Cargando…

Characterization of bidirectional gene pairs in The Cancer Genome Atlas (TCGA) dataset

The “bidirectional gene pair” indicates a particular head-to-head gene organization in which transcription start sites of two genes are located on opposite strands of genomic DNA within a region of one kb. Despite bidirectional gene pairs are well characterized, little is known about their expressio...

Descripción completa

Detalles Bibliográficos
Autores principales: Tu, Juchuanli, Li, Xiaolu, Wang, Jianjun
Formato: Online Artículo Texto
Lenguaje:English
Publicado: PeerJ Inc. 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6585903/
https://www.ncbi.nlm.nih.gov/pubmed/31245179
http://dx.doi.org/10.7717/peerj.7107
_version_ 1783428800471629824
author Tu, Juchuanli
Li, Xiaolu
Wang, Jianjun
author_facet Tu, Juchuanli
Li, Xiaolu
Wang, Jianjun
author_sort Tu, Juchuanli
collection PubMed
description The “bidirectional gene pair” indicates a particular head-to-head gene organization in which transcription start sites of two genes are located on opposite strands of genomic DNA within a region of one kb. Despite bidirectional gene pairs are well characterized, little is known about their expression profiles and regulation features in tumorigenesis. We used RNA-seq data from The Cancer Genome Atlas (TCGA) dataset for a systematic analysis of the expression profiles of bidirectional gene pairs in 13 cancer datasets. Gene pairs on the opposite strand with transcription end site distance within one kb or on the same strand with the distance of two genes between 1–10 kb and gene pairs comprising two randomly chosen genes were used as control gene pairs (CG1, CG2, and random). We identified and characterized up-/down-regulated genes by comparing the expression level between tumors and adjacent normal tissues in 13 TCGA datasets. There were no consistently significant difference in the percentage of up-/down-regulated genes between bidirectional and control/random genes in most of TCGA datasets. However, the percentage of bidirectional gene pairs comprising two up- or two down-regulated genes was significantly higher than gene pairs from CG1/2 in 12/11 analyzed TCGA datasets and the random gene pairs in all 13 TCGA datasets. Then we identified the methylation correlated bidirectional genes to explore the regulatory mechanism of bidirectional genes. Like the differentially expressed gene pairs, the bidirectional genes in a pair were significantly prone to be both hypo- or hyper-methylation correlated genes in 12/13 TCGA datasets when comparing to the CG2/random gene pairs despite no significant difference between the percentages of hypo-/hyper-methylation correlated genes in bidirectional and CG2/random genes in most of TCGA datasets. Finally, we explored the correlation between bidirectional genes and patient’s survival, identifying prognostic bidirectional genes and prognostic bidirectional gene pairs in each TCGA dataset. Remarkably, we found a group of prognostic bidirectional gene pairs in which the combination of two protein coding genes with different expression level correlated with different survival prognosis in survival analysis for OS. The percentage of these gene pairs in bidirectional gene pair were significantly higher than the gene pairs in controls in COAD datasets and lower in none of 13 TCGA datasets.
format Online
Article
Text
id pubmed-6585903
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher PeerJ Inc.
record_format MEDLINE/PubMed
spelling pubmed-65859032019-06-26 Characterization of bidirectional gene pairs in The Cancer Genome Atlas (TCGA) dataset Tu, Juchuanli Li, Xiaolu Wang, Jianjun PeerJ Bioinformatics The “bidirectional gene pair” indicates a particular head-to-head gene organization in which transcription start sites of two genes are located on opposite strands of genomic DNA within a region of one kb. Despite bidirectional gene pairs are well characterized, little is known about their expression profiles and regulation features in tumorigenesis. We used RNA-seq data from The Cancer Genome Atlas (TCGA) dataset for a systematic analysis of the expression profiles of bidirectional gene pairs in 13 cancer datasets. Gene pairs on the opposite strand with transcription end site distance within one kb or on the same strand with the distance of two genes between 1–10 kb and gene pairs comprising two randomly chosen genes were used as control gene pairs (CG1, CG2, and random). We identified and characterized up-/down-regulated genes by comparing the expression level between tumors and adjacent normal tissues in 13 TCGA datasets. There were no consistently significant difference in the percentage of up-/down-regulated genes between bidirectional and control/random genes in most of TCGA datasets. However, the percentage of bidirectional gene pairs comprising two up- or two down-regulated genes was significantly higher than gene pairs from CG1/2 in 12/11 analyzed TCGA datasets and the random gene pairs in all 13 TCGA datasets. Then we identified the methylation correlated bidirectional genes to explore the regulatory mechanism of bidirectional genes. Like the differentially expressed gene pairs, the bidirectional genes in a pair were significantly prone to be both hypo- or hyper-methylation correlated genes in 12/13 TCGA datasets when comparing to the CG2/random gene pairs despite no significant difference between the percentages of hypo-/hyper-methylation correlated genes in bidirectional and CG2/random genes in most of TCGA datasets. Finally, we explored the correlation between bidirectional genes and patient’s survival, identifying prognostic bidirectional genes and prognostic bidirectional gene pairs in each TCGA dataset. Remarkably, we found a group of prognostic bidirectional gene pairs in which the combination of two protein coding genes with different expression level correlated with different survival prognosis in survival analysis for OS. The percentage of these gene pairs in bidirectional gene pair were significantly higher than the gene pairs in controls in COAD datasets and lower in none of 13 TCGA datasets. PeerJ Inc. 2019-06-17 /pmc/articles/PMC6585903/ /pubmed/31245179 http://dx.doi.org/10.7717/peerj.7107 Text en © 2019 Tu et al. http://creativecommons.org/licenses/by/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ) and either DOI or URL of the article must be cited.
spellingShingle Bioinformatics
Tu, Juchuanli
Li, Xiaolu
Wang, Jianjun
Characterization of bidirectional gene pairs in The Cancer Genome Atlas (TCGA) dataset
title Characterization of bidirectional gene pairs in The Cancer Genome Atlas (TCGA) dataset
title_full Characterization of bidirectional gene pairs in The Cancer Genome Atlas (TCGA) dataset
title_fullStr Characterization of bidirectional gene pairs in The Cancer Genome Atlas (TCGA) dataset
title_full_unstemmed Characterization of bidirectional gene pairs in The Cancer Genome Atlas (TCGA) dataset
title_short Characterization of bidirectional gene pairs in The Cancer Genome Atlas (TCGA) dataset
title_sort characterization of bidirectional gene pairs in the cancer genome atlas (tcga) dataset
topic Bioinformatics
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6585903/
https://www.ncbi.nlm.nih.gov/pubmed/31245179
http://dx.doi.org/10.7717/peerj.7107
work_keys_str_mv AT tujuchuanli characterizationofbidirectionalgenepairsinthecancergenomeatlastcgadataset
AT lixiaolu characterizationofbidirectionalgenepairsinthecancergenomeatlastcgadataset
AT wangjianjun characterizationofbidirectionalgenepairsinthecancergenomeatlastcgadataset