Abstract

Tumorigenesis is a multi-stage, dynamic biological process that involves several genetic and epigenetic changes, aberrant non-coding RNA expression, and modifications to the expression profiles of coding genes. We refer to this group of genome-space alterations as the "cancer initiatome." In the genome, long non-coding RNAs are widely expressed and have important regulatory roles in chromatin remodelling and gene regulation. In both normal development and pathological conditions, such as cancer, spatial and temporal heterogeneity in lncRNA expression has been noted. Even though several dysregulated lncRNAs have been examined in malignancies, it is still unclear how lncRNAs contribute to the development of cancer, particularly in the case of esophageal squamous cell carcinoma. From ESCC and matched nearby non-cancerous normal tissues, we performed a genome-wide screen to determine the expression of lncRNAs and coding RNAs. In comparison to matched normal tissue equivalents, we discovered which lncRNAs and coding RNAs were differently expressed in ESCC. Using polymerase chain reaction analysis, we confirmed the conclusion. Additionally, we discovered lncRNAs that are differentially expressed in ESCC and that are co-localized and expressed with differentially expressed coding RNAs. These findings suggest a possible interaction between lncRNAs and nearby coding genes that affects ether lipid metabolism and that this interaction may help to develop ESCC. These findings give strong support for a potential new genetic biomarker of esophageal squamous cell carcinoma.

Minimizing free energy, which is NP-hard, is frequently used to predict RNA secondary structures with pseudoknots. During transcription from DNA into RNA, the majority of RNAs fold in a hierarchical manner where secondary structures emerge before tertiary structures. Because of kinetics, local optimization is frequently used in real RNA secondary structures rather than global optimization. By taking dynamic and hierarchical folding mechanisms into account, the accuracy of RNA structure prediction may be increased. Based on a statistical examination of the actual RNA secondary structures of all 480 sequences from RNA STRAND, which are verified by NMR or X-ray, this study presents a fresh report on RNA folding that is consistent with the golden mean feature. With L standing for the sequence length, the length ratios of the domains in these sequences are roughly 0.382L, 0.5L, 0.618L, and L. The key golden sections of the sequence are just these locations. This feature allows for the building of an algorithm that simulates RNA folding by dynamically folding RNA structures in accordance with the aforementioned golden section points while also predicting RNA hierarchical structures. The Mfold, HotKnots, McQfold, ProbKnot, and Lhw-Zhu algorithms cannot match our algorithm's sensitivity and quantity of predicted pseudoknots. As a result of a novel perspective that is near to natural folding, experimental results follow the RNA folding regulations.