• PC Schwalie, MC Ward, CE Cain, AJ Faure, Y Gilad, DT Odom, P Flicek. Co-binding by YY1 identifies the transcriptionally active, highly conserved set of CTCF-bound regions in primate genomes. Genome Biol 2013;14(12):R148. doi:10.1186/gb-2013-14-12-r148
  [BibTeX] [Abstract]
  

  \textbf{BACKGROUND:} The genomic binding of CTCF is highly conserved across mammals, but the mechanisms that underlie its stability are poorly understood. One transcription factor known to functionally interact with CTCF in the context of X-chromosome inactivation is the ubiquitously expressed YY1. Because combinatorial transcription factor binding can contribute to the evolutionary stabilization of regulatory regions, we tested whether YY1 and CTCF co-binding could in part account for conservation of CTCF binding.
  \textbf{RESULTS:} Combined analysis of CTCF and YY1 binding in lymphoblastoid cell lines from seven primates, as well as in mouse and human livers, reveals extensive genome-wide co-localization specifically at evolutionarily stable CTCF-bound regions. CTCF-YY1 co-bound regions resemble regions bound by YY1 alone, as they enrich for co-bound transcription factors, RNA polymerase II and active histone marks. Although these highly conserved, transcriptionally active CTCF-YY1 co-bound regions are often promoter-proximal, gene-distal sites show similar molecular features.
  \textbf{CONCLUSIONS:} Our results reveal that these two ubiquitously expressed, multi-functional zinc-finger proteins collaborate in functionally active regions to stabilize one another’s genome-wide binding across primate evolution

  @Article{24380390,
  author = {Schwalie PC and Ward MC and Cain CE and Faure AJ and Gilad Y and Odom DT and Flicek P},
  title = {Co-binding by YY1 identifies the transcriptionally active, highly conserved set of CTCF-bound regions in primate genomes},
  journal = {Genome Biol},
  volume = {14},
  number = {12},
  pages = {R148},
  year = {2013},
  doi = {10.1186/gb-2013-14-12-r148},
  abstract = {\textbf{BACKGROUND:} The genomic binding of CTCF is highly conserved across mammals, but the mechanisms that underlie its stability are poorly understood. One transcription factor known to functionally interact with CTCF in the context of X-chromosome inactivation is the ubiquitously expressed YY1. Because combinatorial transcription factor binding can contribute to the evolutionary stabilization of regulatory regions, we tested whether YY1 and CTCF co-binding could in part account for conservation of CTCF binding.
  \textbf{RESULTS:} Combined analysis of CTCF and YY1 binding in lymphoblastoid cell lines from seven primates, as well as in mouse and human livers, reveals extensive genome-wide co-localization specifically at evolutionarily stable CTCF-bound regions. CTCF-YY1 co-bound regions resemble regions bound by YY1 alone, as they enrich for co-bound transcription factors, RNA polymerase II and active histone marks. Although these highly conserved, transcriptionally active CTCF-YY1 co-bound regions are often promoter-proximal, gene-distal sites show similar molecular features.
  \textbf{CONCLUSIONS:} Our results reveal that these two ubiquitously expressed, multi-functional zinc-finger proteins collaborate in functionally active regions to stabilize one another's genome-wide binding across primate evolution},}

Description

Abstract

The genomic binding of CTCF is highly conserved across mammals, but the mechanisms that underlie its stability are poorly understood. One transcription factor known to functionally interact with CTCF in the context of X-chromosome inactivation is the ubiquitously expressed YY1. Because combinatorial transcription factor binding can contribute to the evolutionary stabilization of regulatory regions, we tested whether YY1 and CTCF co-binding could in part account for conservation of CTCF binding.

Combined analysis of CTCF and YY1 binding in lymphoblastoid cell lines from seven primates, as well as in mouse and human livers, reveals extensive genome-wide co-localization specifically at evolutionarily stable CTCF-bound regions. CTCF-YY1 co-bound regions resemble regions bound by YY1 alone, as they enrich for co-bound transcription factors, RNA polymerase II and active histone marks. Although these highly conserved, transcriptionally active CTCF-YY1 co-bound regions are often promoter-proximal, gene-distal sites show similar molecular features.

Our results reveal that these two ubiquitously expressed, multi-functional zinc-finger proteins collaborate in functionally active regions to stabilize one another’s genome-wide binding across primate evolution.

Full details  published in Genome Biology .

Raw Data

The multi-species CTCF and YY1 ChIP-seq reads can be found in ArrayExpress with the accession number E-MTAB-1511.  ChIP-seq fastq data : E-MTAB-1511

The multi-species gene expression data can be found in Arrayexpress with the accession number E-MTAB-424.  RNA-seq fastq data: E-MTAB-424.

Peak calls

Peaks calls made using  CCAT  3.0 can be found in ArrayExpress with the accession number  E-MTAB-1511 , along with the raw data and are also provided below as gff files (chr, peakCaller, TF-species-tissue, start, end, CCATScore, +, ., chr:peakStartEnd).

CTCF peak calls in LCL and Liver
YY1 peak calls in LCL and Liver
CTCF peaks overlapping (or not) YY1 (GRCh37/hg19 and GRCm38/mm10)
CTCF peaks shared with all primates (GRCh37/hg19)
CTCF and YY1 peaks overlapping enriched repetitive elements (LTR13, LTR41, LTR50 and MLTs)

Human peak calls shared in other species

For each human peak call, the sharing status in other species is listed in both LCLs and liver tissue. Primate specific and mammalian EPO alignments were used in the analysis. Columns are labelled in each file.

Format of primate EPO alignment column: (0 not present, 1 shared); species order hsa, ptr, ggo, ppy, mml, pha, cja (soed); all zeros: region not in alignment.

Format of mammal EPO alignment column: (0 not present, 1 shared); species order hsa, mml, cja, mmu; all zeros: region not in alignment.

Human coordinates from GRCh37/hp19 assembly.

Shared peaks in LCL
Shared peaks in liver

• K Stefflova, D Thybert, MD Wilson, I Streeter, J Aleksic, P Karagianni, A Brazma, DJ Adams, I Talianidis, JC Marioni, P Flicek, DT Odom. Cooperativity and rapid evolution of cobound transcription factors in closely related mammals. Cell 2013;154(3):530–540. doi:10.1016/j.cell.2013.07.007
  [BibTeX] [Abstract]
  

  To mechanistically characterize the microevolutionary processes active in altering transcription factor (TF) binding among closely related mammals, we compared the genome-wide binding of three tissue-specific TFs that control liver gene expression in six rodents. Despite an overall fast turnover of TF binding locations between species, we identified thousands of TF regions of highly constrained TF binding intensity. Although individual mutations in bound sequence motifs can influence TF binding, most binding differences occur in the absence of nearby sequence variations. Instead, combinatorial binding was found to be significant for genetic and evolutionary stability; cobound TFs tend to disappear in concert and were sensitive to genetic knockout of partner TFs. The large, qualitative differences in genomic regions bound between closely related mammals, when contrasted with the smaller, quantitative TF binding differences among Drosophila species, illustrate how genome structure and population genetics together shape regulatory evolution

  @Article{23911320,
  author = {Stefflova K and Thybert D and Wilson MD and Streeter I and Aleksic J and Karagianni P and Brazma A and Adams DJ and Talianidis I and Marioni JC and Flicek P and Odom DT},
  title = {Cooperativity and rapid evolution of cobound transcription factors in closely related mammals},
  journal = {Cell},
  volume = {154},
  number = {3},
  pages = {530--540},
  year = {2013},
  doi = {10.1016/j.cell.2013.07.007},
  abstract = {To mechanistically characterize the microevolutionary processes active in altering transcription factor (TF) binding among closely related mammals, we compared the genome-wide binding of three tissue-specific TFs that control liver gene expression in six rodents. Despite an overall fast turnover of TF binding locations between species, we identified thousands of TF regions of highly constrained TF binding intensity. Although individual mutations in bound sequence motifs can influence TF binding, most binding differences occur in the absence of nearby sequence variations. Instead, combinatorial binding was found to be significant for genetic and evolutionary stability; cobound TFs tend to disappear in concert and were sensitive to genetic knockout of partner TFs. The large, qualitative differences in genomic regions bound between closely related mammals, when contrasted with the smaller, quantitative TF binding differences among Drosophila species, illustrate how genome structure and population genetics together shape regulatory evolution},}

Description

To mechanistically characterize the microevolutionary processes active in altering transcription factor (TF) binding among closely related mammals, we compared the genome-wide binding of three tissue-specific TFs that control liver gene expression in six rodents. Despite an overall fast turnover of TF binding locations between species, we identified thousands of TF regions of highly constrained TF binding intensity. Although individual mutations in bound sequence motifs can influence TF binding, most binding differences occur in the absence of nearby sequence variations. Instead, combinatorial binding was found to be significant for genetic and evolutionary stability; cobound TFs tend to disappear in concert and were sensitive to genetic knockout of partner TFs. The large, qualitative differences in genomic regions bound between closely related mammals, when contrasted with the smaller, quantitative TF binding differences among Drosophila species, illustrate how genome structure and population genetics together shape regulatory evolution.

Raw Data

The raw chip-seq data for CEBPA, HNF4A and FOXA1 from the 5 mouse species including the CEBPA and HNF4A KO mices can be found in ArrayExpress with accession number  E-MTAB-1414.
The raw chip-seq data for CEBPA, HNF4A and FOXA1 from rat can be found in ArrayExpress with the accession number  E-MTAB-1415.

SNV data

The raw genomic data used from Keane et al (2011) used to find the SNV can be found  here
The format used for the SNV data and described below is a simplified version of pileup format:
format for a SNV: [Chromosome], [position in reference genome], [reference nt], [nt in the non reference specie]
format for an indel [Chromosome], [position in reference genome], *,[indel]

C57BL/6J vs A_J SNVs
C57BL/6J vs CAST SNVs
C57BL/6J vs SPRET SNVs
C57BL/6J vs Caroli SNVs

Fasta Files of Mouse genomes

The fasta files contain the operational genome sequences for A/J,CAST,SPRET and Caroli. The genomes were constructed for the four mouse species by interverting the single nucleotide variants (SNVs) and integrating the small indels defined above. Because of the indel insertions, the ortholog regions between the mouse reference and those genomes may not have the same coordinate. The mapping tables below allows to do the coordinate mapping.

A_J operational genome
CAST operational genome
SPRET operational genome
Caroli operational genome

Mouse strain genome alignment representation

The files below represent pairwise genome comparison defined from the SNV files described above. See the description below for the meaning of each data field:
field 1 :  chromosome
field 2 :  id (unic for 1 given chromozom)
filed 3 :  Super genome begining fragment (super genome: representation of an genome containing all insertion between two species)
field 4 :  Super genome ending fragment
field 5 :  C57BL6J start of fragment
field 6 :  C57BL6J start of insertion
field 7 :  C57BL6J ending of fragnent
field 8 :  C57BL6J ending of insertion
field 9 :  strain start of fragment
field 10 :  strain start of insertion
field 11 :  strain end of fragment
field 12 :  strain genome end of insertion

if C57BL6 or a strain has an insertion, then the start of fragment = end of fragment and the start of insertion will be 1 and the end of insertion will = to the size of insertion

C57BL/6J vs A_J mapping table
C57BL/6J vs CAST mapping table
C57BL/6J vs SPRET mapping table
C57BL/6J vs Caroli mapping table

Complete peak calls

Peaks calls made using  SWEMBL
See the description below for the meaning of each data field:

chr:  chromosome
start_bl6 :  start of binding region wiht C57BL/6J coordinate system
end_bl6 :  start of binding region wiht C57BL/6J coordinate system
summit_bl6 :  summit of binding region wiht C57BL/6J coordinate system
intensity :  intensity in normalized read count
intensity_class :  intesnity classe
start_own :  start in specie coordinate system
end_own :  end in specie coordinate system
summit_own :  summit position in specie coordinate system

CEBPA peaks calls for 5 mouse species
HNF4A peaks calls for 5 mouse species
FOXA1 peaks calls for 5 mouse species

Peak calls with only mappable region between mouse and rat

See the description below for the meaning of each data field:  chr :  chromosome
start_bl6 :  start of binding region wiht C57BL/6J coordinate system
end_bl6 :  start of binding region wiht C57BL/6J coordinate system
summit_bl6 :  summit of binding region wiht C57BL/6J coordinate system
start_own :  start in specie coordinate system
end_own :  end in specie coordinate system
summit_own :  summit position in specie coordinate system

CEBPA peak calls for 5 mouse species + rat
HNF4A peak calls for 5 mouse species + rat
FOXA1 peak calls for 5 mouse species + rat

Data used for motif descovery

Those fasta files contains the sequences +/-12bp around the summit of the top 2000 peaks.

CEBPA fasta file
HNF4A fasta file
FOXA1 fasta file

Motifs matrices

The position frequency matrices (PFM) used for matrix search.

CEBPA PFM
HNF4A PFM
FOXA1 PFM

Ancestral state of the union of the five mouse strains TFBRs for CEBPA, HNF4A and FOXA1

See the description below for the meaning of each data field:
field 1 :  id of the locus with format chr|start|end
field 2 :  number of most parsimonious solutions
field 3 :  intensity class in C57BL6J
field 4 :  intensity class in Anc1 , the ancestor between C57BL6 and AJ
field 5 :  intensity class in Anc2 , the ancestor between Anc1 and CAST
field 6 :  intensity class in Anc3 , the ancestor between Anc2 and SPRET
field 7 :  intensity class in Anc4 , the ancestor between Anc3 and Caroli

CEBPA ancestral state
HNF4A ancestral state
FOXA1 ancestral state

• AJ Faure, D Schmidt, S Watt, PC Schwalie, MD Wilson, H Xu, RG Ramsay, DT Odom, P Flicek. Cohesin regulates tissue-specific expression by stabilizing highly occupied cis-regulatory modules. Genome Res 2012;22(11):2163–2175. doi:10.1101/gr.136507.111
  [BibTeX] [Abstract]
  

  The cohesin protein complex contributes to transcriptional regulation in a CTCF-independent manner by colocalizing with master regulators at tissue-specific loci. The regulation of transcription involves the concerted action of multiple transcription factors (TFs) and cohesin’s role in this context of combinatorial TF binding remains unexplored. To investigate cohesin-non-CTCF (CNC) binding events in vivo we mapped cohesin and CTCF, as well as a collection of tissue-specific and ubiquitous transcriptional regulators using ChIP-seq in primary mouse liver. We observe a positive correlation between the number of distinct TFs bound and the presence of CNC sites. In contrast to regions of the genome where cohesin and CTCF colocalize, CNC sites coincide with the binding of master regulators and enhancer-markers and are significantly associated with liver-specific expressed genes. We also show that cohesin presence partially explains the commonly observed discrepancy between TF motif score and ChIP signal. Evidence from these statistical analyses in wild-type cells, and comparisons to maps of TF binding in Rad21-cohesin haploinsufficient mouse liver, suggests that cohesin helps to stabilize large protein-DNA complexes. Finally, we observe that the presence of mirrored CTCF binding events at promoters and their nearby cohesin-bound enhancers is associated with elevated expression levels

  @Article{22780989,
  author = {Faure AJ and Schmidt D and Watt S and Schwalie PC and Wilson MD and Xu H and Ramsay RG and Odom DT and Flicek P},
  title = {Cohesin regulates tissue-specific expression by stabilizing highly occupied cis-regulatory modules},
  journal = {Genome Res},
  volume = {22},
  number = {11},
  pages = {2163--2175},
  year = {2012},
  doi = {10.1101/gr.136507.111},
  howpublished = {Advanced online publication: 10 July 2012},
  abstract = {The cohesin protein complex contributes to transcriptional regulation in a CTCF-independent manner by colocalizing with master regulators at tissue-specific loci. The regulation of transcription involves the concerted action of multiple transcription factors (TFs) and cohesin's role in this context of combinatorial TF binding remains unexplored. To investigate cohesin-non-CTCF (CNC) binding events in vivo we mapped cohesin and CTCF, as well as a collection of tissue-specific and ubiquitous transcriptional regulators using ChIP-seq in primary mouse liver. We observe a positive correlation between the number of distinct TFs bound and the presence of CNC sites. In contrast to regions of the genome where cohesin and CTCF colocalize, CNC sites coincide with the binding of master regulators and enhancer-markers and are significantly associated with liver-specific expressed genes. We also show that cohesin presence partially explains the commonly observed discrepancy between TF motif score and ChIP signal. Evidence from these statistical analyses in wild-type cells, and comparisons to maps of TF binding in Rad21-cohesin haploinsufficient mouse liver, suggests that cohesin helps to stabilize large protein-DNA complexes. Finally, we observe that the presence of mirrored CTCF binding events at promoters and their nearby cohesin-bound enhancers is associated with elevated expression levels},}

Description

As published in  Genome Research .

Raw Data

All ChIP-seq reads can be found in ArrayExpress with accession number E-MTAB-941

Peak calls

ChIP-seq peak calls made using SWEMBL are provided in both SWEMBL output and bed file format (chr, start, end) for the mouse embryonic stem cell and mouse liver cell  data.

Aligned reads

Aligned reads in bed file format for all data sets generated for this project can be downloaded  here . Bed files with and without identical duplicate reads are available.

• D Schmidt, PC Schwalie, MD Wilson, B Ballester, A Gonçalves, C Kutter, GD Brown, A Marshall, P Flicek, DT Odom. Waves of Retrotransposon Expansion Remodel Genome Organization and CTCF Binding in Multiple Mammalian Lineages. Cell 2012;148(1-2):335–348.
  [Bibtex]

  @Article{22244452,
  author = {Schmidt D and Schwalie PC and Wilson MD and Ballester B and Gonçalves A and Kutter C and Brown GD and Marshall A and Flicek P and Odom DT},
  title = {Waves of Retrotransposon Expansion Remodel Genome Organization and CTCF Binding in Multiple Mammalian Lineages},
  journal = {Cell},
  volume = {148},
  number = {1-2},
  pages = {335--348},
  year = {2012},
  doi = {10.1016/j.cell.2011.11.058},
  abstract = {CTCF-binding locations represent regulatory sequences that are highly constrained over the course of evolution. To gain insight into how these DNA elements are conserved and spread through the genome, we defined the full spectrum of CTCF-binding sites, including a 33/34-mer motif, and identified over five thousand highly conserved, robust, and tissue-independent CTCF-binding locations by comparing ChIP-seq data from six mammals. Our data indicate that activation of retroelements has produced species-specific expansions of CTCF binding in rodents, dogs, and opossum, which often functionally serve as chromatin and transcriptional insulators. We discovered fossilized repeat elements flanking deeply conserved CTCF-binding regions, indicating that similar retrotransposon expansions occurred hundreds of millions of years ago. Repeat-driven dispersal of CTCF binding is a fundamental, ancient, and still highly active mechanism of genome evolution in mammalian lineages. PAPERCLIP},}

As published in  Cell , a team of geneticists and computational biologists including Paul Flicek of EMBL-EBI and Duncan Odom of the WT Sanger Institute reveal how an ancient mechanism is involved in gene control and continues to drive genome evolution …  more .

Raw Data

The multi-species CTCF and SA1 ChIP-seq reads can be found in ArrayExpress with the accession number E-MTAB-437.  ChIP-seq fastq data : E-MTAB-437
The multi-species gene expression data can be found in ArrayExpress with the accession number E-MTAB-424.  RNA-seq fastq data: E-MTAB-424.

Peak calls

Peaks calls made using  SWEMBL  -R 0.005 can be found in ArrayExpress with the accession number  E-MTAB-437 , along with the raw data and are also provided below as gff files (chr, peakCaller, CTCF_species, start, end, SWEMBLScore, +, ., chr:peakSummit).

    CTCF

Peak calls for 6 species
5-way (placental) shared CTCF binding events (hg19)
Human-mouse-dog shared CTCF binding events (hg19)
5-way (placental) shared CTCF binding events (mm9)

   Cohesin (SA1)

Peak calls for opossum

Motif matrices

Below is the list of position weight matrices (PWM) for the canonical (M1) and the newly discovered (M2) CTCF motifs.
CTCF M1 and M2 PWMs
M1 PWM for each single species

    Peaks with M1 and M2 motifs

Coordinates for peaks that contain at least one M1 (NestedMica nmscan cutoff -15) and at least one M2 PWM match at a 21/22 half-site distance (included as 9 or 8 bp = distance between M1 end and M2 start).
Human peaks with both M1 and M2 PWM hits (stringent cutoff, hg19)
Human peaks with both M1 and M2 PWM hits (very lenient cutoff, hg19)
Mouse peaks with both M1 and M2 PWM hits (stringent cutoff, mm9)
Mouse peaks with both M1 and M2 PWM hits (very lenient cutoff, mm9)

The files contain coordinates of CTCF peaks and summit position, information about the M1 motif (motif column, contains chr and motif start as well as NestedMica nmscan score, with 0 the best match; if several motifs are present inside a peak they are separated by “;”) and information about M2 presence: distance from end of M1 to start of M2 (either 8 or 9), the score of M2 (0 is best match, the lower the values, the worse the match) and the strand.