Transcript Linkage analysis - Scheid Signalling Lab @ York University

Sequencing the Human
Genome
• In 1998, Celera Genomics announced
plans to sequence the human genome…
• …175,000 sequence reads per day,
operating 24 hours a day, 7 days a week
J. Craig Venter
Sequencing the Human
Genome
• Whole genome shotgun approach vs.
Clone by Clone approach
• By-passes the initial work of ordering
clones
• Celera performed about 32 million
sequence reads, each 500 – 1000 bp
Sequencing the Human
Genome
Sequencing the Human
Genome
• IHGSC published sequence reads every
24 hours to prevent patenting of DNA
• Celera had access to IHGSC data
• Debate over whether Celera could have
shotgun sequenced the genome without
IHGSC data
Sequencing the Human
Genome
• Both groups published results simultaneously
• Celera – Science
February 2001
• IHGSC – Nature
February 2001
Sequencing the Human
Genome
Nature 409, 818 - 820 (15 February 2001)
Sequencing the Human
Genome
• Controversy! Science published Celera’s
sequence without requiring deposition to
GenBank
• Celera provides full access, with a catch…
• Celera provided Science with a copy in escrow
Sequencing Your Human Genome
• For $500,000 you can have your DNA
sequenced
• Sequence 1000 individual human
genomes
• “Personalized” medicine
J. Craig Venter
Human Genome
• Legal considerations
– Should DNA, or genes, be patentable?
• In the past, USPTO considered genes as
man-made chemicals
– Copy DNA region, splice it together, and
propagate it in bacteria, etc
Human Genome
• Celera >6500 genes
• Human Genome Sciences >7000
• Incyte >50,000
• Only a fraction may be awarded by
USPTO, and only a fraction of these may
be useful in treating human disease
Human Genome
• 1994 U. of Rochester scientists isolate mRNA
for COX-2 and clone gene
• Suggest that compounds which inhibit COX-2
might provide pain relief from arthritis
• Submit patent application in 1995
Human Genome
• 1998 – Celebrex – inhibitor of
cyclooxygenase-2 (COX-2) introduced as
arthritis medication
• Developed by Pfizer/Searle
• Development began in early-90’s i.e. around
time of U. of Rochester discovery
Human Genome
• April 2000, U. of Rochester awarded patent
covering COX-2 gene and inhibition of the
peptide product thereof
• The same day, U. of Rochester files lawsuit
against Pfizer/Searle to block Celebrex sales
• Claims that Pfizer/Searle infringes on their
patent
• They want royalties from the sale of the
invention
Human Genome
• 2003 – U. of Rochester patent found invalid
• 2004 – Invalidation upheld by higher Court
• U. of Rochester patent did not provide
sufficient example of what the inhibitor would
be…i.e. claims too broad without a working
example
• How will “basic science” performed by
Universities be rewarded?
Human Genome
• Vioxx and Celebrex in news again this
year: increased risk of “cardiovascular
event” i.e. heart attacks
Human Genome
• Gene discovery
– Methods for finding genes
– Easy in prokaryotes
Human Genome
• Gene discovery
– Difficult in eukaryotes
Human Genome
• Gene discovery
– Average gene extends over 27 kb
– Average 8.8 introns
– Average 145 bp
• Extremes:
– Dystrophin gene 2.4 Mb
– Titin gene contains 178 introns, coding for a
80,780 bp mRNA
Human Genome
• Gene discovery
– One approach is to examine “transcriptome”
Human Genome
• Conservation of chromosome/gene
location between organisms
• Synteny
• Exons tend to be conserved between
species
Human Genome
• Human vs. Pufferfish genome
• Pufferfish genome about 1/7th the size of the
human genome with similar number of genes
Human Genome
• Predictive computer programs, e.g.
GENSCAN
• GENSCAN predicts the location of genes
based on splicing predictions, promoter
regions and other criteria
Human Genome
• Online databases have formed to curate
Human genome data
• Ensembl (www.ensemble.org)
Genetic Mapping of Mendelian
Characters
Identifying Disease-Causing
Gene Variations
• Linkage analysis and Positional Cloning
– Clone disease gene without knowing anything
except the approximate chromosomal location
Recombination
• Recombination during meiosis separates loci
– More often when they are farther apart
– Less often when they are close
• Recall discussion of the Genetic Map
– Loci on separate chromosomes segregate
independently
– Loci on the same chromosome segregate as a
function of recombination
Recombination
13-1
13_06.jpg
Linkage analysis
• Linkage analysis locates the disease gene
locus
– Linkage analysis requires
• Clear segregation patterns in families
• Informative markers close to the locus
– Utilize LOD analysis to verify linkage
– Calculate cM distance between Loci
Positional Cloning
• Widely used strategy in human genetics
for cloning disease genes
• No knowledge of the function of the gene
product is necessary
• Strong for finding single-gene disorders
Positional Cloning
• Linkage analysis with polymorphic
markers establishes location of disease
gene
• LOD score analysis, and other methods
are employed
• Once we know the approximate location…
– The heavy molecular biology begins
Positional Cloning
• Example - Huntington’s disease
– CAG…
– Autosomal dominant
– 100% penetrance
– Fatal
– Late onset means patients often have children
Finding the Huntington Gene –
1981-1983
• Family with Huntington's disease found in
Venezuela
• Originated from a single founder - female
• Provided:
– Traceable family pedigree
– Informative meiosis
– Problem was… only a few polymorphic markers
where known at the time
Finding the Huntington Gene
• Blood samples taken
• Check for disease symptoms
• Paternity verified
Finding the Huntington Gene
• By luck, one haplotype segregated very
closely with Huntington disease
• Marker was an RFLP called G8 (later
called D4S10)
Finding the Huntington Gene
Finding the Huntington Gene
• Locate the region to the tip of the short
arm of chromosome 4 by linkage with G8
(D4S10)
• Maximum LOD score occurred at about 4
cM distance, i.e. 4 in 100 meiosis
Finding the Huntington Gene
• Together this started an international effort
to generate YAC clones of the 4 Mb region
• More polymorphisms were found
Finding the Huntington Gene
• Next, find an unknown gene in an
uncharacterized chromosome location
• Locate CpG islands
• Cross-species comparisons
• Further haplotype analysis suggested a 500 Kb
region 3’ to D4S10
Finding the Huntington Gene
• Exon trapping was key
• Compare cloned exons between normal
and Huntington disease patients
Finding the Huntington Gene
Finding the Huntington Gene
• One exon, called IT15, contained an
expanded CAG repeat….
• Mapping to 4 cM – 1983
• Cloning of Huntington gene – 1993
Complex Disease and
Susceptibility
Single gene disorders
Gene
Mendelian Inheritance
High penetrance
Low environmental influence (but
sometimes significant)
Gene
LOD-based linkage analysis works great
Genetic heterogeneity
Disease
Low population incidence
Complex Disease and
Susceptibility
Gene
Gene
Gene
Gene
Environment
Disease A
Disease B
Disease C
Multifactorial disorders
Complex Disease and
Susceptibility
• Single gene disorders
–
–
–
–
–
–
Huntington’s
Fragile X
SCA1
DMD
Werner’s syndrome
Cystic fibrosis
• Multifactorial
–
–
–
–
–
–
–
Heart disease
Cancer
Stroke
Asthma
Diabetes
Alzheimer’s
Parkinson’s
Genetic Component in Complex
Disorders
• Relative risk
lr= frequency in relative of affected person
Population frequency
Genetic Component in Complex
Disorders
• Family Studies
Class of relative
Proportion of genes
shared
Examples
First degree
50%
Parent/child, siblings
Second degree
25%
Grandparent/grandchild, aunt/niece
Third degree
12.5%
Cousins
Genetic Component in Complex
Disorders
Congenital
Malformations
Cleft lip
Pyloric stenosis
General population
0.001
0.001
First degree relatives
X40 (0.04)
X10 (0.01)
Second degree relatives
X7
X5
Third degree relatives
X3
X1.5
• Problem of environmental impact
Genetic Component in Complex
Disorders
Disorder
Breast cancer
Type I diabetes
Type II diabetes
Multiple sclerosis
Peptic ulcer
Rheumatoid arthritis
Tuberculosis
Monozygotic
6.5
30
50
20
64
50
51
Dizygotic
5.5
5
30
6
44
8
22
Genetic Component in Complex
Disorders
Disorder
Alcoholism
Autism
Schizophrenia
Alzheimer’s
Dyslexia
Monozygotic
40
60
44
58
64
Dizygotic
20
7
16
26
40
Genetic Component in Complex
Disorders
• In polygenic diseases, risk (susceptibility)
alleles increase the phenotypic value
• Traits may appear continuously variable
• Traits may appear discontinuous
Genetic Component in Complex
Disorders
•
How to find susceptibility gene?
– Four main approaches
1.
2.
3.
4.
Candidate gene
Parametric linkage analysis
Non-parametric linkage analysis
Population association studies
Candidate gene
• Before searching the whole genome, think
about what genes may be involved
– Eg., Type I diabetes
– Some genes involved in cell-mediated
immunity are located on chromosome 6
(Human leukocyte antigen region)
– Linkage between Type I diabetes and HLA
was closely examined
• After a small genomic region is isolated,
determine best candidate gene
Parametric Linkage Analysis
• Standard LOD score analysis, as used for
single-gene disorders
Parametric Linkage Analysis
• Eg., breast cancer susceptibility genes
• Collect family history of >1500 breast cancer
patients
– Some family histories showed multiple cases
occurring at early ages – could be a Mendelian allele
segregating
– Best model suggested a dominant single-gene allele
with a population frequency of 0.0006 – this
suggested about 5% of total breast cancers
Parametric Linkage Analysis
• Eg., breast cancer susceptibility genes
• Collect family history of >1500 breast cancer
patients
– Now, look for families with multiple breast cancer
cases with early onset
– Genotype family members and look for linkage
– Linkage (significant LOD score) to breast cancer was
found to a marker on 17q21
Parametric Linkage Analysis
• Eg., breast cancer susceptibility genes
• Collect family history of >1500 breast cancer
patients
– The gene involved was cloned, like other single-gene
disorders
– Breast cancer (BRCA) 1 gene– tumor suppressor
gene involved in genomic stability
– LOH leads to high penetrance of breast cancer, as
well as ovarian cancer
Parametric Linkage Analysis
• Eg., breast cancer susceptibility genes
• Collect family history of >1500 breast cancer
patients
– However, examination of BRCA1 mutations outside of
affected families suggests lower penetrance
Parametric Linkage Analysis
• Other successes in finding Mendelian risk
factors in polygenic diseases
– HNPCC – non-polyposis colon cancer
• MSH1, MLH1, PMS1, PMS2
– FAP – familial polyposis colon cancer
• APC
– Premature heart disease - hypercholesterolemia
• Mutation of the LDL receptor
Parametric Linkage Analysis
• Familial hypercholesterolemia
– Autosomal dominant
Parametric Linkage Analysis
• Familial hypercholesterolemia
• 200 mg/dl - 350 mg/dl - dietary, common
• 400 mg/dl - 600 mg/dl - heterozygous,
uncommon
• >600 mg/dl - homozygous, rare
Parametric Linkage Analysis
• Familial hypercholesterolemia
• Autosomal dominant; allele frequency about
1:150
Parametric Linkage Analysis
• Spectacular misfires as well:
– Bi-polar disease (manic depression)
– Initial linkage to HRAS and INS on
chromosome 11
– LOD scores of 4.08 and 2.63
– Two individuals in extended family
misdiagnosed
– Lowered LOD score to 1.03 and 1.75
Non-parametric Linkage Analysis
• Genomic regions surrounding risk alleles will be
inherited from a common ancestor in affected
individuals to a greater frequency than by
chance – also called autozygosity mapping
• Search for commonly inherited regions by
polymorphic microsatellites, SNP’s, etc.
• High throughput analysis critical
Non-parametric Linkage Analysis
• Common to use Affected Sib-Pairs (ASP)
• Collect genotypic data for 100’s of ASP
• 300+ microsatellite markers genotyped for 10cM
coverage
• Look for significant IBD (>chance occurrence)
Non-parametric Linkage Analysis
• IBD: if parental alleles differ at locus, then
sibs that have both alleles in common are
identical by decent
• IBS: if parental alleles are not know, then
we can only say sibs are identical by state
Population association studies
• Association studies are carried out on
populations
• Look for alleles that segregate with the
disease in a whole population
– Direct causation
– Natural selection
– Linkage disequilibrium
Population association studies
• Linkage disequilibrium
• Combination of alleles at two closely
linked loci occur more often than expected
by chance from population frequencies
• Recombination reduces linkage
disequilibrium
Population association studies
• Linkage disequilibrium vs. Linkage
Mapping
– Mapping is performed on families with few
informative meiosis; LD is determined on
populations after many generations
– Mapping will show linkage over large
distances; LD is visible only over short
distances
Genetic Component in Complex
Disorders
• How to find susceptibility genes?
– Four main approaches
1.
2.
3.
4.
Candidate gene
Parametric linkage analysis
Non-parametric linkage analysis
Population association studies
Summary
• Family, adoption and twin studies provide evidence of
genetic component to complex disease
• Risk of disease is the combined effect of polygenes
influenced by environment, thus termed multifactorial
• Combined affect of many common alleles each providing
a small effect, or of a few uncommon alleles with large
effect
• Candidate gene, parametric and non-parametric linkage
analysis, and population association analysis are used to
find risk factors for multifactorial disease