Molecular Genetics

SlideSet8-12.pdf

Home >Biology homework help >Molecular Genetics

Back to the basics: Pedigrees and inheritance patterns

Autosomal dominant • affects either sex • transmitted by either sex • 50% chance of transmission * (*) for affected x unaffected with affected being heterozygous

Autosomal recessive • affects either sex • affected people often have unaffected

parents (asymptomatic carriers) • 25% chance of being affected

Achondroplasia (dwarfism) Type 1 Waardenburg syndrome (including deafness) Huntington’s disease (adult onset neurological deterioration)

only need one chromosome copy to have disease

Equally

very rare to mate with someone with same autosomal dominant disease

Unborn baby

= related by blood

Not sex-linked

X-linked recessive • affects mainly males • mother unaffected carrier • no male to male transmission • 50% transmission in males

X-linked dominant • affects either sex (females>males) • males transmit to daughters only (100%) • females transmit to 50% of offspring • females more mildly or variably affected

Y-linked • affects only males • all affected have affected father (no known human character besides maleness itself)

Red-green color blindness Hemophilia Duchenne muscular dystrophy

since its X linked it cant be transmitted through a Y chromosome

due to X chromosome inactivation

more likely to be lethal in males, so that's why it is seen higher rate in females

no known disease that looks like this!

:-. -

Locus heterogeneity: same clinical phenotype can result from mutations at several independent loci (example: congenital deafness). This can lead to genetic complementation.

Allelic heterogeneity: mutations in one gene can lead to different clinical manifestations (example: Duchenne vs. Becker muscular dystrophy).

Duchenne MD = deletions of multiple exons leading to out-of-frame mutations Little to no functional DMD protein is produced.

Becker MD = point mutations or deletions that restore frame Some level of functional DMD protein is produced. Severity of symptoms varies according to protein levels and functionality

same clinical manifestation produced by different genes

ex) BRCA1/2 phenotype of mutations within these genes have very different phenotypes for severity of breast cancer more disfunction will cause worse diseaes

mutations in gene A or gene B can cause the same disease

Ex aaBB and AAbb can both have the disease if they mate, their children will be hers so none of them will have the disease most human traits are controlled by many genes not only one gene

both deaf affected people

none of the children are deaf

they are all carriers for mutant alleles not do not show phenotype

mutations in same gene can cause different diseases

Dosage effect genes... proteins are usually not a monomer, they are usually multiple subunits. If 3/6 subunits are mutated, it will not function if its a dominant mutation even though 3 of the subunits are functional.

less severe than Duchenne MD

Mutate many differentgenes and result in same disease

Different mutation in SAME gene

- - - -

r o

- -

C l

Further complications to basic Mendelian pedigrees

Common recessive condition leads to appearance of pseudo-dominant inheritance patterns

Cystic Fibrosis (1 / 2,000 individual is a carrier)

High frequency of recessive alleles?

Geographical distribution of endemic malaria and mutant hemoglobin alleles in Africa T.E. Wellems, R.M. Fairhurst Nat Genet, 37 (2005)

heterozygote advantage?

CFTR gene- transport of chloride ions in mucous, respitory problems. mutant alleles may be common since they provide advantage against respitory infections in cold areas like Europe

High transmission rate, affects males and females...O is mutant allele married to asymptomatic carrier. There are many carriers in the populations. frequency of recessive alleles in population leads to large prevalence of recesive disease

can appear to be a dominant condition, but actually just high frequency of recessive alleles in form of carriers in population

Single cell anemia common to Africa. Hb mutation is thought to be an advantage against malaria

all affected are double O (00)

⑥ 0 @ ⑥

• O O

↳

Incomplete penetrance

Polydactily: autosomal dominant, 70% penetrance; BRCA1: autosomal dominant, 75-85% lifetime risk

Age-related penetrance.

Huntington syndrome, a dominant form of dementia due to triplet repeat amplification.

Curve A: probability of showing the disease phenotype as a function of age; curve B probability that an unaffected individual with an affected parent carries the disease allele as a function of age.

carriers are unaffected since there is incomplete penetrance, 70% risk of getting it, 30% chance of not getting. Depends of expression of mutation, by gene expression.

ONLY FOUND IN AUTOSOMAL DOMINANT

Grandma is affected

Mom is not affected

But her child is affected??

curve A: past age 20, there is increased likelihood as age increases B: as you reach 30+, you probably don't have the disease

If you receive the mutant chromosome 70% chance that will show

(Mom was a carrier)

We all have the BRCA 1/2 gene its whether or not mutation is present

positive: 50% chance to transmit to children. can select non-mutated gene so can prevent transmit on onto children negative: Find out that this disease could kill you. Can destroy the family structure

Gene Test?

> 3bases repeated

Variable expressivity: only some of the symptoms are presented by patients

Waardenburg syndrome

New spontaneous (de novo) mutations

color part of shape if they have some of the traits.

family has no history, there may be de novo mutations that cause new disease

(dominant condition)

Inbreeding: an X-linked recessive condition shows apparent male to male transmission and an affected female

Inbreeding effects

Inbreeding: happens! (cultural, geographical, religious) concentrates rare recessive alleles in a family & results in high frequency of recessive diseases

Nonsyndromic Congenital Retinal NonAttachment (NCRNA) Ghiasvand et al. (2011) Nature Neuroscience. Affects 1% of Kurdish people living in a group of neighboring villages in North Khorasan, Iran.

heterozygosity decreases homozygosity increases

inbreeding increases homozygousity inbreeding concentrates rare recessive alleles

females are carriers .

concentration of mutation

Inc frequency of rare mutations

Inbreeding concentrates rare recessive alleles (this can be useful for mapping disease-associated genes)

Nature Genetics (2015)

• Icelandish population is recent (est. ~900 AD) • The isolation of Iceland means more consanguinity • Good health records • Best characterized population genetically (deCODE

Genetics)

Iceland- recently populated. found 7,000 LOF autosomal SNPS ( knockout) in thousands of genes. what happens when you knock out genes?

consanguinity- blood relation

Inbreeding concentrates rare recessive alleles (this can be useful for mapping disease-associated genes)

Nature (2017)

can provide insight into gene function by gene knockout in humans.

APOC3 prevents against heart disease. its a good mutation. can lower triglycerides in the blood if you have this mutant allele.

Founder effects also lead to reduced genetic diversity Nature Genetics (Sept. 2017)

The caste system in India has led to the existence of thousands of reproductively isolated populations with evidence of profound genetic bottlenecks.

founder effect- some local human poplations can be founded by a few individuals. The founders are the ancestors of a village. the founders result from a bottleneck. There is a high degree of consignuity

This study looked at S Asia, viewed as many small groups of populations. Could identify 81 unique groups within the population that descent from founder events more extreme than those in the past. High rates of recessive disease due to founder effects (not inbreeding). .

we know diseases in jewish pop, can observe for presence of disease, can do genetic testing etc

• The size of human families tend to be small and collecting large pedigrees is often impossible.

• Careful, well-preserved medical records are often not available

But there are exceptions…

Nature Genetics 38, 184 - 190 (2006) Published online: 22 January 2006; | doi:10.1038/ng1728 Spectrin mutations cause spinocerebellar ataxia type 5 Yoshio Ikeda1, 2, 10, Katherine A Dick1, 2, 10, Marcy R Weatherspoon1, 2, Dan Gincel5, Karen R Armbrust1, 2, Joline C Dalton1, 2, Giovanni Stevanin6, Alexandra Dürr6, Christine Zühlke7, Katrin Bürk8, H Brent Clark3, 4, Alexis Brice6, Jeffrey D Rothstein5, Lawrence J Schut9, John W Day2, 4 & Laura P W Ranum

Limitations of human genetics how do we know if a gene is genetic? look at inheritance patterns, look at twins

people move, die etc.

ex. intellectual disability requires very careful diagnosis etc.

ataxia- defects in walking. can be linked to seizures Concordance: 60% likely to share concordance vs 40% ex) congenital deafness: 30% genetic (made up of 40% syndrome, 60% unknown genetic causes). 40% is acquired, due to infections of ears, injury. 30% is unknown causes.

Twins: share 100% of genome Fraternal: Brothers and Sisters who happen to be twins: 50% on average but can range from 0-100% depending on meiosis outcome.

ex. twins are genetically identical. they should have the same disease frequency etc. concordance rate: rate of disease without impact from environment

Gene mapping

The gene hunter minimal tool kit:

• Genetically-determined trait (disease, physical attribute, …) with precise phenotypic description

• Families in whom the trait is segregating (large # of offspring, high # of generations, good medical records, inbreeding)

• Markers (high density, high polymorphism, easy/cost-effective to assay)

• Goal of the game: identify markers that are as closely genetically linked as possible to the unknown disease gene (i.e. F(rec) is as low as possible)

there are a lot of markers along the genome, multiple alleles of a marker

Genetic mapping using linkage analysis

An ancestral mutation (A) for a dominant disorder segregates through human pedigrees…

…and the chromosomes get broken down through meiotic recombination

…but sequences in the immediate vicinity of the disease mutation remain relatively constant because two closely linked sequences tend to recombine at very low frequencies

Goal of the game: identify markers that show tight genetic linkage (low or no recombination) with the (unknown) disease-causing mutation

unknown disease gene, marker A is super close to the gene

gives us info on the recombination fraction/ frequency = theta

identify the markers that are in the same haplotype with the mutant allele.

affected

To be able to correctly infer genetic linkage we need to:

- Understand what genetic markers are - Understand what the effect of DNA recombination is on the inheritance of

DNA segments across generations - Understand how patterns of recombination (crossover) operate

Genetic markers used in linkage analysis

Two key attributes of markers are density and polymorphism!

use Restriction enzymes to cut and map SNPS near RE sites

ton of alleles, highly variable can use in forensics

only 2 chromosomes- so only 2 different alleles per locus. but high density because they are everywhere

The maximal recombination fraction between two markers is 50%

Recombination Fraction is the unit for measuring distance between marker

RF is ideally 0 for linked trait and marker

R= recombinant

Total#of Never more than recombinantChr

501. RF Total # of gametes

Counting the number of chiasmatas during meiosis informs about the total number of genetic exchanges, i.e. the size of the genetic map

Each chromosome undergoes 1-2 obligatory crossover at meiosis

Old modern

SNP mapping in families can help identify recombination sites

Also, sharing of chromosome regions between individuals is most often due to inheritance from an ancestor (identity by descent, IBD)

Credit Dr. Graham Coop - UCD

Effect of crossing over

Genetic relatedness decreases every generation

1 2

3 4

5 6

7 8

9 10 11 12

13 14

15 16

17 18

19 20

21 22

Your genome in your Mum

1 2

3 4

5 6

7 8

9 10 11 12

13 14

15 16

17 18

19 20

21 22

Your genome in your Maternal

grandma

1 2

3 4

5 6

7 8

9 10 11 12

13 14

15 16

17 18

19 20

21 22

Your genome in your Maternal

granddad

1 2

3 4

5 6

7 8

9 10 11 12

13 14

15 16

17 18

19 20

21 22

Your genome in your Dad

1 2

3 4

5 6

7 8

9 10 11 12

13 14

15 16

17 18

19 20

21 22

Your genome in your Paternal

grandma

1 2

3 4

5 6

7 8

9 10 11 12

13 14

15 16

17 18

19 20

21 22

Your genome in your Paternal

granddad

Credit Dr. Graham Coop - UCD

inherited bits and pieaces of

Patent& grandmother 1grandfather 's

gene

sister brother

SNP-based mapping of shared blocks between sibs

white- no sharing due to random segregation... one copy from mom and other from dad so likely that sibilings got opposite ones dark blue-both chromosomes are shared (2)-Both got the same chromosomes from mom&dad Light blue.-shares 1 chromosome

Average about 50 % sharing

Your maternal Grandmother’s Contribution to you

Your maternal Grandmother’s Contribution to your cousin

Overlap of Grandmother’s Contribution to you & cousin

Genetic relatedness decreases non-linearly along family tree

Credit Dr. Graham Coop - UCD

Genetic relatedness decreases non-linearly along family tree (now with 3rd cousins)

● ● ● ● ●

● ●

5 10 15

0 10

00 20

00 30

00 40

00 50

Generations ago

N um

be r o

f a nc

es to

● ● ● ● ●

● ●

●

● ● ● ● ● ●

●

● ●

● ● ● ● ●

5 10 15

0. 0

0. 2

0. 4

0. 6

0. 8

1. 0

Generations ago

P ro

b. a

nc es

to r i

s ge

ne tic

a nc

es to

You have vast numbers of ancestors… but most of them are not genetic ancestors

Ancestors (2n)

Ancestors who contributed to your genome

Credit Dr. Graham Coop - UCD

prob Not genetically I related

, passed 11 .

l l

Genetic ancestors

All ancestors

Credit Dr. Graham Coop - UCD

way more ancestors than genetic ancestors

T j parent grandparent

Many Americans likely will have many identifiable 3rd and 4th cousins (10s -- 100s) in large genomic databases. Likely have vast numbers of >5th cousins but they won’t be easily identifiable.

Credit Dr. Graham Coop - UCD

Implications for genealogy hunters… Establishing genetic ancestry can be tricky...

The vast majority of Warren’s DNA — 95 percent — indicated European ancestors. But five genetic segments were identified, with 99 percent confidence, as being associated with Native American ancestry. The largest segment identified was on Chromosome 10.

“While the vast majority of the individual’s ancestry is European, the results strongly support the existence of an unadmixed Native American ancestor in the individual’s pedigree, likely in the range of 6-10 generations ago,” the report said.

https://www.washingtonpost.com/politics/2018/10/18/ just-about-everything-youve-read-warren-dna-test-is- wrong/

… and useful!

Golden state killer identified through matches with 3rd and 4th cousins in GED Match database.

What your 3rd and 4th cousins do may end up affecting you!!

Meiotic recombination is a complex multi-step process that requires a large number of proteins

DSB

Resection

Strand invasion

CO or NCO intermediate

Resolution

Baudat et al., (2014) Nat Reviews Genetics

displacement loop: blue strand pairing against the homolog trying to find exactly complementary region to the sequence

proteins polymerize on to the single strand & searchs the homolog for the complementary region of DNA

RAD 51= most Important protein,the one actually performing thehomology search

PMDR9 to open DNA

form DSB

cut back ends to make ssDNA

D loop formed by strand invasion

resolution

most are non crossovers produced by gene conversion

crossover, chiasmata

Binding of MostofDNA

PRDM 9Occurs found lat loops .

at the tip of loopsof 100ps DNA

recombination occurs at

broken the axis

exposesthe

singlestrand Rare

- Most common

Meiotic recombination is a highly regulated and orchestrated process

Baudat et al., (2014) Nat Reviews Genetics

• Meiosis initiates via the introduction of programmed DSBs • Many more DSBs are created than will become crossovers • Breaks repaired through NCO pathways might permit proper

alignment through homology sensing

To find the right alignment non-crossover can still be useful . So breaking chromosomein 100's of places, the chromosomes can interact with eachother,sensing eachotherand make sure alignment is correct.

MEI4- works with PMRD9 creates on avg 150 DSB per cell we only have 1-2 crossovers and 43 chromosomes should only have 40 breaks on avg

yH2AX- DNA damage response DMC1/RAD51 repair homologs DSB

SYP3- holds homologs together. chiasmata form

pairingsynapses

↳ gehrxomo

DNA Dama ! some

marker Double strand breaks .

Recombination hotspots are genetically determined by the PRDM9 protein binding to a 13-mer motif

Recombination hotspots vary between individuals, between sexes and between populations

so what this means people from west African ancestry have a different Variant of PRDM9 then people of European ancestry. and as a result the protein binds to slightly different different sites such that some sites are only active in people from west African and industry ancestry because they are bound to the west African version of PRDM9 and some sites are only active in the European ancestry because they are bound by the European ancestry variant of PRDM9. which really tells you that location over combination hotspot is genetically determined and controlled by this one protein called PRDM9.

PRDM9 directs recombination away from genesso genes are not broken

13 bp sequence which is a recombination hotspot

there are 2500 hotspots that are active in Africans but not Europeans, there are genetic variants ( alleles) in different people that determines where recombination will be

no Cpg islands since it is meiosis specific ( expressed only in germ cells undergoing meiosis)

more recombinations in Africans vs Eurasian ancestry on average

PRDM9 functions by opening up chromatin and allowing Double Strand DNA (DSB) break formation

Baudat et al., (2014) Nat Reviews Genetics Grey et al., (2018) PLoS Genetics

PR/SET: Histone Methyltransferase catalytic domain KRAB: Histone binding module (function unclear) C2H2 zinc finger: DNA binding module

(1) (2) (3) 1. Binding to site and

chromatin modification

2. Recruitment to chromosome axis

3. DSB formation by SPO11

has a ZF protein that binds 13 bp sequence in DNA in the form of chromatin

PR SET domain is a catalytic domain that acts as a methyl transferase to add methyl group to H3K4me3 Histone 3, lysine 4 so the chromatin becomes open. allows it to open dna to break and repair

opensup the ← additional 3 histone tail chromatin . methyl groups chemically modified(Inc accessibility)

Brings the binding

site

to the central

axis

steps for PRDMI towork

synaptonomal complex Doublestrand bleak

PRDM9 and its binding sites evolves quickly!

Detection of PRDM9-dependent sites in inter-specific strains of mice. (A) PRDM9Dom2 and PRDM9Cst differ both in the number of zinc finger domains and sequence-specific identity of individual domains. (B) Identification of PRDM9-specific chromatin binding sites over a representative 100-kb window of chromosome 1. Peaks can be unique (Boxed) or shared between the two strains of mice.

Dom. Cst.

Baker et al., (2014) Genome Research

Consensus binding sites to PRDM9 alleles from Mus domesticus (left) and Mus castaneus (right)

Carries variants of prdm9 So therefore the bin ding sites of protein will be slightly different. So the dam and meitok recombination location will be different between ppl of different ancestry

Each finger corresponds to one base -linear arrangement of fingers are different.

where DNA breaks IS NOT random, dictated by location of PRDM 9 Because PRDM9 = product of genes. The genetic recombination & hotspots ate genetically determined

amino acid codes .

Genetic recombination rates vary between individuals

These patterns tend to be inherited as a result of distinct PRDM9 alleles: genetic control of DNA recombination!

Different alleles of PRDM 9 slome are Not good at binding, so low recombination frequency -because of the way oocytes do meiosis (are vulnerable to aneuploidy ) A good way to buffer/balance aneuploidy is to have more crossovers. -people with more crossovers are able to have children later in life because their oocytes have chromosomes that have Not apart fallen

Much more susceptible for aneploidy and therefore much less likely to have children late.

some individuals have different recombiation rates.

high recombination is passed on

more crossovers- less likely to have NDJ in meiosis if there are more of them

what are the genes involved in controlling meiotic recombination?

each dot is a single person

aufoiiakotorrt

Genetic maps are non-random and sex-specific

Males tend to recombine next to telomeres. This may reflect the fact that the first events of chromosome synapsis occur near telomeres from yeast to vertebrates.

Females tend to recombine away from telomeres AND have higher density of crossovers. Could be due to evolutionary pressure to avoid non-disjunction during MI arrest.

We used genotypes from high-density single nucleotide polymorphism (SNP) markers of 2,315 individuals and their children from two Caucasian populations to characterize meiotic recombinations.

Chowdhury et al., 2009 PLoS Genetics

red- females blue- males. higher at telomeres both have no recombination at centromere

High density analysis of recombination rates

Females are more likely to recombine more than males. More within the chromosome (middle) instead of at tips like males.if crossover is at the tip, much more likely to dissolve over time.Who are now two or three crossovers They are less likely to fall apart

Reflects the evolutionary pressure on oocytes

peaks-_ recombination hotspots

Malestend to recombine closer to the telomeres .

heend of chromosome

High density analysis of Double Strand Breaks (DSBs)

PAR DSBs in males

Sex-specific pattern of DSB usage

Brick et al., (2018) Nature

female MAPOFDSBON

Both specific germ cells(sperm& oocytes )

oocytes (

sperm

male specific

Fine scale mapping of recombination events reveals “hotspots” and “deserts”

recombination is not constant across the entire genome. there are hotspots which are the spikes. deserts- not no recombination activity- genes in one unit always segregate together. they never get broken from each other. It is flanked by recombination hotspots

Segment of DNA that does not tend to be broken by recombination and is inherited as a block I >

NO Recombination :haplotype block likely inherited asablock

together

✓

Haplotypes and Linkage Disequilibrium

• Haplotype Block: segment of DNA that is flanked by recombination hotspots and therefore tend to be inherited as a block.

SNPs within a haplotype block segregate as one unit. This implies that a particular haplotype block can be “tagged” by one SNP – simplifies genotyping

• Linkage Disequilibrium: a mathematical measure of linkage between markers. Low LD indicates that markers are often broken apart by recombination High LD indicates that markers are highly correlated (tightly linked) = co-exist on one haplotype

you dont need to genotype every SNP. if you know one SNP in the block you can infer the rest without sequencing them all

low LD- markers are often broken by recombination within a haplotype ( there is a recombination hotspot between them. high LD-The markers are highly correlated and inherited together since there is no recombination. ( never broken by recombination)

LD and recombination are inversely correlated

SNPS located on the same haplotype block tend to not recombine & if they dont recombine = linkage disequilibrium

High LD : low recombination

he there needs tobe a hotspotbit them.

October 27, 2005

ENr131 2q37.1 ENm014 7q31.33

Comparison of LD and recombination at two ENCODE regions

Over 1 million SNPs typed in 269 individuals

The HapMap project first enabled large-scale mapping of recombination events and haplotype blocks

derived recombination maps and found recombination hotspots and deserts and allowed mapping of haplotype blocks

Haplotype Map project -

Not just diversity

also recombination

density

(Family trios)

Relationship between recombination rates, haplotype lengths and gene locations across 19q13

Fig. 10

• No obvious relationship with gene density • Recomb rate varies with PRDM9 binding site density. • PRDM9 sites appear to have evolved from Alu/THE1 transposable elements

high LD- little recombination

short haplotype length- lots of recombination every horizontal line= different person

genelocation doesntmatter

recombine alot

PRDM bindingsite

Genetic recombination is directed away from functional genomic elements in mice Brick et al., (2012) Nature

Interesting given that PRDM9 has been lost from several branches of the Vertebrate tree (Birds, Crocodiles, Amphibians) and specific species (Dogs)

Baker et al., (2017) eLife

PRDM9 can read & modify patchessanity

Length of LD spans across the genome (HapMap project)

Fig. 15 Of note: - LD is lower at chromosome ends -> chromosome ends have higher recombination rates - Centromeric regions show high LD -> counter selected for recombination - X chromosome has highest LD; it only undergoes recombination in females - African populations have lower LD than European

telomeres- males like to recombine around telomeres

for every chromosome

lower recombination, 1/2 time there is no homolog since males so less recombination overall

recombination is suppressed

inmales

First task: Identifying recombinants vs. non-recombinants

Practical aspects of genetic linkage mapping Gene mapping = identifying markers that are genetically linked to an unknown gene of interest.

Use recombination to position genes on the genome .

In close proximity i. recombine less frequently If on the same haplotype block they won't recombine at all or rarely .

* recombination -40W

very close proximity & high linkage

> a

pad MOM

Identifying informative vs un-informative matings

In this example, A1 is given as linked to a dominant disease gene in first generation.

Disease gene still linked to A1

Uninformative with regards to

recombination

No information Unclear if disease gene is linked with

A1 or A2

Disease gene still linked to A1

Uninformative with regards to

recombination

Disease gene now linked to A1

Informative with regards to

recombination

eithercanbe

from mom or dad

- lostsome info

At Az D=dominant

NO D-

' -a a.mnfrtamt.at,

ti: recombination

Distance bit marker & disease gene Al - - Az The likelyhood of crossover is high • ma

chromosome has atleast IN2 crossover. centromere so further distance in chromosome

very likely to crossover. ×

Not - Max recombination ratebit 2 chromosome IS 501. → likely 501. = NObetter than random assortment

informative - The 2 markers at the end of the chromosomeD - - d will show a high frequency of recombination .

informative' ffD µA2 -

Rgemcoarnnebri ,

nation freq will be much

- Much less likely for DSM

908hitmnegsuameeunnahpiox.pe Free hai p f Istianghethifn'8Ye "

- High LD

Establishing “linkage phase”

Is the specific allele of the marker you are following physically linked to disease gene (on same chromosome)?

In this example, A1 is given as linked to dominant disease gene in first generation.

grandparents are dead so we cant genotype them! we cannot esablish this linkage phase, the diseased allele could be on A1 or A2 due since there could have been a recombination

- we can't tell who is recombinant or not

disease gene is on the A1 chromosome

there was no recombination even in III-1 since we do not see A6 marker in III

in most pedigrees it is hard to determine linkage. Makes it hard to determine RF.

Gene mapping = identifying markers that are genetically linked to an unknown gene of interest. 1ststep: establishing linkage phase

recombinant bc it is affected but disease is on A1 normally, now it is on A2 allele

lost

recombination DON't know if )freq-- 1 mutation "

BHAmarker linked to Al

Ot Az an & Unknown gene

recombinant Is unknown. (But Notcertain)climes =Lost phase

2nd step: Determine recombination frequencies of flanking markers, obtain evidence for linkage between disease allele and markers

• This needs to be done in the context of pedigrees, following the segregation of the desired trait

• To have sufficient statistical power, collecting data from many individuals is necessary

• The larger the recombination fraction is, the more informative samples are necessary to distinguish between linkage and random assortment. more recombination could lead to being confused with random assortment.

sample sizes .4 is very close to random assortment, would need 343 people to do this signifigantly

change markers , keep measuring recombination freq

(More the better)

- -frequency

(frequency )

Information increases .

Logarithm of Odds Ratio (LOD) scores

Provide a mathematical formula: ➢ that helps interpret potential linkage between trait and a marker ➢ that does not depend on knowledge of linkage phase ➢ and does not depend on knowledge of the actual recombination fraction

LOD score = Z = LOG ( )Likelihood that pedigree is explained by linkage to marker Likelihood that pedigree shows

no linkage

EXTRA BONUS: * LOD scores are additive across families * LOD scores can be calculated for many markers at once, even chromosome-wide

odds non recombinant/ linkage- (1-theta)^#non recomb X # recombinant

odds no linkage- (1/2)^#non recombinant

theta is a variable that can vary between 0 and .5

way toevaluate whetherthe marker or trait is linked to the disease gene

(mayor may not be linked)

Above 1 ④ →completely random

Below 1 ① assortment

LOD score analysis from 2 point mapping

Z > 3 or Z< -2 show significant evidence for or against linkage, respectively

At low recombination freq, there is significant evidence against linkage. Around 25% recombination freq, there is significant evidence for linkage. 1. Marker IS NOT closely linked to the disease gene 2. There IS Slg evidence for linkage between the marker & disease gene around 0.25% of recombination freq

Anything above 3= thousand times likely that pedigree is explained by linkage → considered to be statistical significant evidence for linkage So close to the disease marker SO likely on the same haplotype or unlike

LOD score improves as recombinant freq increase. → significant evidence against linkage → random assortment of markers → Marker is no way near the disease

Ideal outcome → Z score = above 3 → As recombination freq increases, goes down to 0 → strong evidence that pedigree is explained for linkage at low recombination freq

Meansmarkerson the same

chromosome

Ihbtt Intermediate

Not informative Not significant

[ Yet.fm#8innaotn- frequency

evidence

T '

X 1013=0 NOInfo

¥ § NOW -2,100 times B- rely that pedigree is0.8 Ee explained by random ± = assortment .

→statistical significant evidence against linkage

all curves converge at 0.5 because recombination at 50% is about same as being on 2 diff chromosomes

RED- for small theta, there is evidence of no linkage ( not close). at large theta from .15-.3 there is linkage not random assortment. - prob on the right chromosome but still pretty far from diseased gene

PURPLE- no linkage for small RF, no info after that

GREEN- desired result. tightly linked, small RF are significant. the marker is close.

Multipoint LOD score across a region

Candidate region LOD score nearly constant indicates little if any recombination = haplotype block

can do this genome wide. each dot is a LOD score

abrubt transitions- recombination hotspots

haplotype on avg is 100,000 bp

Region of interest

diseased gene of interest

Further narrowing down regions by detecting rare recombination events within haplotypes

Darier-White disease: dominant inherited skin disorder mapped to a haplotype on 12q

Disease gene locates to the interval between D12S84 and D12S129

each number is an allele

red one- it is a rare recombinant

mom is homozygous for 5 5 so we cant determine where the crossover was

marker 2 and above does not cause disease. the disease haplotype is 52622.

622 is unaffected. so 622 does not contribute to disease New diseased haplotype is 52.

autosomal dominant disease

different family has different markers.

We can assume 838 is the diseased haplotype.

2 I

¥25] 24 2×2] l l 6 5 93 25 57

↳Beabletonarrow 25 ss

interval of hotspot

f) ① ←

Dad markers

r Unknown exact crossover location

9 R # Utt From

a.EE#ed Udnaafatected mom d l

Please ChalhlsNot located in 622

Mapping of a gene for autosomal-recessive profound congenital deafness in a large inbred family lots of inbredding. autosomal recessive so afected people must

be homozygous

likely that flanking markers above and below (marker not involved) contain the diseased region

62 is shared among affected individuals ONLY

Identifying a candidate gene region is often the beginning of a long road…

human genome browser

1:13:48 of lecture

started mouse set #

Genetic mapping of complex traits

• Most human traits are not controlled by one gene with a large influence on the phenotype ➔ NOT Mendelian.

• Instead, most traits are controlled by many genes exerting subtle effects on overall phenotype.

• Environment also exerts a significant effect on many traits (e.g. Type II diabetes)

Identifying causal variants is hard!

Adopted Twin Studies separated at birth ↳ environment taken out and evaluate the influence of the genome

Polygenic models of inheritance can account for the distribution of complex traits in populations

Rationalize that there is genetic contribution?

⇐ I

£ I

fitness

Polygenic diseases manifest when the balance between “good” and “bad” alleles tips towards presenting a phenotype – liability threshold

Sibs of affected people tend to carry more liability alleles

relatives BY being related to affected individual we know the # ofbad allele.

good bad

balance .

VISK IS much

higher

more likely to

carry liability alleius

An example of a disease with sex-specific thresholds

Pyloric stenosis: - affects about 1 in 2,000 births - narrowing of the opening in the stomach - Genetic and environmental causes - affects boys 4 times more often than girls

Green: general population Blue: sibs of affected males Red: sibs of affected females

How to deal with complex traits? 1- Determine if the trait is genetically determined

Twin studies. Concordance rate is significant between MZ twins Concordance rate is higher in MZ than DZ and unrelated individuals

2- Attempt to identify families showing near-Mendelian segregation See BRCA1 mapping example below

3- Develop adequate genetic mapping methods Association studies

MOMO

MZ > Dz fraternal

Mapping the BRCA1 breast and ovarian susceptibility gene (1994)

Caveat: BRCA1 mutations are responsible for only ~20% of familial risk and even less for spontaneous cases.

GWAS studies have identified at least 100 other genetic loci associated with increased risk

First incarnation: Sib-pair mapping.

Affected sib pairs are likely to share segments of DNA closely linked to the disease allele

Drawback: shared regions are usually huge!

Association studies aim to identify shared haplotypes between affected individuals

On average, full siblings share 50% of their genome

Dominant Recessive

We’re all sibs!

On average, 2 unrelated individuals in the UK share ancestors ~22 generations ago (circa 1500)

This implies that at some low frequencies, we all share alleles that are Identical by Descent (IBD)

This enable Association Studies in all populations!

World population= ancestors

implies that at some low frequency

we might all share alleles that are

inherited inthe big humantree .

Association studies

• aim to identify markers that are shared within populations of unrelated “affected” individuals (independent of inheritance mode, genetic linkage, recombination frequency, etc…)

• rely on the fact that our genome is composed of millions of haplotype segments which are segments of DNA that are flanked by recombination hotspots and therefore tend to be inherited as blocks

• rely on the fact that even unrelated individuals tend to have shared ancestry

➢ Individuals affected by diseases tend to share the mutation-containing haplotype segment(s) compared to the general population

Limitations: • Require many, many patients to uncover common variants • Hard to ensure homogeneity of cohorts (limited reproducibility) • Only capture a limited fraction of genetic susceptibility • Often identify variants “near” genes but provide little clue as to what the variants actually do to contribute to disease

Mapping of shared, ancestral fragments within unrelated patients with Nijmegen breakage syndrome (NBS)

Common region to all patients NBN gene: chr8:90,945,564-90,996,899

Regions broken by rec. events

Nijmegen syndrome: autosomal-recessive Characterized by chromosome breakage, microcephaly, immunodeficiency, susceptibility to cancer

First step: identified a haplotype block with strong linkage with disease.

Second step: look for ancestral segments share across unrelated patients.

Looked for shared region withth haplotypes

association study :

affected individuals are more likely to share disease alleles than

controlled individuals .

Manolio et al., Nature 2009

Various types of genetic “architectures” encountered in human genetics

Strong impact on the phenotype

very rare

Latest thoughts on disease and genetic architecture in human populations

- mutation concentrated in the family - or acquired new mutations

Examples of successful gene mapping of human disease traits

Genetics of obesity as seen through Genome Wide Association Studies (GWAS)

Obesity – 2010 update These variants only explain 6-11% of the genetic susceptibility…

GIPR: Gastric inhibitory peptide receptor – defects lead to impaired insulin signaling and higher blood glucose levels (link to T2D) FTO: m6A RNA demethylase / signals to brain circuitry / circadian rythm MC4R: Melanocortin 4 receptor POMC: Pro-opiomelanocortin BDNF: Neuronal transcription factor involved in brain-specific expression and splicing NRXN3: Neurexin 3 – receptor and cell adhesion in nervous system NEGR1: Neuronal Growth regulator

Manhattan plot

All Neuronal genes

Obesity – Oct 2017 update Nature Genetics

• Confirms major involvement of Central Nervous System in obesity • Discovers new role for immune function in obesity risk

• Double # of loci associated by BMI by analyzing a distinct population

How about Alzheimer’s Disease?

APOE: Apolipoprotein E – role in cholesterol metabolism CLU: Clusterin – prevents stress-induced protein aggregation BIN1: Bridging Integrator 1 ?? PICALM: phosphatidylinositol binding clathrin assembly protein

RISK factors

role In

Chloesterol metabolism

Alzheimer’s Disease cont’d: GWAS studies can have very high resolution if sufficient SNPs are tested

highly shared .

significant

peaks should

not really be read along

shared the right

Dots -_

satistical association values

zoomed in on a small region

How about traits such as intelligence?

Sniekers et al., Nature Genetics May 2017

negatively correlates

- #ignnitticana . >

high intelligence correlates w/

Nature (2016)

Genetic correlations between EduYears and other traits

positive negative

An example of modern GWAS resource

GWAS for height

identified 100 s of genes related to height

y differs BH people

Personal Genomics: hope, hype, risks

• Knowledge of genetics variants and their functional significance is steadily increasing

• Variants have significance for disease risks but also a variety of traits and geographical origins.

• Personal genomics is on the rise and private individuals can now obtain information on their own genomes at low cost.

Where we are today:

Personal Genomics: hope, hype, risks

Hope: • If I am at risk for a particular disease, knowledge is power

❑ “Passive” attitude: prepare for what’s to come ❑ Active attitude: correct the mutations (in self or in offspring)

Genetics + DNA editing = End of Genetic Diseases

Risks: • Genetic discrimination (confidentiality of information)

GATTACA http://www.imdb.com/title/tt0119177/ • Human germline editing?

https://en.wikipedia.org/wiki/He_Jiankui_affair

https://www.youtube.com/watch?v=k99bMtg4zRk CRISPR-Cas9 ("Mr. Sandman" Parody) | A Capella Science

Hype: • Do we really know what we are doing?

Aaaa

a Aaaa a Aaaa

1- Avoiding Transmission of Genetic Disease – Peter Braude / King’s College, London

2- Genetic Basis of Human Disease and Implications for Germline Editing – Eric Lander / MIT

Avoiding Transmission of Genetic Disease

Professor Peter Braude Division of Women’s Health

Kings College, London

Centre for Preimplantation Genetic Diagnosis

Avoiding Transmission of Genetic Disease

WHY? • The health legacy of genetic disorders • Reproductive options for couples with history of

recurrent genetic risk HOW? • Prenatal Genetic Diagnostic (PGD) and

Prenatal Genetic Screening (PGS)

Centre for Preimplantation Genetic Diagnosis

Health legacy of genetic disorders

Sporadic early embryonic loss Recurrent pregnancy loss Anatomical abnormality Mental disability Neonatal and childhood death Chronic disease and early demise Late onset disease

Centre for Preimplantation Genetic Diagnosis

Recurrent: As a result of inherited disorders

Sporadic:

Random, often age related

Two kinds of genetic risk

Centre for Preimplantation Genetic Diagnosis

Autosomal Recessive Genetic Disease

1:4 free of disease; 1:2 carriers – but asymptomatic; 1:4 will be affected

Cystic fibrosis Spinal muscular atrophy Sickle cell disease Tay-Sachs disease Fanconi Anaemia

Autosomal Dominant Genetic Disease

Huntington’s Disease Achondroplasia Marfan Syndrome Neurofibromatosis Retinoblastoma

50:50 (1:2) chance of passing on disease

Haemophilia

Genetic disease

Women are carriers but generally don't have symptoms; only males affected

from mom

Reproductive options for those with serious recurrent genetic risk

Gamete donation Adoption Remain childless

Reproductive roulette

Prenatal diagnosis and termination of pregnancy

Centre for Preimplantation Genetic Diagnosis

highriskof early termination

pregnancy

needleto collectDNA

placenta

Preimplantation Genetic Testing

perform genetictesting 10h the 1 Cell .

Preimplantation Genetic Testing

Detection of genetic information in an embryo made by examining a representative sample taken at a preimplantation stage of development

Centre for Preimplantation Genetic Diagnosis

Early human development in vitro

Late Day 1 2-cellsDay 1 Fertilised egg Day 2 4-cells Day 3 8-cells

Day 4 Morula

Day 5 BlastocystDay 6 Hatching blastocystLate Day 6 Hatched blastocyst

ICM

not fused yet

differentiation occurred

Tissues for Preimplantation Biopsy

Polar Body Blastomere Trophectoderm

egg cleavage stage blastocyst

collect from polar area that will body become placenta / or

Recurrent: As a result of inherited disorders

Sporadic:

Random, often age related

Two kinds of genetic risk Two kinds of genetic test

PGD

To diagnose known genetic condition

PGS

To screen for random aneuploidy to improve infertility outcome

Centre for Preimplantation Genetic Diagnosis

Diagnostic

screening

The principle of PGD

Select for transfer to the patient

affected affectedaffected unaffectedunaffected

Centre for Preimplantation Genetic Diagnosis

Questions: - selection of embryos - oocyte collection / IVF

- expensive - Best

oocyte to

grow is not

clear

affordability & equity

(quality of

being fait )

- not coveredby

insistence .

classic IVF

A broad look at PGD / PGS: ESHRE PGD Consortium members by country

number of centres: 11

number of centres: 86

number of centres: 27

Total number of centres: 124

(June 2015)

ESHRE 2015

Indications PGD & PGS

1000

2000

3000

4000

5000

6000

7000

PGS

social s exing

sexing for X-linked disorders

chromosomal abnormal ities

monogenic dis ease

ESHRE 2015

54,589 cycles

ESHRE 2015 Coonen: Unpublished and still to be verified data

clearly Hsing

y down

syndrome +

aneuploidy

Monogenics

Chromosomal

Sexing X linked

Aneuploidy

Social sexing 33 741 (58.0%)

668 (1.1%)

12 885 (22.2%)

9 081 (15.6%)

ESHRE 2015 Coonen: Unpublished and still to be verified data

REASONS FOR EMBRYO BIOPSY ESHRE Consortium data I-XV

Based on 54,589 cycles

Range of PGD Cases for 2014

Centre for Preimplantation Genetic Diagnosis

191 biopsy cases

CISAC

Huntington . Fibrosis

PGS leads to over-estimation of genetic risk due to cleavage stage mosaicism

15, 490 - 491 (2009)

Ext NK e I.ee#iiEiet et et r

µ 4¥

May die

& bereplaced by

normal embryo IS NOT Uniform .

cells .

48% of cells tested normal at blastocyst stage

52% not normal

Ermanno Greco, M.D. Maria Giulia Minasi, M.Sc. Francesco Fiorentino, PhD. European Hospital & Genoma Molecular Genetics Laboratory Rome, Italy

cells w/ chromosomal

abnormalities ended up dying and embryo is populated by normal cells

In the light of recent re-evaluations of the risk benefit ratio, it seems that PGS has been oversold

“Rather than improving outcome for childless couples, PGS encourages the waste of healthy embryos which are excluded from transfer to the uterus.”

“The procedure just appears to increase costs and complexities of IVF. Its utilization, at present, should therefore be acknowledged as highly

experimental and refuted in routine IVF care.”

Number of PGS cycles reaching a plateau?

500

1000

1500

2000

2500

3000

3500

4000

4500

PGS cycles over time

ESHRE 2015ESHRE 2015 Coonen: Unpublished and still to be verified data

Per Egg Collection Per Embryo Transfer

Results of 2091 cycles in one PGD only centre (1998 -2014)

Centre for Preimplantation Genetic Diagnosis

Ongoing Pregnancy Rate

Success rate of IVF after PGD is 30-45% doesn't effect

success rate

similar to

IVF

Human Diseases and Traits

Rare, Mendelian Cystic fibrosis, Huntington Disease, Diastrophic Dysplasia …

Common, polygenic Heart disease, Alzheimer’s Schizophrenia, Height, Obesity Intelligence? . . .

Avoid all cases of severe genetic disease Eliminate disease alleles from population

Eliminate disease risk ’Enhance’ human population

Goals of editing intervention:

Eric Lander - MIT

Rare Mendelian Disease: Dominant

D+ ++

D+ D+ ++ ++

Heterozygous parent Half of offspring affected Half of offspring unaffected

Can use pre-implantation diagnostics (PGD) PGD+germline editing adds relatively little

DD ++

D+ D+ D+ D+

Homozygous parent All offspring affected

Germline editing would be useful Homozygotes are extremely rare For Huntington’s disease,

only dozens of cases found worldwide

t.tt/FFHtO*; Aa

Rare Mendelian Disease: Recessive

m+ m+

mm m+ +m ++

Heterozygous unaffected parents One-quarter of offspring affected

To avoid affected offspring: Can use pre-implantation diagnostics (PGD) PGD+germline editing adds relatively little

To avoid most cases of devastating genetic diseases, the most important intervention would be ensuring access to genetic testing so carrier couples know they are at risk

To eliminate disease alleles from population, we’d all need to use IVF – since we all carry multiple disease genes in heterozygous state

NO need for

editing

- O

Rare Mendelian Disease: Recessive

m+ m+

mm m+ +m ++

Heterozygous unaffected parents One-quarter of offspring affected

To avoid affected offspring: Preimplantation diagnostics available (PGD) PGD+germline editing adds relatively little

mm mm

mm mm mm mm

Homozygous parents All offspring affected

Germline editing would be useful Very rare, unless brought together by disease E.g.: Deaf parents with mutations in same gene

Human Diseases and Traits

Rare, Mendelian Cystic fibrosis, Huntington Disease, …

Common, polygenic Heart disease, Alzheimer’s Schizophrenia, Height, Obesity Intelligence? . . .

Avoid all cases of severe genetic disease Eliminate disease alleles from pop’n

Decrease disease risk ‘Enhance’ human population

Common, Polygenic Disease

Genetic variants have modest effects Handful: 3-5-fold 99+%: <1.2-fold

Why? Selection keeps strong alleles at

low frequency Disease processes are buffered

20 '

a effect

on thephenotype

strong alleles don't

have a strongphenotype .

Common, Polygenic Disease

Genetic variants have modest effects Handful: 3-5-fold 99+%: <1.2-fold

Why? Selection keeps strong alleles at

low frequency Disease processes are buffered

Schizophrenia

Population 1% risk C4 gene 1.1% risk

Polygenic risk score • Top 100 loci (statistically significant) Top decile: ~3% risk

• Top 10,000 loci (only a fraction significant) Top decile: ~10% risk

Common, Polygenic Disease

Is there a free lunch? Genetic variants have ‘pleiotropic’ effects

and environmental interactions

Inflammatory bowel disease Lower risk of: Higher risk of:

FUT2 Norovirus Crohn's & Type 1 diabetes IFIH1 Type 1 diabetes Crohn's disease RNF186 Ulcerative colitis Chronic kidney disease

Viral infection Lower risk of: Higher risk of:

CCR5 HIV West Nile (13x higher risk for fatal cases)

⑦ Are there any detrimental sideeffects,yes .

-effect multiple

diffprocesses .

Common, Polygenic Disease: Germline editing

Avoid deleterious variants? Most have very small effects

Those with large effects usually rare: treat like Mendelian

Bestow protective variants with large effects? Very few examples overall

Moreover, want common (to assess impact in homozygotes) ideally, with no downsides (undesired pleiotropic effects)

can we NO

canyou edit the genome to allow protective variants w/ strongeffect?

↳ downside Unknown

Common, Polygenic Disease: Germline editing

Avoid deleterious variants? Most have very small effects

Those with large effects usually rare: treat like Mendelian

Bestow protective variants with large effects? Very few examples overall

Moreover, want common (to assess impact in homozygotes) ideally, with no downsides (undesired pleiotropic effects)

Best candidates: • ApoE2, 3 vs. ApoE4 (3%) Higher Alzheimer’s risk • PCSK9 null (<2%) Lower LDL levels, heart attack risk

Still, we have incomplete knowledge about pleiotropic effects If alleles are so good, why aren’t they at higher frequency?

Bad

• remove ApoE4 In favor of APOEL ,3 Few genes to edit )

The world’s first CRISPR-edited babies was a fiasco Target: CCR5 gene – HIV cell surface receptor Rationale: individuals naturally carry truncated copies of CCR5 and are immune to HIV. Father was HIV positive. Goal was to avoid HIV transmission. Ethically: consent was given by parents but using vague language. Study was cloaked in secrecy and not adequately reviewed. Preventing HIV transmission to offspring in IVF is routinely done with sperm washing, rendering editing unnecessary. Medically: edited babies may now face shortened lifespan. In addition, they may have higher risk of infection by influenza and West Nile.

Summary and Conclusion Rare, Mendelian diseases

• Vast majority of cases can be addressed by IVF and PGD • Some cases of compelling need, although rare • If we wish to avoid devastating genetic diseases, most important intervention

is ensuring couples have access to genetic testing to know they are at risk

Common, Polygenic diseases • Thousands of genes being identified

— revealing disease processes, pointing to therapeutic hypotheses • For vast majority of variants, impact on risk is small • Currently, at most a few plausible variants for editing

Conclusion • Genetic basis of human disease is complex • We still have a lot to learn • Before making permanent changes

to the human gene pool, we should use great caution