Exam3Lec3GenomeWideAssociationStudies Flashcards
Medelian Inheritance
Dominant
99% of time this is due to a mutant polypeptide (Part of a big protein complex)
The mutant protein from one allele gets mixed with the protein from the wild type. This corrupts the protein assembly.
1 allele mutated for disease. Mutant polypeptide comes together to make protein and it corrupts it, 1 allele messes up the whole protein.
Medelian Inheritance
Recessive
Example is metabolic pathways; The other allele can make up for the mutated one
Both alleles mutated to cause disease.
For example you have one allele from mom, one allele from dad, one can still make an enzyme.
Medelian Inheritance
Penetrance of the gene/mutation
How a dominant mutation will present in a clinical setting, is could be mild, severe
How much an allele is having an impact on disease presentation
Mutation= DNA; Mutant=Protein
What is an example of a dominant disease?
Osteogenesis imperfecta
What is Osteogenesis imperfecta?
One mutant allele is making a protein that’s part of a bigger complex. If the cells pick the mutant protein from the pool of proteins, the protein will become part of the complex and corrupt it
Collagen assembles from the CN terminus. If the mutant is @ the N terminus it will not have much of an impact (end of collagen)
IF the mutant AA is picked up @ the C terminus, it will ruin everything else (beginniing)
incorporated into larger protein
As we assmeble collagen we pick up mutated protein from dad and it corrupts collagen
What is an example of a recessive disease?
PKU
What is PKU?
Recessive examples often include METABOLIC PW w/ polypeptides working as enzymes. As long as one allele (From mom or dad) is a wild type, there is no disease presentation. UNLESS there is an unusual stress
The protein works INDEPENDENTLY, this differs then collagen that mixes together its polypeptide
involved in metabolic pathways.
Independently working (1 from dad, 1 from mom), if you mutate one of them there is no disease presentation, b/c the other one is working indepemndently and is able to completete the pathway.
Issues with penetrance
What role does modifier genes play?
Modifier genes play a role in the development of cancer. They can reduce or increase the chance of inheritance of the mutation of BRCA 1/2.
BRCA1/2 are a dominantly inherited mutation, BUT a reduced penetrance when inherited just one mutation.It is Not a 100% chance of cancer if you have either/just BRCA1 or BRCA 2 mutation
dominat mutation, but not penetrance, There is an increased chance of cancer, but you dont automatically get cancer b/c you need many mutations
What is BRCA1 responsible for?
formation of kinetochore and is mutated in breast cancer
What is BRCA 2 responsible for?
Recombination repair and is mutated in breast cancer.
Tracking down disease gene in pedigree
What is a haplotype?
a group of alleles in an organism that are inherited together from a single parent (Person Specific)
you keep haplotype block (the generic DNA block), and it does not go througn recom
What is located between haplotypes?
recombination hotspots (in between haplotypes) that protect long segments of DNA. It goes through a lot of recombination so it has a greater chance of switching its alleles when creating germ cells.
haplotype block, recomb. hotspot, haplotype block, recomb hotsopt
What role does VNTRs play with diseased genes?
VNTRS are markers in the genome, they they stay the same surrounding the disease gene. VNTRs that surround the disease gene will have a different # of repeats between the homologs. different # of repeats between homologs. With this we can identify between homologs which is which. We use PCR to find VNTRs and the more number of copis we have the less cyles needed to find disease gene
when doing medellian inheritance we are looking for VNTRs
How do we track the “d” disease through generations via VNTRs?
In this example, the father has a disease gene, and had specific vntrs surrounding that gene. The female does not have this. VNTRs surrounding the disease gene stays the same as it goes down generations, so we can track the disease. We dont know whats causing the disease, so we go and track it down, through medellian inherticance to find out what gene is causing the disease.
Dominant because only one allele and offspring have the disease
VNTRs detectable w/PCR at a specific VNTR @ a specific location
This method of checking for disease if we don’t know the exact cause of the disease
Solid fill = disease state
See same VNTRs for the disease state
Where is the disease NOT located and why?
It is not located at VNTR postion S129,354, or 79.
This is because the mother had a recombination event, this allows us to further locate the disease. We knew the disease was on this chromosome, but if recombination happens, we can narrow down the gene that is causing disease.
The highlighted section “4, 2, 3” are Homologs that can be swapped out Via recombination, but the individual Will still retain the disease ( this helps us narrow disease)
This method is efficient for inherited diseases within a family; when the disease is inherited in a Mendelian-manner
As we are tracking down through inheritance, we are using VNTRs, if there was a recomb event, we know that where that recomb event happened (4,2,3 ) the gene thats mutated could not be there bc VNTRs changed, so it is NOT a posible position for mutation
What do we need to make to see which gene is mutated around the VNTRs?
Exome
exome=exons
One we narrow down to few vntr’s we need to make an exome to locate where disease occured.
Explain exon capture: exome
Capture Bait” Oligonucleotides (OG) will be attached to something that can easily be recovered
OG attached to beads that will sediment very easily in a test tube with the oligonucleotides. Will WC BP with the sequence in our genome and we get ride of the non-exonic DNA (98%) then we do massive parallel sequencing with the exonic DNA
Take seq that we know,fragment them, and use captiure bait. OG is attached to bead of all the exons (with the seq we know), put complimentary seq and let them wcbp. We wash out all the extra stuff and we are left with just the exome, bc all we care abt are the genes that are actually going to become proteins. We line them up and use massive parallel sequencing to find the sequence. We cpmpare normal genome to the sequnce we get
Explain steps in identifying a disease gene
- Use VNTRs to narrow down where there is a SNP or single mutation (location)
- Once we narrow it down, we fragment it, use capture bait w/og that can wcbp with just the exons,wash away the rest.
- Use massive parallel sequencing to get seq we want and then compare to normal genome,.
could ko/ki genes in identifying step
True or false: Mendelian = Dichotomous
TRUE there is small a and big a, Two small aa, you are fine.
We use VNTRs for mendelian inheritance of a disease then use genome browser
How do we define “candidate region” if no pedigree; and if disease is non-Mendelian, i.e., “quantitative/non -inherited”? (Compared to Mendelian)
Look at Recombination
Quantitative trait; i.e. quantifying how much glucose is in the blood ( more of a progression)
Dichotomous trait: means either you have the disease or you don’t ( for ex you either have osteogenesi imperfecta or you don’t)
Are Recombination events evenly distributed along the chromosome?
NO There are section of the genome that stay together (haplotype block) and there are section with high recombination (Recombination hotspots)
There are Recombination Hotspots
The DNA in between= haplotype block
Haplotype Block and linkage disequilibrium
identical throughout all the population except for SNPs
Linkage disequilibrium = all the alleles are travelling together because genome doesn’t divide randomly (no recomb/ crossing over)
You look for haplotype blocks which stays together and see if there is a mutation in the block and if there was, we can figure out if it is the cause of disease
Genes (and SNPs) within a haplotype block have a much greater chance, statistically, of remaining linked; this is referred to as Linkage Disequilibrium
Humans have how many haplotype blocks thoughout their genome?
200
What is this picture an example of?
Haplotype block.
76% of people have the orange sequence, 18% have the green. If anything changes in either of these sequences it is SNP (snps=in the haplotype block there was a mutation)
Genes (and SNPs) within a haplotype block have a much greater chance, statistically of being what?
Remaining linked, this is linkage disequilibrium
chaces are you will not go through recomb in a haplotype block
What borders haplotype blocks?
Recombination hotspots
blocks are H blocks
What does this photo represent?
Another view of haplotype blocks and recombination hotspots: Recombination hotspots border sections of the genome. So in this photo, peaks are the RH and spaces b/w peaks are HB.
In the non-medelian case, how do we identify the candidate region?
We use the HB. We follow the alleles for the SNPs in a large population; and thereby follow the haplotype block.
We have the whole genome, we can locate different region of the genome that differ, look at its clinical significance with the disease, and determine if this SNP increases risk for disease
Non-Mendelian
versus Mendelian:
Quantitative traits (non mend) and multi-gene effects (versus dichotomous traits).
Multi gene effects means that multiple genes play a role in getting disease, question is how many genes contribute to the quantitative traits. You look at suscceotibitly to getting to disease, and once you hit that thresholdm you have a very high chance of getting disease.
Explain graph A,B,D in this picture
A: Medalian disease (aa: no disease, Aa: dominant disease)
B: aabb: no disease,
small and big letters: progression of disease (pre-diabetic)
AABB: high subsceptibiltiy to disease
D: As we add more genes that are relevant, it starts to impact the intensity of the population. You get a gradient of disease presentation making it qualitative.
Now that we have a gradient, how does inheritance play a role?
Higher chance of obtaining disease in family (1st degree relatves) due to inheritance, the threshold is less. The threshold is less bc you dont need all alleles to cause the disease b/c you inhertied those snps
Which diabetes has a largers risk for inheritance?
Type 1 diabetes
What are Somatic cell genetic changes, e.g., changes that could make MZ twins (identical) different from one another?
- Mutations in sisters due to errors in DNA replication
- Copy number changes due to DNA replication, i.e., strand slippage (Like Huntingtons)
- Copy number changes due to sister chromatid exchange
- Epigenetics, after a DNA alteration?
Which X Chromosome is picked
Environment still considered to be major factor
* Such as different exposure to viruses
* Environmental factors when see diseases develop in one but not both monozygotic twins
How we define a candidate region is no pedigree and if the disease is quantitative?
Lets say 80+ y/o and have type 2 diabetes: all have at least 1 upper case allele for the 3 genes
review, slide 22 TA
We look for imbalances compared to the controls. We already sequenced out the whole genome, we know where HB are, if we can find snp in HB and locate it, we can see where gene was and see if it relates to diabtes.
We find that 18/26 have upper-case A, meaning that 69% has uppercase A. We find a SNP allele that is on the same haplotype block as the upper-case A allele that contributes to 69% of the cases as we see in the 18/26 that have diseases. We know that gene,** if mutated**, is related to type 2. Multiple genes have the ability to contribute
STUDY: Genome-Wide Association Analysis indentifies Loci for Type 2 diabetes and Triglyceride levels.
DISCOVERY BASED=They didn’t have a hypothesis, they looked at the whole genome with their genome wide associated study and simply looked for a snp and found differenves in HB.
Statistical significance of people with Type II diabetes having the rs10811661 polymorphic allele (A “G” for example)
The higher up the more statitical signifiance. SO this means the higher you go, the more overepresentation of allele you have, the more clinically sig, so pt’s with type 2 have more A to G changes.
So this snp (rs10811661) was significant enough where alot of pt samples had that snp at specific HB.
Polymorphism is linked to a gene causing Type II diabetes.
OMIM=Link SNP to disease
RS= SNP = designation
What does statistical significance refer to?
To calculate and determine if a specific SNP is occurring more frequently than if everything were random .
Is it statitistically sig or is it just a regular snp?
What does “linked” mean, when a trait or gene is linked to a SNP?
SNP to the gene with Upper A in this example)
We want to define a haplotype and look at genes in that haplotype for genes that are relevant to type 2 DM
What does this gene do, how is it linked to a disease?
A brief, simple example of “statistical significance” in the association of a polymorphic nucleotide residue with a disease:
slide 27 review
People with the disease: 69% have a G SNP allele (SIG); 31% have either C, A, T SNP alleles, ie, the G SNP allele is in linkage disequilibrium with the A gene allele (upper case allele)
People without the disease: 25% have either C, T, G, or A. (no sig)
Both groups have nucleotide residue T 90% of the time. (no sig)
Compare the 69% with the G SNP to the 25% with either C,T,G,A.
Th 69% G is significant
IF the “nearest known gene” is really causing the disease, then what would we expect?
An obvious (new??) function that is connected to Type II diabetes development.
A mutation in the gene that affects the gene function in the patients with Type II diabetes.
Type II diabetes is diabetes where insulin is present but not functioning, that is, insulin cannot stimulate glucose uptake. If the nearest gene encoded the insulin receptor AND The insulin receptor had a mutation in the Type II diabetics, THEN The SNP analysis would have led researchers to what conclusion?
69% of ppl with type 2 have A flipped to a G. We find out where that SNP is and find nearest gene, and type 2 diabetes we assume it has something to do with receptor. Remember that just bc you have snp does not mean you will get it, it is quantitative, you have likelihood.
Type 1 diabetes is associates with a mutation on which chromosome?
6,. If you add a snp, it alters your HLA class2 gene, and Imune system attacks the pancreas that produces insulin.
Explain how SNPs are assoiated with prostate cancer
in region 2, that SNP is PCAT1= this gene produces long non-coding RNA that promotes cell proliferation
in region 3, that SNP is POUSFIB= encodes for a TA for stem cell pluripotency
in region 1 that SNP is CASCB=Long coding RNA that might regulate MYC (Oncogene)
Mutate any of these, causes cancer
name for a combination of alleles that are linked together on the same physical chromosome
Haplotype
In the absence of recombination, each haplotype is inherited as a ____, and may be used to track gene flow in populations.
unit
Identical sequences throughout all pop (Except SNPs) ~200 in humans
Haplotype
Mutation of allele in haplotype
SNP
How Dominant mutation will present clinically
Penetrance
Multiple SNPs that could possible cause disease
Quantitative trait loci (QTL)-
True or false: Dichotomous trait=Mendelian inheritance
TRUE
True or False: If a gene represents a disease variant (i.e., has a variant nucleotide), that variant will be within a haplotype block and thereby associated with particular SNP alleles
Thus, by finding a SNP allele (A, T, G, or C) that is over-represented in the disease group, we can find the gene with the disease variant, i.e., on the same haplotype block that has the SNP allele that is over represented in the disease group. We look to see what is the closest gene or SNP and how does it relate to the disease.
True
True or False Mendelianly inherited disease genes can be tracked with VNTRs and similar genome “markers”, until isolated by recombinations within families and sequenced for mutatins.
true
Type II diabetes is due to lack of response to insulin, thus leading to excess _________ in the blood. Type II diabetes represents numerous gene variations, together leading to _________ traits, such as different levels of glucose, all of which will represent Type II diabetes, due to a cutoff value for glucose levels.
glucose, quantitative,
multiple genes contribute to getting type 2 diabetes
Associations of specific polymorphisms with a disease or trait can be established by ___________ tests, ie, determinations of whether a certain variant is found associated with a certain trait are dependent on a statistical analysis of significance.
statistical
The availability of the entire human genome, and knowledge of SNP’s throughout the genome, allows a ___________ based approach to determining whether certain regions of the genome are associated with a disease.
discovery
The availability of the entire genome allows for simultaneous discoveries regarding associations of different regions of the _______________ with diseases.
genome
Snps also represent ____ in the human genome. Basicallt SNP haplotypes fall b/w _____ of recombination. However, this is an oversimplification for modern, sophisticated GWAS’s b/c postions of recombination hotspots can vary throughout the population.
haplotypes, hot spots
www.gwascentral.org provides a link to ____ results
GWAS
