Introduction to Regulatory Genomics

1
Introduction to Regulatory Genomics

• CS195-L 4/22/08

2
Overview

• DNA gt RNA gt Protein
• Exons/Introns
• Alternative Splicing
• Transcription
• Promoter, Transcription Factorsgt enhance binding
  to RNA Polymerase IIgt allows for initiation of
  transcription.

3
Transcription in eukaryotic organisms

• Cis-regulatory sequences
• Short (usually 5-10nt) DNA sequences associated
  with each gene
• Degenerate similar sequences confer equivalent
  binding site

4
Transcription in Eukaryotic Organisms

• Variable positions relative to gene proximal or
  distal
• Sequences often cluster together forming
  "cis-regulatory modules
• Modules act independently to direct transcription
  of gene

5
Transcription Factors

• DNA-binding proteins that bind cis-regulatory
  sequences and can enhance or repress
  transcription
• Bind cooperatively to adjacent sites
• TFTF interactions are relatively weak and
  nonspecificsmall changes (point mutations) can
  have large effects.
• Combinatorial control of transcription

6
(No Transcript)
7
The framework

• Animal body plans and their structure/function
  are the result of development processes in time
  and space
• Development is mediated by the regulation of
  thousands of genes in time and space.

8
The framework

• This regulation is largely done by the
  interaction of DNA-binding proteins called
  transcription factors with other TFs and gene
  promoters.
• Often, binding sites upstream of the gene
  promoter as also involved in these regulatory
  processes.

9
The framework

• The DNA sequence of these binding sites can
  change over evolutionary time due to mutations.
• Alterations in the animal body plan are caused by
  changes in the organization of these regulatory
  sequences.

10
Regulatory Information and the Genome

• The genome contains the information for all cell
  types in the body.
• In order for an organism to develop, different
  sets of genes need to be expressed or repressed
  at the right place and time.
• If this does not occur properly, the effects are
  usually lethal.

11
Regulatory Information and the Genome

• Often, what is regulated in development are the
  expression of transcription factors.
• These factors then influence the expression of
  other TFs as well as intercellular signals.

12
General Principles of Organization of Regulation
in Development

• 1) Signalling affects regulatory gene expression
  Cells often send chemical signals to other cells
  which help to pattern gene expression in space.
• This means that gene regulatory elements often
  contain sequence that is responsive to
  intracellular signal transduction pathways.

13
General Principles of Organization of Regulation
in Development

• Intracellular signal transduction pathways are
  the biochemical pathways by which the cell takes
  external signals and changes gene expression.
• 2) Development is controlled by networks of genes
  (often coding for TFs) that regulate other genes.

14
General Principles of Organization of Regulation
in Development

• Transcription factors often target hundreds of
  genes. Which genes are actually activated (or
  repressed) depends on the combination of TFs
  present at the promoter region of the gene.

15
Gene Regulatory Networks

• Davidson conceives of these gene regulatory
  networks in the following way each node is a
  gene, which takes multiple inputs.
• Depending on the specific combination of inputs,
  the gene provides multiple outputs to other genes
  in the network.

16
Regulatory Genes perform multiple roles in
development

• The number of regulatory genes is limited, and
  all animals use more or less the same number of
  DNA-binding motifs used by different
  transcription factors.
• Transcription factors are often required for
  different processes at different times in
  development, and they are often used for many
  unrelated purposes in the life cycle.

17
C-value paradox
18
C-value paradox

• Genome size does not scale with complexity of the
  organism
• Number of protein-coding genes does not scale
  with complexity of the organism
• Biologists assume this is due to more complex
  ways of regulating genes.
• Alternative splicing also different domains to be
  swapping in and out of proteins.

19
Overview of Regulatory Architecture

• The short size and degeneracy of regulatory DNA
  motifs means that they will occur at random in
  enormous number.
• Functional distribution of these motifs is highly
  non-random.
• Functional regulatory elements that have been
  isolated consist of relatively dense clusters of
  distinct sites recognized by diverse DNA binding
  proteins.

20
Overview of Regulatory Architecture

• Specific clusters of sites specify regulatory
  activity.
• Cis-regulatory modules produce unique regulatory
  outputs in time and space in the organism.
• In other words, these functional groupings are
  associated with certain cells at certain times in
  development.

21
Overview of Regulatory Architecture

• Davidson defines enhancers as cis-regulatory
  modules located many kB distant from the basal
  transcriptional apparatus
• AKA the BTA the promoter, the transcriptional
  start site, etc.
• Enhancers communicate with the BTA by DNA
  looping.

22
Overview of Regulatory Architecture

• Others might function as silencers by preventing
  proteins from binding to each other or to the
  BTA.
• Only by experimentally verifying the function of
  target sites within a cis-regulatory element do
  we understand what the genomic regulatory
  sequence means.

23
Gene Regulatory Networks

• At the periphery of developmental gene networks
  are the sets of protein-coding differentiation
  genes that define particular cell types.
• These genes do not have outputs affecting other
  developmental genes.
• Developmental gene networks progressively specify
  exclusionary fates for cell types.

24
Cell types and the Genome

• All the specificity for each cell in the organism
  is ultimately contained in the genome.
• Spatial expressionEve2 as example
• One of the most important regulatory objectives
  in development is the control of spatial gene
  expression.

25
Cell types and the Genome

• Cis-regulatory modules are sufficient to drive
  spatial expression, given the presence of the
  appropriate TF input.
• Ectopically incorporated cis-regulatory modules
  are sufficient to generate correct patterns of
  spatial gene expression

26
Regulation

• Methylation of DNA-inhibits transcription
• Methylation (transcriptional repression) and
  acetylation (transcriptional activation) of
  nucleosomal histones
• Polycomb proteins remodel chromatin so TFs cant
  bind to promoters
• microRNAs repress expression via
  post-transcriptional binding.

27
Cis-regulatory modules

• CRMs non random clusters of certain target sites
  that usually span a few hundred bases.
• Modules dictate where, when, and how genes are
  expressed in development
• CRMs can be repressive or activating
• Output can be considered the combinatorial
  product of multiple operations.

28
Cis-regulatory modules

• CRMs present their information as inputs.
• The DNA binding proteins must bind and these
  outputs must be effectively communicated
  elsewhere to incorporate changes, such as the
  basal transcription apparatus or molecules
  associated with it.

29
Computational Identification

• CRMs tend to be highly conserved functional
  significance inferred from less than expected DNA
  divergence.
• Putative binding sites can be detected upon
  examination of highly conserved sites.

30
More on CRMs

• Davidson calls the active TFs that drive
  different regulatory states by appearing and
  activating genes at different times and place
  drivers.
• These drivers do not contain the whole story.
• There are many more tightly and specifically
  bound sites than just the subset occupied by
  drivers.

31
CRM example cyIIIa

• Network of conditional logic interactions
  programmed into the DNA

32
(No Transcript)
33
The combinatorial cis-regulatory logic code

• Genomic regulatory code how do different
  combinations of binding sites regulate genes?
• Sorin is working to solve this problem in
  conjunction with Eric Davidson and Ryan Tarpine.

34
Steps to doing this computationally

• Identify target sites (and the factors that bind
  them)
• Interpret the functions mediated by these sites
  and factors
• Find rules for combining these functions to infer
  overall cis-regulatory output.

35
Steps to doing this computationally

• Need a sufficiently useful, discriminatory and
  general target site database.
• A given TF can activate or repress, depending on
  combinations of protein partners.
• Want to find functional combinations of target
  sites/proteins.

36
Logic Functions

• Certain sites within Modules can be activated in
  a cooperative fashion and associate with other
  modules
• Can develop logic functions for the proper
  expression of a gene.