Decoding the epigenome to navigate human health

EPINOMICS

Decoding the epigenome to navigate

human health

Paul Giresi

EPA Workshop 9/2/2015

EPINOMICS 2

Epigenome is the instructions for using genome hardware

Hardware Software

Genome Epigenome

EPINOMICS 3

Instructions encoded within non-coding sequence

Genes

98% 2%

Approximately 40% of genome encodes for the

regulation of gene expression

EPINOMICS 4

Access to regulatory information controlled by epigenome

EPINOMICS 5

Techniques for measuring epigenomic features

Output Direct measures

RNA DNA

Methylation

Protein-DNA

Mapping

Accessibility

EPINOMICS 6

Landscape of technologies for measuring epigenome

TECHNOLOGY LANDSCAPE

Co

st

Sequencing

Microarray MeDIP

Bisulfite

conversionChIP

DNase

hypersenitivity

?

RNADNA

Methylation

Protein

Mapping

Open or active

regions

Scop

e

Static ~1% of genome Dynamic

Output of

epigenome

Assay only single

epigenetic factor

All bound

TFs

EPINOMICS 7

Rapid and efficient technology for profiling the epigenome

ATAC-seq

EPINOMICS 8

✔

✔

✔


ChIP-seq DNase-seq ATAC-seq

Input

requirement107 cells 107 cells 100-50,000 cells

Sample prep

time

2 days 4 days 2 hours

(30 min hands-on)

Outputs

• Measures single

factor only

• High skills required

to reproduce

• All proteins bound to

genome in 1 reaction

• Limited factors

can be probed

• Higher-order

chromatin compaction

EPINOMICS 9

Types of epigenomic features measured using ATAC-seq

DNA-binding proteins

Chromatin compaction

Active regulatory elements

EPINOMICS 10

Identification of active regulatory elements from limited number of cells

Closed Active

EPINOMICS 11

Compatible with fresh and archived blood samples

EPINOMICS 12

Landscape of hematopoietic development

HSC

MPP

CMP

GMP

LMP

MEP

CD8

B cells

Monos

CD4NK

EPINOMICS 13

Active regulatory elements are fingerprint of cell identity

TET2 TET2 mRNA

HSC

CMP

GMP

MEP

CD8+ T cells

NK cells

100 kb0 5000

EPINOMICS 14

Rapid and sensitive detection of epigenomic modulationC

D4

+ T

ce

ll a

cti

va

tio

n

PM

A/io

no

myc

in

0 hr

1 hr

2 hr

4 hr

EPINOMICS 15





Implications

Dynamic readout of cell

state and functioning

EPINOMICS 16

Binding locations of hundreds of factors read out at once

ATAC-seq

Ge

no

me

-wid

e C

TC

F m

oti

fs

Bound

Unbound

EPINOMICS 17

Detection of therapeutic modulation of epigenomic factors

Resting

Distance from binding site (bp)

Bin

din

g s

co

re

NFAT

CD4+ T cell activation

EPINOMICS 18

Detection of therapeutic modulation of epigenomic factors

Resting

Activated

Distance from binding site (bp)

Bin

din

g s

co

re

NFAT

CD4+ T cell activation

EPINOMICS 19




Dynamic

readout of all

protein-DNA

interactions


Implications



EPINOMICS 20

Distribution of fragment lengths driven by chromatin

Paired-end sequencing

EPINOMICS 21

Distribution of fragment lengths driven by chromatin

ATAC DNA

EPINOMICS 22

Single basepair resolution nucleosome positioning

Detection of nucleosome

organization at transcription start

sites

Detection of nucleosome

positioning with loss of chromatin

remodeling complex

EPINOMICS 23

Determination of functional epigenomic states

EPINOMICS 24

Real time monitoring of epigenomic changes in patients

Week 0

Patient treatment

time course

Pati

en

t 20

Week 1

Week 2

Week 3 *Vo

rin

osta

t

* Treatment stopped

50 Fragment length (bp) 600

EPINOMICS 25



Implications

Determination of

functional epigenetic

states


Dynamic

readout of all

protein-DNA

interactions


Implications



Mapping epigenomic landscape to create purposeful navigation

EPINOMICS 26

Reference Maps

• Genome variants (1000 Genomes)

• Many regulators (ENCODE)

• Many many regulatory elements

Navigation

• Fast: clinical time scale

• Sensitive: clinical samples

• Actionable: clinical decision criteria

Measuring impact of environmental factors using plant epigenomics

EPINOMICS 27

• Continuously sample soil, air and water

• Material relative easy and inexpensive to obtain

• Able to sample from same plant over time

• Can perform follow-up experiments in controlled lab environment

Acknowledgements

EPINOMICS 28

Team

Fergus Chan - Co-founder

Tracy Nance, PhD - Bioinformatics

Marie Brennan, MD/PhD - Physician

John Latham, PhD - Scientist

Matt Negulescu, Software engineer

Anupama Joshi, Software engineer

Advisers

Howard Chang, MD/PhD - Stanford

Will Greenleaf, PhD - Stanford

Mike Snyder, PhD - Stanford

Anshul Kundaje, PhD - Stanford

Robert Tibshirani, PhD - Stanford

Joseph Ecker, PhD - Salk

EPINOMICS 29

Decoding the epigenome to navigate human health

Documents

Transcript of Decoding the epigenome to navigate human health