Construction Of Chromosome Conformation Capture Database And A Potential Method For CircRNA Identification

Chromosome conformation plays an important role in gene regulation. Chromosome Conformation Capture (3C) is one of the most important technologies to study chromatin interaction between gene sites. For the lack of high-resolution chromatin interaction database, we constructed a database of 3C (3CDB; http://3cdb.big.ac.cn). We have searched more than 2000 published articles containing 3C technology in PubMed and over 3000 articles in Google Scholar since 3C technology was first proposed in 2002.482 articles with useful information of 3C results were achieved by scanning the full text. From these articles, we extracted 3319 interactions manually in 308 cell lines from 17 species, with total 73 restriction enzymes and their combinations. Moreover, for inconsistent 3C experiment protocols in different labs and variety of 3C results, we developed 3C data evaluation system to assess the reliability and precision of each data in 3CDB. To a certain extent, our 3CDB will be able to fill in the blank of chromatin conformation database. It provides important resources for precise positioning of chromatin interaction and building exact models between the distal regulatory elements.Circular RNA (circRNA) is a class of RNA molecules with two ends,3’, 5’-phosphodiester bond covalently linked to form a circle. The circRNAs have been proven to be a class of abundant, stable and ubiquitous non-coding RNA. Most circRNAs are composed of exons, while only a small part of introns, and partly by the combination of them. Almost all circRNAs are located in the cytoplasm except a few in nucleus. Current evidences suggest that the coding potential for circRNAs was limited. At present, there are some tools having been developed for circRNA identification, such as circRNA_finder, CIRCexplorer, CIRI, find_circ and Mapsplice. These tools are based on the presence of back-splicing junction reads in the RNA-seq data. However, since the few number of back-splicing junction reads, it is unreliable for traditional tools to compute circRNAs expression levels. Besides finding back-splicing junction reads, we consider that the storage order in Fastq of paired-end reads from Illumina RNA-seq may also provide additional information for circling. We found that the chiastic storage paired-end reads flanking the back-splicing junction site were as many as the back-splicing junction reads. Therefore, we proposed a draft scheme to identify circRNA, called CircRNA Identified by Chiastic Alignment(CICA). CICA will be an important complement to the current circRNA identification algorithms and circRNA expression analysis.