Tan G et al. (2019), CNEr: A toolkit for exploring extreme noncoding...

ECB-ART-47396

PLoS Comput Biol 2019 Aug 26;158:e1006940. doi: 10.1371/journal.pcbi.1006940.

Show Gene links Show Anatomy links

CNEr: A toolkit for exploring extreme noncoding conservation.

Tan G , Polychronopoulos D , Lenhard B .

???displayArticle.abstract???
Conserved Noncoding Elements (CNEs) are elements exhibiting extreme noncoding conservation in Metazoan genomes. They cluster around developmental genes and act as long-range enhancers, yet nothing that we know about their function explains the observed conservation levels. Clusters of CNEs coincide with topologically associating domains (TADs), indicating ancient origins and stability of TAD locations. This has suggested further hypotheses about the still elusive origin of CNEs, and has provided a comparative genomics-based method of estimating the position of TADs around developmentally regulated genes in genomes where chromatin conformation capture data is missing. To enable researchers in gene regulation and chromatin biology to start deciphering this phenomenon, we developed CNEr, a R/Bioconductor toolkit for large-scale identification of CNEs and for studying their genomic properties. We apply CNEr to two novel genome comparisons-fruit fly vs tsetse fly, and two sea urchin genomes-and report novel insights gained from their analysis. We also show how to reveal interesting characteristics of CNEs by coupling CNEr with existing Bioconductor packages. CNEr is available at Bioconductor (https://bioconductor.org/packages/CNEr/) and maintained at github (https://github.com/ge11232002/CNEr).

???displayArticle.pubmedLink??? 31449516
???displayArticle.pmcLink??? PMC6730951
???displayArticle.link??? PLoS Comput Biol
???displayArticle.grants??? [+]

Species referenced: Echinodermata
Genes referenced: LOC100887844 LOC115925415 LOC590297 meis1

???attribute.lit??? ???displayArticles.show???

	Fig 1. CNEr workflow.(A) A typical pipeline of identification and visualisation of CNEs. (B) Illustration of scanning an alignment for CNEs. The scanning window moves along the alignment for conserved regions. The exons and repeats regions are skipped during the scanning by default.
	Fig 2. Horizon plot of CNE density around key developmental genes along D. melanogaster as reference.(A) H15 and mid genes are spanned by arrays of CNEs. Despite the much lower CNE density from D. melanogaster and Glossina, a CNE cluster boundary shows up that is consistent with CNEs from other Drosophila species. (B) The CNE cluster around ct gene is missing in the comparison of D. melanogaster and Glossina since no CNEs are detected. This implies that this region undergoes a higher CNE turnover rate. (C, D) The same loci as in (A, B) are shown on the Ancora browser in order to compare the normal CNE density plot with the horizon plot. Notations: ensGene, Ensembl gene track; Glossina 21/30, G. morsitans 70% identity over 30 bp; droAna2 49/50, D. ananassae 98% identity over 50 bp; dp3 48/50, D. pseudoobscura 96% identity over 50 bp; droMoj2 48/50, D. mojavensis 96% identity over 50 bp; droVir2 48/50, D. virilis 96% identity over 50 bp.
	Fig 3. The number of CNEs between D. melanogaster and several other species, including G. morsitans.70% over 30 bp, 80% over 30 bp, 90% over 30 bp in vs. D. ananassae, 70% over 30 bp in vs. D. pseudoobscura are not counted due to the threshold being too low for close species.
	Fig 4. Over-represented GO biological process terms ranked by GeneRatio.The gene ratio is defined as the number of genes associated with the term in our selected genes divided by the number number of selected genes. The universe for over-representation test is all Drosophila with GO annotations. The p-values are adjusted using "BH" (Benjamini-Hochberg) correction and the cutoff is 0.05. The visualisation is done by clusterProfiler [35]. (A) GO enrichment for genes nearest to Drosophila and Glossina CNEs. (B) GO enrichment for genes in the missing CNEs clusters compared between Drosophila and Glossina.

	Fig 6. Horizon plot of CNE density at Meis loci on sea urchin Strongylocentrotus purpuratus.The density plots of CNEs detected at similarity threshold 96% (48/50), 98% (49/50) and 100% (50/50) over 50 bp sliding window, in Lytechinus variegatus comparison, are shown in three horizon plot tracks. The boundaries of CNE clusters from various thresholds are mutually consistent.
	Fig 7. Pairwise overlap analysis of CNEs demonstrates association with (A) Sox2, (B) Pou5f1 and (C) Nanog binding regions. In all three cases, permutation tests with 1000 permutations of CNEs are shown. The number of overlaps of the randomized regions with the test regions of interest (in this case, Sox2, POU5F1 and Nanog) are depicted in grey. Those overlaps of the randomized regions cluster around the black bar that represents the mean. In green, the number of overlaps of the actual regions (CNEs in this case) with the test regions is shown and is proved to be much larger than expected in all cases. The red line denotes the significance limit.

References [+] :

Anderson, Mapping the Shh long-range regulatory domain. 2014, Pubmed