Cargando…

Construction of mate pair full-length cDNAs libraries and characterization of transcriptional start sites and termination sites

To identify and characterize transcript structures ranging from transcriptional start sites (TSSs) to poly(A)-addition sites (PASs), we constructed and analyzed human TSS/PAS mate pair full-length cDNA libraries from 14 tissue types and four cell lines. The collected information enabled us to define...

Descripción completa

Detalles Bibliográficos
Autores principales: Matsumoto, Kyoko, Suzuki, Ayako, Wakaguri, Hiroyuki, Sugano, Sumio, Suzuki, Yutaka
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2014
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4176323/
https://www.ncbi.nlm.nih.gov/pubmed/25034687
http://dx.doi.org/10.1093/nar/gku600
Descripción
Sumario:To identify and characterize transcript structures ranging from transcriptional start sites (TSSs) to poly(A)-addition sites (PASs), we constructed and analyzed human TSS/PAS mate pair full-length cDNA libraries from 14 tissue types and four cell lines. The collected information enabled us to define TSS cluster (TSC) and PAS cluster (PAC) relationships for a total of 8530/9400 RefSeq genes, as well as 4251/5618 of their putative alternative promoters/terminators and 4619/4605 intervening transcripts, respectively. Analyses of the putative alternative TSCs and alternative PACs revealed that their selection appeared to be mostly independent, with rare exceptions. In those exceptional cases, pairs of transcript units rarely overlapped one another and were occasionally separated by Rad21/CTCF. We also identified a total of 172 similar cases in which TSCs and PACs spanned adjacent but distinct genes. In these cases, different transcripts may utilize different functional units of a particular gene or of adjacent genes. This approach was also useful for identifying fusion gene transcripts in cancerous cells. Furthermore, we could construct cDNA libraries in which 3′-end mate pairs were distributed randomly over the transcripts. These libraries were useful for assembling the internal structure of previously uncharacterized alternative promoter products, as well as intervening transcripts.