Epidemiology 217 Epidemiology 217 Molecular and Genetic Molecular and Genetic Epidemiology I Epidemiology I Course Director: Course Director: John Witte John Witte Professor of Epidemiology Professor of Epidemiology & Biostatistics & Biostatistics
Jan 16, 2016
Epidemiology 217Epidemiology 217Molecular and Genetic Molecular and Genetic
Epidemiology IEpidemiology ICourse Director: Course Director:
John WitteJohn Witte
Professor of Epidemiology & Professor of Epidemiology & BiostatisticsBiostatistics
OutlineOutline
I.I. Class DetailsClass Details
II.II. Introduction to Molecular and Introduction to Molecular and Genetic Epidemiology: an exampleGenetic Epidemiology: an example
III.III. Notes on Human GeneticsNotes on Human Genetics
IV.IV. Determining the genetic Determining the genetic component to disease before component to disease before looking at DNAlooking at DNA
V.V. ExperimentExperiment
I. DetailsI. Details
Classes: 10 Tuesdays, 1:15-2:45 pmClasses: 10 Tuesdays, 1:15-2:45 pm Office hours: by appointment Office hours: by appointment
Course Director: Course Director: John S. Witte, PhDJohn S. Witte, PhDPhone: 415-502-6882Phone: 415-502-6882email: email: [email protected]@humgen.ucsf.edu
Teaching Assistant: Nerissa Ko, MDTeaching Assistant: Nerissa Ko, MDPhone: 415-353-1489Phone: 415-353-1489email: email: [email protected]@ucsfmedctr.org
Course website:Course website:www.epibiostat.ucsf.edu/courses/schedule/mol_methodsi.htmlwww.epibiostat.ucsf.edu/courses/schedule/mol_methodsi.html
GoalsGoals Learn about:Learn about:
common molecular and genetic measurescommon molecular and genetic measures determining the genetic contribution to diseasedetermining the genetic contribution to disease searching for disease-causing genes, searching for disease-causing genes, and their interaction with environmental factorsand their interaction with environmental factors pharmacogenomics; proteomics; and pharmacogenomics; proteomics; and
bioinformatics.bioinformatics.
Main goal:Main goal: develop a framework for develop a framework for interpreting, assessing, and incorporating interpreting, assessing, and incorporating molecular and genetic measures in your own molecular and genetic measures in your own research. research.
SyllabusSyllabus
DateDate LecturerLecturer Title / ContentTitle / Content
1/21/2 Eric Jorgenson Eric Jorgenson Introduction to Molecular Introduction to Molecular and Genetic Epidemiologyand Genetic Epidemiology
1/91/9 Joe WiemelsJoe Wiemels Molecular and Genetic Molecular and Genetic Measures Measures
1/161/16 John WitteJohn Witte Candidate Gene Studies I: Candidate Gene Studies I: Design Design
1/231/23 John WitteJohn Witte Candidate Gene Studies II: Candidate Gene Studies II: AnalysisAnalysis
1/301/30 Eric JorgensonEric Jorgenson Genome-Wide Studies: Genome-Wide Studies: LinkageLinkage
2/62/6 Eric JorgensonEric Jorgenson Genome-Wide Studies: Genome-Wide Studies: AssociationAssociation
2/132/13 Neil RischNeil Risch Ethnicity and Race in Ethnicity and Race in Genetic EpidemiologyGenetic Epidemiology
2/202/20 John WitteJohn Witte Proteomics and Proteomics and BioinformaticsBioinformatics
2/272/27 Kathy GiacominiKathy Giacomini PharmacogenomicsPharmacogenomics
3/63/6 Joe WiemelsJoe WiemelsJohn WitteJohn Witte
Incorporating Molecular Incorporating Molecular and Genetic Measures; and Genetic Measures; Putting it all TogetherPutting it all Together
Homework AssignmentsHomework Assignments
Count for 70% of grade (30% for final exam).Count for 70% of grade (30% for final exam). Weekly readings and / or brief problem sets.Weekly readings and / or brief problem sets. Readings give important background Readings give important background
information, and should be completed before information, and should be completed before the start of the corresponding lecture. the start of the corresponding lecture.
Problem sets are due at 8 pm on Mondays, so Problem sets are due at 8 pm on Mondays, so we can discuss the following day.we can discuss the following day.
Late assignments are not accepted.Late assignments are not accepted.
Web ResourcesWeb Resources
Video from UAB Short course on statistical Video from UAB Short course on statistical genetics: genetics: http://www.soph.uab.edu/ssg_content.asp?id=1http://www.soph.uab.edu/ssg_content.asp?id=1174174
Dorak’s notes on genetics: Dorak’s notes on genetics: http://dorakmt.tripod.com/genetics/http://dorakmt.tripod.com/genetics/
Strachan & Read’s Human Molecular Genetics:Strachan & Read’s Human Molecular Genetics:
http://www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=http://www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=hmghmg
II. Introduction: how was II. Introduction: how was lunch?lunch?
?
Impact of folate, B12, and homocysteine on cognitive function?
How can we measure How can we measure these factors?these factors?
Problems?
Can we improve our Can we improve our measurements?measurements?
Look at circulating levels in plasma
What else will impact What else will impact these levels?these levels?
Methylene tetrahydrofolate reductase Methylene tetrahydrofolate reductase (MTHFR): (MTHFR): e.g., catalyzes the last step in conversion of e.g., catalyzes the last step in conversion of folic acid to its active form, 5-folic acid to its active form, 5-methyltetrahydrofolate (5MTHF). methyltetrahydrofolate (5MTHF).
MTHFRMTHFR gene gene
Single nucleotide polymorphisms (SNPs) in Single nucleotide polymorphisms (SNPs) in MTHFRMTHFR::
C677T C677T (C and T are alleles; CC, CT, TT are genotypes)(C and T are alleles; CC, CT, TT are genotypes)
A1298C A1298C (A and C are alleles; AA, AC, CC are genotypes)(A and C are alleles; AA, AC, CC are genotypes)
E.g., if an individual is homozygous for the 677TT SNP, E.g., if an individual is homozygous for the 677TT SNP, MTHFR enzymatic activity can decrease by 50%.MTHFR enzymatic activity can decrease by 50%.
This may in turn reduce cognitive function. This may in turn reduce cognitive function.
Locus
Diet, plasma, & genotype Diet, plasma, & genotype interactioninteraction
Look at how these work in Look at how these work in conjunction with each other to affect conjunction with each other to affect cognitive functioning!cognitive functioning!
III: Notes on Human III: Notes on Human Genetics: DNAGenetics: DNA
Human ChromosomesHuman Chromosomes
Human Chromosome 21Human Chromosome 21
TelomeresTelomeres
CentromereCentromere
p stands for petitp stands for petit
q stands for grandq stands for grand
21q22.1 is 21q22.1 is pronounced twenty-pronounced twenty-one q two two point one q two two point oneone
Human Genome Human Genome StatisticsStatistics
• 3,253,037,807 basepairs3,253,037,807 basepairs
• 21,774 known genes21,774 known genes• 1,036 novel genes1,036 novel genes• 1,069 pseudogenes1,069 pseudogenes• 3,976 RNA genes3,976 RNA genes• 270,661 exons270,661 exons
http://www.ensembl.org/Homo_sapiens/index.html
Transcription and Transcription and TranslationTranslation
Potentially Functional Potentially Functional Regions of a GeneRegions of a Gene
cis regulator
Amino acid codingAmino acid coding
RNA processing
Transcription regulation
promoter
Human Genome Human Genome VariationVariation
• 3,253,037,807 basepairs3,253,037,807 basepairs• Mutation rate ≈ 10Mutation rate ≈ 10-8 -8 per bp per per bp per
generationgeneration• 65 new mutations expected in each 65 new mutations expected in each
personperson
• Compare two copies in any one personCompare two copies in any one person• 1 variant per 1,331 basepairs1 variant per 1,331 basepairs• 2,444,055 variants2,444,055 variants
• Most variants are oldMost variants are old
Mutation and MeiosisMutation and Meiosis
IV. How can we tell that a IV. How can we tell that a disease is genetic?disease is genetic?
Without looking at DNA…Without looking at DNA…
1.1.Ecologic Studies (Migrant Ecologic Studies (Migrant Studies)Studies)
2.2.Familial Aggregation:Familial Aggregation: Family StudiesFamily Studies Twin StudiesTwin Studies
3.3.Segregation analysesSegregation analyses
Break: How about you?Break: How about you?
Research interests?Research interests?
Background / training in molecular / Background / training in molecular / genetics?genetics?
Ever used or considered using Ever used or considered using molecular or genetic measures in molecular or genetic measures in clinical research?clinical research?
IV.1 Ecologic Studies IV.1 Ecologic Studies (Migrant)(Migrant)
Weeks, Population. 7th ed. London: Wadsworth Publishing Co 1999
Example: Cancer (SMRs)Example: Cancer (SMRs)
Not Not U.S.U.S.
U.S.U.S.BornBorn
U.S.U.S.Cauc.Cauc.
CancerCancer JapanJapan Born Born
Stomach Stomach (M)(M)
100100 7272 3838 1717
Intestine Intestine (F)(F)
100100 218218 209209 483483
Breast (F)Breast (F) 100100 166166 136136 591591
(MacMahon B, Pugh TF. Epidemiology: Principles and Methods. Boston: Little, Brown and Co, 1970:178.)
IV.2 Familial AggregationIV.2 Familial Aggregation
Does disease tend to run in families?Does disease tend to run in families? Example: Men who have a brother or Example: Men who have a brother or
father with prostate cancer have 2-3 times father with prostate cancer have 2-3 times the risk of men without a family history. the risk of men without a family history.
Possible study designs:Possible study designs:1.1.Case-controlCase-control: compare the family history : compare the family history
between cases versus controls.between cases versus controls.
2.2.CohortCohort: view the family members of the cases : view the family members of the cases and controls as two cohorts, one exposed (i.e., and controls as two cohorts, one exposed (i.e., to a case), the other not exposed.to a case), the other not exposed.
Twin StudiesTwin Studies
MZ Twins (Identical)MZ Twins (Identical)
Twin 1
Twin 2
Both alleles are shared identical by descent (IBD)
DZ Twins (Fraternal)DZ Twins (Fraternal)
Twin 1
2 1 1 0
Twin 2 can be any of the four
IBD can be 2, 1, or 0
DZ Twins (Fraternal)DZ Twins (Fraternal)
Twin 1
100% 50% 50%0%
Average sharing is 50%
Twin StudiesTwin Studies
Compare the disease concordance ratesCompare the disease concordance rates
of MZ (identical) and DZ (fraternal) twins.of MZ (identical) and DZ (fraternal) twins.
DiseasDiseasee
YesYes NoNo
YesYes AA BB
NoNo CC DD
Twin 1
Twin 2
Then one can estimate heritability (the proportion of the variance of an underlying disease liability due to common genes), and environmentality.
Concordance = 2A/(2A+B+C)
Example of Twin Study: Example of Twin Study: Prostate CancerProstate Cancer
TwinTwin ConcordaConcordant pairs nt pairs (A)(A)
DiscordanDiscordant pairs t pairs (B+C)(B+C)
ConcordanConcordancece
MZMZ 4040 299299 0.210.21
DZDZ 2020 584584 0.060.06
Heritability: 0.42 (0.29-0.50)Non-shared Environment: 0.58 (0.50-0.67)
Lichtenstein et al NEJM 2000 13;343:78-85.
• Twin registry (Sweden, Denmark, and Finland)
7,231 MZ and 13,769 DZ Twins (male)
IV.3 Segregation AnalysisIV.3 Segregation Analysis
Evaluate whether the pattern of disease among Evaluate whether the pattern of disease among relatives is compatible with a single major relatives is compatible with a single major gene, polygenes, or simply shared environment.gene, polygenes, or simply shared environment.
Fit formal genetic models to data on disease Fit formal genetic models to data on disease phenotypes of family members.phenotypes of family members.
The parameters of the model are generally The parameters of the model are generally fitted finding the values that maximize the fitted finding the values that maximize the probability (likelihood) of the observed data.probability (likelihood) of the observed data.
If there appears to be a single major gene, then If there appears to be a single major gene, then one can estimate its dominance, penetrance, one can estimate its dominance, penetrance, and allele frequency.and allele frequency.
Taste TestTaste Test
Bimodal Distribution of Bimodal Distribution of PTCPTC
PTC Distribution
0
5
10
15
20
25
30
35
40
45
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Raw PTC Score
Num
ber
of Subje
cts
Recessive traitRecessive trait
First PTC Family StudyFirst PTC Family Study
ChildrenNo. of families Can taste Can not taste
Both parents can taste 40 90 16One parent can taste, the other can not 51 80 37Neither parent can taste 9 0 17
L. H. Snyder Science 1931
SummarySummary Molecular measures can improve upon Molecular measures can improve upon
conventional questionnaire-based conventional questionnaire-based measurements.measurements.
Genetics can impact many exposures and Genetics can impact many exposures and diseases.diseases.
We can assess the heritability with studies of We can assess the heritability with studies of populations and families, including:populations and families, including:
1.1. Migrant studiesMigrant studies
2.2. Familial aggregation studiesFamilial aggregation studies
3.3. Family/Twin studiesFamily/Twin studies
4.4. Segregation analysesSegregation analyses