A novel reannotation strategy for dissecting DNA methylation patterns of human long intergenic non-coding RNAs in cancers.
Bottom Line: We classified 2461 lincRNAs into two categories and three subcategories, according to their promoter methylation patterns in tumors.LincRNAs with resistant methylation patterns in tumors had conserved transcriptional regulation regions and were ubiquitously expressed across normal tissues.The methylated and demethylated lincRNAs identified in this study provide novel insights for developing cancer biomarkers and potential therapeutic targets.
Affiliation: College of Bioinformatics Science and Technology, Harbin Medical University, Harbin 150081, China.Show MeSH
Related in: MedlinePlus
Mentions: We combined the lincRNA methylation profiles with clinical annotations and identified a subset of lincRNAs with methylation values showing a trend associated with OS in BRCA, LUSC and UCEC. We used a validation set as an independent data set to validate candidate reliability. For UCEC, we obtained 23 PA lincRNAs in the validation set from 30 lincRNAs in the discovery set (FDR < 0.01). For BRCA, we validated five lincRNAs associated with OS from the top 10 lincRNAs in the discovery set (FDR < 0.01). For LUSC, we obtained seven PA lincRNAs (Supplementary Tables S7–S9). For example, BRCA patients with lower methylation level of lincRNA XLOC_009284 had better prognosis (Figure 5A). LUSC patients with relatively lower methylation level of lincRNA XLOC_009367 showed poorer prognosis (Figure 5B). For UCEC, patients with the highest methylation level of lincRNA XLOC_007617 had a better prognosis than patients with lower methylation level (Figure 5C). GAS5, a lincRNA linked to apoptosis that is involved in progression of some types of cancers, was significantly correlated with prognosis for UCEC (53,54). The lincRNA MEG3 was associated with OS in the LUSC discovery set but not in the validation set, which inhibits proliferation of non-small cell lung cancer cells and induces apoptosis by affecting p53 expression (55). We hope to further validate candidates from the discovery set in the future, using a suitable tumor set. Besides, BRCA1, the ovarian cancer marker PCG, showed no correlation between methylation level and OS in a previous study (14). There were 42 lincRNAs showed a correlation between the methylation level and OS in BRCA, LUSC, UCEC, KIRC and LGG, some of which showed a negative correlation between methylation and expression in corresponding tumors (Pearson's correlation test; Supplementary Table S10), suggesting their potential as novel prognostic biomarkers.
Affiliation: College of Bioinformatics Science and Technology, Harbin Medical University, Harbin 150081, China.