Ch. 10.4

Front 1

Of thousands of known human disease genes

Back 1

can ID only small # based on specifics of abnormal condition

Front 2

Sickle cell anemia & thalassemias

Back 2

diseases affecting RBC

Front 3

97% dry wt of RBC consists of

Back 3

hemoglobin, so researchers directed attention to genes encoding polypeptides making up this oxygen-carrying protein as likely causes of diseases

Front 4

Positional cloning

Back 4

used to ID defects causing hereditary diseases

Front 5

Object is to obtain info about

Back 5

unknown location of disease gene by finding polymorphic loci to which mutation is genetically linked

Front 6

Bc we know from human genome seq exact position of each locus

Back 6

discovering anonymous DNA polymorphisms closely linked to disease gene allows focus search for mutation on small region of single chrom

Front 7

From candidate genes w/in region

Back 7

gene responsible for disease can be found by looking for mutations that appear consistently in patients

Front 8

Positional cloning

Back 8

straightforward extension of linkage, but instead of tracking two gene phenotypes, you track one locus and second by direct DNA genotyping of each person.

Front 9

You are looking for same thing at both loci

Back 9

variation in DNA, but phenotype IDs variant indirectly

Front 10

Instead of dealing with 2-3 loci at a time

Back 10

can use DNA microarrays to follow millions of anonymous loci in each person in the pedigree

Front 11

First goal of positional cloning

Back 11

discover DNA marker that shows linkage to disease locus

Front 12

DNA microarrays are so densely packed w polymorphic loci

Back 12

many of them must in fact be linked to any given Mendelian disease gene

Front 13

If microarray had just 1000 molecular markers spread out over entire human genome

Back 13

would be average 3 Mb apart

Front 14

Disease causing mutation would have to lie

Back 14

within 3 Mb of one polymorphic locus on microarray & two must be genetically linked

Front 15

Modern DNA microarrays can simultaneously

Back 15

analyze millions of polymorphisms in a person’s DNA so positional cloning method potentially could map disease genes even more precisely

Front 16

Neurofibromatosis

Back 16

dominant, fully penetrant, autosomal condition, rare but affects > 100,000 Americans

Front 17

Disease causes nervous tissue to

Back 17

proliferate uncontrollably, forming tumorous bumps under skin

Front 18

Altho tumors usually benign

Back 18

can damage nerve cells & sometimes develop into malignant cancers

Front 19

Microarray data includes

Back 19

multipoint info about behavior of millions of DNA loci w respect to each other

Front 20

Can now pinpoint particular crossover events occurring during

Back 20

production of a single gamete to positions bw 2 polymorphisms on same chrom

Front 21

Not every mating bw 2 people provides

Back 21

interpretable info about relative positions of any 2 given loci & human family sizes are small so its difficult to obtain sufficient data for precise mapping

Front 22

Phase problem can be resolved in either of two circumstances

Back 22

if you know genotypes of two loci in both of a person’s parents, you may determine which alleles came from one parent & which from other, would need to have genotyping info about affected parent

Front 23

If two loci are sufficiently close together

Back 23

can infer probable phase bc linked alleles should segregate with each other more often than not

Front 24

Even if you know phase in doubly heterozygous parent

Back 24

mating may not provide any useful info about whether two loci are linked

Front 25

Basic requirement for genetic mapping

Back 25

at least one parent is double heterozygote

Front 26

Even if a mating is noninformative for linkage of disease gene w particular SNP locus

Back 26

multipoint analysis on microarrays usually provides a way for scientists to overcome constraint bc microarray will likely contain other nearby SNPs that will be informative

Front 27

W millions of polymorphic loci on a DNA chip, should be possible in theory to map disease genes extremely accurately even if certain crosses are uninformative

Back 27

but resolution of positional cloning is always limited in practice by # people human geneticists can track in families in which disease is segregating

Front 28

If scientists have mapped disease gene to within 1 cM of DNA polymorphism

Back 28

they have examined phenotypes of at least 100 members of such families & genotyped on microarrays DNA of all people

Front 29

Lod score

Back 29

similar to chi squared statistic to determine whether data sufficient to conclude w confidence whether a disease gene & DNA marker are genetically linked

Front 30

Lod score used in human genetics bc

Back 30

better handles small # of data pts while allowing data obtained from many different pedigrees to be combined

Front 31

Lod score statistic calculated from

Back 31

ratio of probability of obtaining particular set of results in a pedigree if 2 loci unlinked (assuming particular RF value) & chance of observing same result in loci unlinked

Front 32

Lod score statistic is base

Back 32

10 logarithm of likelihood ratio

Front 33

Convention adopted by human geneticists

Back 33

Lod score greater than or equal to 3 indicates 2 loci linked

Front 34

Lod score of 3 means

Back 34

1000 times more likely that 2 loci are linked than not (bc 3 = log1000)

Front 35

Bc it is a log function, Lod scores from diff pedigrees

Back 35

may simply be added so you know when enough data to conclude disease allele linked to marker

Front 36

In human genome, average gene density

Back 36

1 gene per 100 kb DNA

Front 37

Might be possible to find clues for genes in region by

Back 37

looking for changes in patients in amts or sizes mRNA transcripts or protein products of genes

Front 38

Most generally useful strategy

Back 38

use PCR to amplify DNA from all candidate genes in all available patients then sequence all these PCR products

Front 39

If patients all had identifiable mutations in one candidate gene

Back 39

particularly mutations that might affect aa seq of gene’s protein product, very strong evidence for ID that candidate as actual disease gene

Front 40

Some genetic conditions always caused by

Back 40

same single mutation in a single gene, ie DNA sequencing would reveal the same mutation in the genomic DNA of all patients and carriers of sickle cell anemia

Front 41

Allelic heterogeneity

Back 41

displayed by other genetic diseases that can be caused by variety of different mutations in same gene

Front 42

Cystic fibrosis

Back 42

recessive autosomal genetic condition inherited by 1 child in every 2500 born from 2 parents of European descent

Front 43

Children w disease have symptoms

Back 43

from abnormally viscous secretions in lungs, pancreas, sweat glands & several other tissues, most die before age of 30

Front 44

Positional cloning strategies

Back 44

allow narrow search for causative gene to 400 kb region bw 2 DNA markers on chrom 7 with 3 candidate genes

Front 45

CFTR

Back 45

encodes cystic fibrosis transmembrane receptor that allows chloride ions to pass thru cell membranes

Front 46

Both CFTR copies in all cystic fibrosis patients

Back 46

found to contain mutations that would alter aa seq of protein or prevent normal amts of protein from being synthesized, so CFTR clearly responsible for cystic fibrosis

Front 47

Mutation delta F508 (removes phenylalanine F from position 508 of protein)

Back 47

accounts for 2/3 all mutant CFTR alleles worldwide

Front 48

Remaining alleles consist of

Back 48

>1500 different rare mutations

Front 49

Compound heterozygotes

Back 49

aka trans heterozygotes, one copy of chrom 7 has one mutation in CFTR & other has different CFTR mutation

Front 50

Disease results because

Back 50

neither chrom 7 can encode normal transmembrane receptor & in effect two recessive CFTR mutations fail to complement each other

Front 51

Ivacaftor

Back 51

drug for patients who have one specific CFTR mutation G551D (change glycine at 551 to aspartic acid)

Front 52

Mutant protein encoded by allele assembles

Back 52

properly into cell membrane but G551D protein is inefficient in transporting Cl- ions across membrane

Front 53

Ivacaftor interacts specifically at

Back 53

cell surface w G551D mutant CFTR protein, enhancing its ability to transport Cl-

Front 54

Treatment has been

Back 54

very effective in preventing symptoms but only accounts for 4% all mutant CFTR alleles in human population

Front 55

Challenging to develop drugs that can counter more prevalent delta F508 mutation

Back 55

bc defective CFTR protein it encodes cannot fold up properly & so is not inserted into cell membranes

Front 56

Locus heterogeneity

Back 56

diseases caused by mutations in one of two or more different genes, ie deafness

Front 57

Complex traits

Back 57

aka quantitative traits, ie high BP, many diff genes can influence phenotype even in single individual

Front 58

DNA microarrays w millions of SNPs sample

Back 58

only small proportion of variation bw human genomes & can suggest only a disease gene’s general chromosomal location

Front 59

Disease gene ID eventually requires

Back 59

DNA sequencing to correlate disease phenotype w actual mutations

Front 60

If we could cheaply & accurately sequence all of nucleotides in affected person’s genome

Back 60

whole genome sequence must include causative mutation

Front 61

Unlike positional cloning (first goal to find marker)

Back 61

whole genome goal is to find directly a DNA alteration that IS the disease allele

Front 62

Whole genome sequencing still costly enough

Back 62

researchers often economize by sequencing just portion of genome corresponding to protein-coding exons, often informative bc many tho far from all disease-causing mutations alter aa seq of protein

Front 63

Whole exome sequencing

Back 63

enrich (by hybridization to cDNA seq) for genomic DNA fragments that correspond to exons of all genes then sequence these fragments

Front 64

Exome (collection of all exons of all genes)

Back 64

constitutes less than 2% of whole-genome DNA, so sequencing it requires many fewer sequencing reads than whole genome seq

Front 65

High throughput or massively parallel sequencing

Back 65

similar to Sanger but individual DNA mlcs synthesized by DNA pol are anchored in one place, methods control base addition temporally so each base can be ID before next one added, in some systems sensitivity of detection is so high that single mlc DNA can be monitored w/o need for cloning or PCR amplification steps

Front 66

Combination of three innovations allows sequencing machines

Back 66

to record successive addition of nt to each of millions of growing DNA mlcs in real time

Front 67

Patient’s whole exome or whole genome seq should include

Back 67

sequence differences responsible for genetic disease but possession of this sequence info does not guarantee that geneticists will be able to ID responsible mutations

Front 68

Technical problem

Back 68

no genome sequence is 100% accurate or complete

Front 69

All sequencing methods have

Back 69

low but real error rate in identifying nt & random sampling of DNA fragments will leave some regions of genome unsequenced

Front 70

Issues can be minimized by

Back 70

coverage of 10+ genome equivalents but cannot be eliminated completely

Front 71

Amount of variation in human genomes is so huge that

Back 71

our ability to deal w whole genome sequences is still limited & responsible mutation has yet to be identified

Front 72

Logic of whole genome or whole exome sequencing requires that

Back 72

DNA variants that are disease alleles will be rare in population, allows predictions about which of variations in genome could be responsible for disease, depending on pedigree (sex linked, autosomal, incomplete penetrance?)

Front 73

In case of rare dominant condition

Back 73

highly likely that patient should be heterozygous for causative allele

Front 74

Related patients should have

Back 74

same rare mutant allele whereas unrelated patients might have diff mutations in same gene

Front 75

If condition is recessive

Back 75

focus on rare mutations homozygous in patient’s genome, particularly if related even distantly

Front 76

If condition recessive & parents unrelated

Back 76

patient could be compound heterozygote w 2 diff mutant alleles in same gene, so look in patient’s DNA for gene affected by 2 diff mutations

Front 77

If inheritance pattern shows sex linkage, search for candidate genes limited to

Back 77

X-chromosome, which is excluded in case of autosomal

Front 78

DNA seq info from patient’s relatives particularly useful in

Back 78

narrowing list of candidate polymorphisms

Front 79

SNP genotyping of relatives on microarrays

Back 79

could narrow search to region bw 2 known SNPs

Front 80

Positional cloning & whole genome sequencing

Back 80

not mutually exclusive approaches to disease identification but can provide complementary info

Front 81

Tho more expensive, better yet are

Back 81

comparisons of patient’s whole genome or exome sequence with those of parents and/or siblings

Front 82

Bro & sis had Miller syndrome

Back 82

rare condition affects development of face & limbs but neither parent affected, suggest recessive autosomal with two children inherited mutant alleles from heterozygous, carrier parents

Front 83

To find Miller syndrome gene

Back 83

sequenced entire genomes of both children & both parents

Front 84

Identical regions

Back 84

affected bro & sis had same maternally & paternally derived alleles

Front 85

Nonidentical regions

Back 85

sibling share no alleles

Front 86

Haploidentical maternal regions

Back 86

siblings have same allele from mother but diff alleles from father