Scribd
Upload
42 views

Gene Ontology and Functional Enrichment: Genome 559: Introduction To Statistical and Computational Genomics

This document discusses gene ontology and functional enrichment analysis. It introduces the Gene Ontology project, which provides a standardized vocabulary and annotation of gene functions. Functional enrichment analysis identifies which biological processes or molecular functions are overrepresented among a set of genes of interest compared to what is expected by chance. The document outlines how enrichment analysis was previously done in an ad hoc manner and how it is now done systematically using the Gene Ontology, gene annotations, and statistical tests like the modified Fisher's exact test. This allows identifying cellular functions that may be important for a given condition based on differentially expressed genes.

Uploaded by

alicia1990
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
42 views

Gene Ontology and Functional Enrichment: Genome 559: Introduction To Statistical and Computational Genomics

This document discusses gene ontology and functional enrichment analysis. It introduces the Gene Ontology project, which provides a standardized vocabulary and annotation of gene functions. Functional enrichment analysis identifies which biological processes or molecular functions are overrepresented among a set of genes of interest compared to what is expected by chance. The document outlines how enrichment analysis was previously done in an ad hoc manner and how it is now done systematically using the Gene Ontology, gene annotations, and statistical tests like the modified Fisher's exact test. This allows identifying cellular functions that may be important for a given condition based on differentially expressed genes.

Uploaded by

alicia1990
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 30

Gene Ontology and

Functional Enrichment
Genome 559: Introduction to Statistical and
Computational Genomics
Elhanan Borenstein

A quick review
 The parsimony principle:
 Find the tree that requires the
fewest evolutionary changes!

 A fundamentally different method:


 Search rather than reconstruct

 Parsimony algorithm
1. Construct all possible trees
2. For each site in the alignment and for each tree count the
minimal number of changes required
3. Add sites to obtain the total number of changes required
for each tree
4. Pick the tree with the lowest score

A quick review cont


 Small vs. large parsimony
 Fitchs algorithm:
1. Bottom-up phase: Determine the set of possible states
2. Top-down phase: Pick a state for each internal node

 Searching the tree space:


 Exhaustive search, branch and bound
 Hill climbing with Nearest-Neighbor Interchange

 Branch confidence and bootstrap support

From sequence to function


Which molecular processes/functions are
involved in a certain phenotype - disease,
response, development, etc.
(what is the cell doing vs. what it could possibly do)

Gene expression profiling

Gene expression profiling


 Measuring gene expression:
 (Northern blots and RT-qPCR)
 Microarray
 RNA-Seq

 Experimental conditions:






Disease vs. control


Across tissues
Across time
Across environments
Many more

Different techniques, same structure

genes

conditions

Back in the good old days


1. Find the set of differentially expressed genes.
2. Survey the literature to obtain insights about the
functions that differentially expressed genes are
involved in.
3. Group together genes with similar functions.
4. Identify functional categories with many differentially
expressed genes.

Conclude that these functions are important in


disease/condition under study

The good old days were not so good!


Time-consuming
Not systematic
Extremely subjective
No statistical validation

What do we need?
 A shared functional vocabulary
 Systematic linkage between genes and functions
 A way to identify genes relevant to the condition under
study
 Statistical analysis
(combining all of the above to identify cellular
functions that contributed to the disease or
condition under study)

 A way to identify related genes

What do we need?
Gene Ontology

 A shared functional vocabulary

Annotation

 Systematic linkage between genes and functions


 A way to identify genes relevant to the condition under
study
Fold change,
Ranking, ANOVA

 Statistical analysis

Enrichment
analysis, GSEA

(combining all of the above to identify cellular


functions that contributed to the disease or
condition under study)
Clustering,
classification

 A way to identify related genes

The Gene Ontology (GO) Project


 A major bioinformatics initiative with the aim of
standardizing the representation of gene and gene
product attributes across species and databases.
 Three goals:
1. Maintain and further develop its controlled vocabulary of
gene and gene product attributes
2. Annotate genes and gene products, and assimilate and
disseminate annotation data
3. Provide tools to facilitate access to all aspects of the data
provided by the Gene Ontology project

GO terms
 The Gene Ontology (GO) is a controlled vocabulary,
a set of standard terms (words and phrases) used for
indexing and retrieving information.

Ontology structure
 GO also defines the relationships between
the terms, making it a structured vocabulary.
 GO is structured as a directed acyclic graph,
and each term has defined relationships to
one or more other terms.

GO domains
 Three ontology domains:
1. Molecular function: basic activity or task
e.g. catalytic activity, calcium ion binding
2. Biological process: broad objective or goal
e.g. signal transduction, immune response
3. Cellular component: location or complex
e.g. nucleus, mitochondrion

 Genes can have multiple annotations:


For example, the gene product cytochrome c can be described by the molecular
function term oxidoreductase activity, the biological process termsoxidative
phosphorylation and induction of cell death, and the cellular component
terms mitochondrial matrix and mitochondrial inner membrane.

Go domains
Biological process

Molecular function

Cellular component

Ontology and annotation databases

eggNOG
Clusters of Orthologous
Groups (COG)

The nice thing about standards is that


there are so many to choose from
Andrew S. Tanenbaum

What do we need?
 A shared functional vocabulary
 Systematic linkage between genes and functions
 A way to identify genes relevant to the condition under
study
GO annotation

 Statistical analysis
(combining all of the above to identify cellular
functions that contributed to the disease or
condition under study)

 A way to identify related genes

Picking relevant genes


 In most cases, we will consider differential expression
as a marker:
 Fold change cutoff (e.g., > two fold change)
 Fold change rank (e.g., top 10%)
 Significant differential expression (e.g., ANOVA)
(dont forget to correct for multiple testing, e.g., Bonferroni or FDR)

Gene study set

Enrichment analysis
Functional
category

Signaling category contains 27.6% of all genes


in the study set - by far the largest category.
Reasonable to conclude that signaling may be
important in the condition under study

# of genes in
the study set

Signaling

82

27.6

Metabolism

40

13.5

Others

31

10.4

Trans factors

28

9.4

Transporters

26

8.8

Proteases

20

6.7

Protein synthesis

19

6.4

Adhesion

16

5.4

Oxidation

13

4.4

Cell structure

10

3.4

Secretion

2.0

Detoxification

2.0

Enrichment analysis the wrong way


Functional
category

Signaling category contains 27.6% of all genes


in the study set - by far the largest category.
Reasonable to conclude that signaling may be
important in the condition under study

# of genes in
the study set

Signaling

82

27.6

Metabolism

40

13.5

Others

31

10.4

Trans factors

28

9.4

Transporters

26

8.8

Proteases

20

6.7

Protein synthesis

19

6.4

Adhesion

16

5.4

Oxidation

13

4.4

Cell structure

10

3.4

Secretion

2.0

Detoxification

2.0

Enrichment analysis the wrong way


 What if ~27% of the genes on the array are involved
in signaling?
 The number of signaling genes in the set is what expected by chance.
 We need to consider not only the number of genes in the set for each
category, but also the total number on the array.
Functional
category

 We want to know which category


is over-represented (occurs more
times than expected by chance).

# of genes in
the study set

% on
array

Signaling

82

27.6%

26%

Metabolism

40

13.5%

15%

Others

31

10.4%

11%

Trans factors

28

9.4%

10%

Transporters

26

8.8%

2%

Proteases

20

6.7%

7%

Protein synthesis

19

6.4%

7%

Adhesion

16

5.4%

6%

Oxidation

13

4.4%

4%

Cell structure

10

3.4%

8%

Secretion

2.0%

2%

Detoxification

2.0%

2%

Enrichment analysis the right way


 A statistical test, based on a null model
Assume the study set has nothing to do with the specific
function at hand and was selected randomly, would we be
surprised to see a certain number of genes annotated with this
function?
The urn version: You pick a set of 20 balls from an urn that
contains 250 black and white balls. How surprised will you be to
find that 16 of the balls you picked are white?

Modified Fisher's Exact Test


 Let m denote the total number of genes in the array
and n the number of genes in the study set.
 Let mt denote the total number of genes annotated
with function t and nt the number of genes in the study
set annotated with this function.

Modified Fisher's Exact Test


 Let S be a set of size n, sampled randomly without
replacement from the entire population of m genes,
and let t the number of genes in S annotated with t.
 The probability of observing exactly k genes in S
annotated with t is:
hypergeometric
distribution:

Modified Fisher's Exact Test


 We are interested in knowing the probability of seeing
nt or more annotated genes!
 We can simply sum over all possibilities:

 This is equivalent to a one-sided Fisher exact test

So what do we have so far?


 A shared functional vocabulary
 Systematic linkage between genes and functions
 A way to identify genes relevant to the condition under
study
 Statistical analysis
(combining all of the above to identify cellular
functions that contributed to the disease or
condition under study)

 A way to identify related genes

Still far from being perfect!


 A shared functional vocabulary
 Systematic linkage between genes and functions
Considers only a few genes

Arbitrary!

 A way to identify genes relevant to the condition under


study
Ignores links between
GO categories

Limited hypotheses

 Statistical analysis
(combining all of the above to identify cellular
functions that contributed to the disease or
condition under study)

 A way to identify related genes

Simplistic null model!

You might also like