LINE retrotransposons characterize mammalian tissue-specific and evolutionarily dynamic regulatory regions

Posted on by

Description

To investigate the mechanisms driving regulatory evolution across tissues, we experimentally mapped promoters, enhancers, and gene expression in liver, brain, muscle, and testis from ten diverse mammals. The regulatory landscape around genes included both tissue-shared and tissue-specific regulatory regions, where tissue-specific promoters and enhancers evolved most rapidly. Genomic regions switching between promoters and enhancers were more common across species, and less common across tissues within a single species. Long Interspersed Nuclear Elements (LINEs) played recurrent evolutionary roles: LINE L1s were associated with tissue-specific regulatory regions, whereas more ancient LINE L2s were associated with tissue-shared regulatory regions and with those switching between promoter and enhancer signatures across species. Our analyses of the tissue-specificity and evolutionary stability among promoters and enhancers reveal how specific LINE families have helped shape the dynamic mammalian regulome.This work was published in  Genome Biology .


First posted as a preprint to  BioRxiv .

Data access

The raw and processed high-throughput sequencing data are available in ArrayExpress. The ChIP-seq datasets have accession number  E-MTAB-7127 , and matched RNA-seq experiments  E-MTAB-8122. For the reannotation of genomes, additional RNA-seq dataset were produced and are also available under the accession number  E-MTAB-8118.

Scripts

The scripts used to run all the analyses are available as  Additional file 9: Data S1  associated with the  Genome Biology  manuscript. The script for projecting .bed file coordinates across species using Ensembl EPO whole genome alignments is included in the manuscript supplementary, but also available on  GitHub  as an example Ensembl Compara API script.

Final regulatory region calls per species

The results of postprocessing histone enrichment peaks to final regulatory region definitions across tissues is available  here . These files combine all per-tissue histone peaks calls within each species to define cross-tissue activity of active promoters, active enhancers and primed enhancers. For more details, see the Materials and methods section of the manuscript published in  Genome Biology .

File naming convention: 

{Species}_regRegions_allTissue_parsed.txt


The columns in the tab-delimited file correspond to:

Column number Description
1 Chromosome (Ensembl convention)
2 Start coordinate (Ensembl convention)
3 End coordinate (Ensembl convention)
4 Comma separated list of all regulatory region signatures at this location; multiple regulatory regions if different between tissues.
5 Comma separated list of all tissues with regulatory signatures
6 Unique IDs (see more details below)


Unique ID naming convention: 

{Species}_{Tissue}_{HistoneCombination}_{uniqueNumber}


For example: 

Mouse_Testis_H3K4me1-H3K27ac-H3K4me3_1


The unique IDs include the combination of histone peaks used to call regulatory regions within each species.
H3K27ac+H3K4me3 – active promoters, i.e. regions with both H3K27ac and H3K4me3
H3K27ac-H3K4me3+H3K4me1 – active enhancers, i.e. regions with H3K27ac and H3K4me1, but not H3K4me3
H3K4me1-H3K27ac-H3K4me3 – primed enhancers, i.e. region with only H3K4me1

All evolutionarily maintained regulatory regions

The results of finding all maintained regulatory regions – i.e. those regulatory regions in a species that align to a regulatory region in another species – are available  here . Briefly, these files were generated by taking all the final regulatory region calls per species (see above) and using Ensembl EPO alignments to check whether a regulatory region call in another species aligns with at least one base overlap. Every possible reciprocal pairwise comparison is represented by a seperate file. For the definition of evolutionarily maintained regulatory regions and more method details, see the Materials and methods section of the manuscript published in  Genome Biology .

File naming convention: 

{Species1}_regRegions_allTissue_mainRegs_to_{Species2}_active.txt

This file reports all regulatory regions from Species 1 that align to a regulatory region in Species 2, as well as the regulatory signature, tissue of activity and position, in both species.

The columns in the tab-delimited files correspond to:

Column number Description
1 Alignment of query region from Species 1 (see Column 4) to Species 2. Species 2 chromosome (Ensembl convention)
2 Alignment of query region from Species 1 (see Column 4) to Species 2. Species 2 start coordinate (Ensembl convention)
3 Alignment of query region from Species 1 (see Column 4) to Species 2. Species 2 end coordinate (Ensembl convention)
4 Species 1 query region used to align to Species 2. Syntax Chromsome:Start-End (Ensembl convention)
5 Comma separated list of regulatory signature(s) in Species 1
6 Comma separated list of all tissues with regulatory signatures in Species 1
7 Unique IDs in Species 1 (see more details above under Final regulatory region calls)
8 Species 2 regulatory region chromosome (Ensembl convention)
9 Species 2 regulatory region start coordinate (Ensembl convention)
10 Species 2 regulatory region end coordinate (Ensembl convention)
11 Comma separated list of regulatory signature(s) in Species 2
12 Comma separated list of all tissues with regulatory signatures in Species 2
13 Unique IDs in Species 2 (see more details above under Final regulatory region calls)

Final normalised RNA-seq data

The results of RNA-seq normalisation across tissues and filtering is available  here . Briefly, the cufflinks suite was used to normalise RNA-seq across Ensembl annotated genes and transcripts and lowly covered genes/transcripts were removed using an FPKM cutoff. For more details, see the Materials and methods section of the manuscript published in  Genome Biology .  File naming convention:  {Species}_{genes/isoforms}.fpkm_table_FPKMfilter
Genes files correspond to all Ensembl annotated genes.
Isoform files correspond to all Ensembl annotated transcripts.

Repeat masking of genomes

Repeats were masked with RepeatMasker and the RepBase database, for more information please see the methods section published in  Genome Biology . The gtf formatted results of repeat masking are available  here .