Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or...
Transcript of Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or...
![Page 1: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/1.jpg)
Integrative Analysis of Human and Mouse
Regulomes
John A. Stamatoyannopoulos, M.D. Depts. of Genome Sciences & Medicine
University of Washington
![Page 2: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/2.jpg)
Living genome features
![Page 3: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/3.jpg)
Functional elements
-‐ promoters-‐ enhancers -‐ silencers -‐ insulators -‐ etc.
![Page 4: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/4.jpg)
Mapping human regulatory DNA
![Page 5: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/5.jpg)
Functional elements
-‐ promoters-‐ enhancers -‐ silencers -‐ insulators -‐ etc.
![Page 6: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/6.jpg)
Mapping regulatory DNA using nucleases (DNase I)
DNase I
Sequencing
![Page 7: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/7.jpg)
Mapping regulatory DNA DNase I hypersensitive sites precisely mark regulatory DNA
![Page 8: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/8.jpg)
Mapping regulatory DNA DNase I hypersensitive sites precisely mark regulatory DNA
DNaseI hypersensi5ve site (DHS)
Regulatory DNA
![Page 9: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/9.jpg)
Mapping regulatory DNA DNase I hypersensitive sites precisely mark regulatory DNA
DNaseI hypersensi5ve site (DHS)
![Page 10: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/10.jpg)
Mapping regulatory DNA DNase I hypersensitive sites precisely mark regulatory DNA
DNaseI hypersensi5ve site (DHS)
Transcription factor footprints
![Page 11: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/11.jpg)
DNaseI hypersensi5ve site
Promoters
Mapping regulatory DNA DNase I hypersensitive sites precisely mark regulatory DNA
(DHS)
Enhancers
~100,000 – 250,000 elements per cell type (0.5-1.5% of genome)
Thurman et al, Nature, 2012
![Page 12: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/12.jpg)
>400 cell/tissue types and developmental states studied to date >95% from primary cells and tissues
![Page 13: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/13.jpg)
Mapping the human regulatory genome c. 2015
The human genome encodes at least 4 million DNaseI hypersensitive sites à Virtually all (>>99%) are tissue/lineage or cell type-selective elements à >95% of these are distal non-promoter elements à ~50% of DHSs are ‘memory sites’ – persistent marks of prior cell states
![Page 14: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/14.jpg)
What other information is encoded in regulatory DNA patterns?
![Page 15: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/15.jpg)
DHS patterns in fully differentiated cells encode memories of prior cell fate decisions
Stergachis et al, Cell, 2013
![Page 16: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/16.jpg)
Extensive forward propagation of regulatory information during cell differentiation
Stergachis et al, Cell, 2013
![Page 17: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/17.jpg)
~1/2 of DHSs in definitive cells are ‘memory sites’ that encode information about prior cell states
Systematic persistence of DHSs during differentiation
Stergachis et al, Cell 2013
![Page 18: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/18.jpg)
Mapping the human regulatory genome c. 2015
The human genome encodes at least 4 million DNaseI hypersensitive sites à Virtually all (>>99%) are tissue/lineage or cell type-selective elements à >95% of these are distal non-promoter elements à ~50% of DHSs are ‘memory sites’ – persistent marks of prior cell states à Nearly 1 million elements can be linked with likely target genes by co-activation across cell types à Individual cell types have hundreds to thousands of DHSs that are completely unique for that cell type
The genome encodes at least 20 million regulatory factor recognition sites à Each cell type likely encodes ~2-5 million transcription factor footprints à The average cell type utilizes a recognition ‘lexicon’ of ~2-300 ‘words à We are closing in on a complete recognition lexicon for human TFs
![Page 19: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/19.jpg)
Mapping the human regulatory genome c. 2015
The human genome encodes at least 4 million DNaseI hypersensitive sites
We have little idea what most of these do
The genome encodes at least 20 million regulatory factor recognition sites à Each cell type likely encodes ~2-5 million transcription factor footprints à Each cell type utilizes a recognition ‘lexicon’ of ~2-300 ‘words à We are closing in on a complete recognition lexicon for human TFs
![Page 20: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/20.jpg)
Not just ‘enhancers’: Most regulatory regions likely encode novel and complex activities that will take some time to sort out
Mercer et al., Nature Genetics 2013
![Page 21: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/21.jpg)
Mapping the human regulatory genome c. 2015
The human genome encodes at least 4 million DNaseI hypersensitive sites
We have little idea what most of these do
The genome encodes at least 20 million regulatory factor recognition sites à Each cell type likely encodes ~2-5 million transcription factor footprints à Each cell type utilizes a recognition ‘lexicon’ of ~2-300 ‘words à We are closing in on a complete recognition lexicon for human TFs
![Page 22: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/22.jpg)
Mapping the human regulatory genome c. 2015
The human genome encodes at least 4 million DNaseI hypersensitive sites
We have little idea what most of these do
The genome encodes at least 20 million regulatory factor recognition sites
Every regulatory region is built differently, and every TF must do its job (and cooperate with other TFs)
in its local context.
![Page 23: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/23.jpg)
Origins
How did the regulatory genome arise?
![Page 24: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/24.jpg)
Part I: Mouse and human regulatory regions Evolutionary dynamics of regulatory
DNA regions
Part II: Transcription factors and networks
Conservation of trans vs. cis regulatory circuitry
![Page 25: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/25.jpg)
Creating comprehensive maps of mouse regulatory DNA marked by DNase I hypersensitive sites (DHSs)
– 44 cells/Kssues studied
– 1.3 million disKnct DHSs
– Avg. 150,000 per cell/ Kssue type
IntegraKve comparison with ~3 million DHSs from 230human cell/Kssue types
![Page 26: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/26.jpg)
Comparative analysis of mouse and human regulatory DNA
Align sequence to human genome through
pair-wise alignment
Overlap aligned segments with human
DHSs (any cell type)
![Page 27: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/27.jpg)
Comparative analysis of mouse and human regulatory DNA
Align sequence to human genome through
pair-wise alignment
Overlap aligned segments with human
DHSs (any cell type)
Species- specific DHS Shared DHS
![Page 28: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/28.jpg)
Comparative analysis of mouse and human regulatory DNA
Align sequence to human genome through
pair-wise alignment
Overlap aligned segments with human
DHSs (any cell type)
![Page 29: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/29.jpg)
Pervasive turnover of
regulatory DNA in placental mammals
![Page 30: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/30.jpg)
Pervasive turnover of cis-regulatory DNA during mammalian evolution
~75% of DHSs restricted to
placental mammals
![Page 31: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/31.jpg)
Pervasive turnover of cis-regulatory DNA during mammalian evolution
50% non-aligning DHSs restricted to
murids
![Page 32: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/32.jpg)
Pervasive turnover of cis-regulatory DNA during mammalian evolution
The vast majority of mouse and human regulatory DNA is Placental mammal-specific and has undergone rapid evolution
![Page 33: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/33.jpg)
Evolutionary mechanism #1:
Functional repurposing of regulatory DNA
![Page 34: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/34.jpg)
Brain
sc e
Extensive functional ‘repurposing’ of regulatory DNA
1.3 million mouse DHSs
21% same tissue
14% different tissue
Mouse DHS Human DHS
Muscle Cell type
Mu l
Brain
![Page 35: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/35.jpg)
Brain
Extensive functional ‘repurposing’ of regulatory DNA
1.3 million mouse DHSs
21% same tissue
14% different tissue
Mouse DHS Human DHS
Muscle Muscle Cell type
Brain
![Page 36: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/36.jpg)
Simple but pervasive sequence changes underlie tissue repurposing
Mechanism for functional repurposing:
TF binding site turnover
Conservation of TF binding site
![Page 37: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/37.jpg)
Simple but pervasive sequence changes underlie tissue repurposing
Mechanism for functional repurposing:
TF binding site turnover 1.3 million mouse DHSs
35% DHS in human (any cell type)
Conservation of Novel binding site, TF binding site novel function
![Page 38: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/38.jpg)
Simple but pervasive sequence changes underlie tissue repurposing
Mechanism for functional repurposing:
TF binding site turnover 1.3 million mouse
Conservation of Novel binding site, TF binding site novel function
Conserved TF binding sites are significantly enriched in DHSs with conserved activity
11% ≥1 conserved TF binding site
24% No conserved TF binding sites
DHSs
![Page 39: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/39.jpg)
Conservation of cis-regulatory architecture vs. individual elements
Tissue-specific DHS landscape
![Page 40: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/40.jpg)
Conservation of cis-regulatory architecture vs. individual
elements
Tissue-specific DHS landscape
– 21% of mouse DHS landscape is shared with a corresponding human tissue
![Page 41: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/41.jpg)
Conservation of cis-regulatory architecture vs. individual
elements
Tissue-specific DHS landscape
– 21% of mouse DHS landscape is shared with a corresponding human tissue
– 11% has a conserved TF binding site
![Page 42: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/42.jpg)
Conservation of cis-regulatory architecture vs. individual
elements
Tissue-specific DHS landscape
– 21% of mouse DHS landscape is shared with a corresponding human tissue
– 11% has a conserved TF binding site
Given divergent regulatory landscapes, what is maintaining functional conservation in mouse and human?
![Page 43: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/43.jpg)
Evolutionary mechanism #2:
Conservation of global cis-regulatory ‘content’
![Page 44: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/44.jpg)
Conservation of overall cis-regulatory content
![Page 45: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/45.jpg)
Conservation of overall cis-regulatory content
![Page 46: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/46.jpg)
Conservation of overall cis-regulatory content
![Page 47: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/47.jpg)
Conservation of overall cis-regulatory content
Despite poor conservation of individual binding sites, the overall proportion of regulatory DNA ‘real estate’ available
to each TF in each organism remains nearly constant
![Page 48: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/48.jpg)
Rigid conservation of global TF recognition landscapes Every TF, every cell type ! Different sequence targets, same occupancy fraction
Content of cis-regulatory Individual cis-regulatory elements compartment
Many cis-regulatory architectures TF activity patterns are highly with convergent phenotypic constrained (i.e., trans-
outcomes environment)
![Page 49: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/49.jpg)
Recap The regulatory DNA landscape has undergone wholesale rewiring during the mouse-human interval
Humans and mice share a core mammalian regulon encoding cell identity and lineage programs
Regulatory DNA landscape evolution involves
• Extensive repurposing of elements from one tissue context to another
• Continuous ‘re-evolution’ on the same ancestral DNA template
• Strict conservation of the proportion of regulatory DNA encoding binding sites for each transcription factor
![Page 50: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/50.jpg)
Part I: Mouse and human regulatory regions Evolutionary dynamics of regulatory
DNA regions
Part II: Transcription factors and networks
Conservation of trans vs. cis regulatory circuitry
![Page 51: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/51.jpg)
Footprinting the mouse genome
![Page 52: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/52.jpg)
![Page 53: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/53.jpg)
![Page 54: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/54.jpg)
![Page 55: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/55.jpg)
Conservation of TF recognition repertoires
![Page 56: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/56.jpg)
Deriving a mouse cis-regulatory lexicon
![Page 57: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/57.jpg)
Deriving a mouse cis-regulatory lexicon
![Page 58: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/58.jpg)
Deriving a mouse cis-regulatory lexicon
Human footprint-derived motifs from Neph, Vierstra et al. Nature 2012
![Page 59: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/59.jpg)
Mouse-specific motifs are largely selective for ES cells
Human footprint-derived motifs from Neph, Vierstra et al. Nature 2012
![Page 60: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/60.jpg)
Conservation of trans regulatory circuitry
![Page 61: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/61.jpg)
Building direct TF networks using TF footprints
Node: Transcription factor Edge: Regulatory interaction between
2 TFs
Neph, Stergachis et al. Cell 2012
![Page 62: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/62.jpg)
![Page 63: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/63.jpg)
Direct TF footprint-derived networks accurately recapitulate known TF network relationships
Kim et al., 2008
Naidu et al. 1995; Yun & Wold 1996; Ramachandran et al.
Neph, Stergachis et al. Cell 2012
2008
![Page 64: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/64.jpg)
TF-to-TF connections are cell-selective
![Page 65: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/65.jpg)
TF-to-TF connections are cell-selective
![Page 66: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/66.jpg)
Conservation of TF-to-TF connections
![Page 67: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/67.jpg)
Conservation of TF-to-TF connections
![Page 68: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/68.jpg)
Conservation of global TF network architecture
Neph, Stergachis et al. Cell 2012
![Page 69: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/69.jpg)
Conservation of global TF network architecture
![Page 70: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/70.jpg)
Conservation of global TF network architecture
![Page 71: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/71.jpg)
Conservation of fine network architecture
![Page 72: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/72.jpg)
Conservation of fine network architecture
![Page 73: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/73.jpg)
Stepping away from the genome: Where evolution is really acting
![Page 74: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/74.jpg)
Evolutionary engineering of a regulatory genome
![Page 75: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/75.jpg)
Evolutionary engineering of a regulatory genome
![Page 76: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/76.jpg)
Evolutionary engineering of a regulatory genome
![Page 77: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/77.jpg)
Evolutionary engineering of a regulatory genome
![Page 78: Integrative Analysis of Human and Mouse Regulomes · Virtually all (>>99%) are tissue/lineage or cell type-selective elements >95% of these are distal non-promoter elements ~50% of](https://reader036.fdocuments.us/reader036/viewer/2022081615/5fd7441c48dd262deb059723/html5/thumbnails/78.jpg)
Acknowledgements - Mouse
Key experiments/analyses Mouse DNaseI pipeline groupJeff Vierstra, Andrew Stergachis, Shane Neph Miaohua Zhang, Rachael Byron, Peter Sabo,Eric Rynes, MaMhew Maurano, Eric Haugen, Theresa Cantwell, Kristen LeeRichard Sandstrom , Richard Humbert, Shinny Vong, Vaughan Roach, Erica GisteBob Thurman, Alex Reynolds, Sandra Stehling-‐Sun Benjamin Vernot, Wenqing Fu
CollaboratorsMark Groudine (FHCRC)M.A. Bender (FHCRC)Piper TreuKng (UW ComparaKve Med)Elhanan Borenstein (UW Genome Sciences) Joshua Akey (UW Genome Sciences) Shelly Heimfeld (FHCRC)Mouse ENCODE ConsorKum
Funding: NHGRI (MouseENCODE)