widespread transcription across 70%-90% of the genome with approximately 9000 small RNAs, 1100 microRNAs (miRNAs; http://microRNA.org), 32,000 long noncoding RNAs (lncRNAs), and 11,000 expressed pseudogenes (Bernstein et al., 2012;Volders et al., 2013). Based on increasing evidence, it has become clear that this nonprotein-coding portion of the genome is critical in managing the greater complexity of higher eukaryotes during development and in different physiological and/ or disease states, especially cancer. While most of the focus is on miRNAs as small noncoding RNAs, other long or small noncoding RNAs are attracting attention in normal and in cancer cells. Here we provide an overview of currently known noncoding RNA elements (Figure 1), their functions, and their connections with disease states, especially cancer (Table ). In the first part, lncRNAs and their subclasses will be discussed. The second part focuses on small noncoding RNAs.
Long noncoding RNA classeslncRNAs are transcripts of longer than ~200 nucleotides with little or no protein-coding capacity (it is possible that some lncRNAs encode small peptides). In humans, with exceptions, lncRNAs are generally polyadenylated and are spliced to generate a small number of exonic regions (Derrien et al., 2012). On the other hand, evidence suggests some lncRNAs to be processed and stabilized by other mechanisms such as RNase P cleavage to generate a mature 3' end, capping by snoRNP complexes at both ends, or formation of circular structures (Zhang et al., 2013a).The lncRNA genes are positioned throughout various chromosomal regions of the genome. Therefore, the accepted classification of lncRNAs is generally based on their locations, whether they are intergenic, intronic, or found in the antisense strand [also known as natural antisense transcripts (NATs)] or upstream to annotated protein-coding genes. With the advancements of sequencing methods, along with other noncoding RNAs,