Title: Rick Kittles, Ph'D'
1The Role of Diverse Populations in Personalized
Genetic Medicine
TTCTTTCATGGGGAAGCAGATTTGGGTACCACCCAAGTATTGACTCACCC
ATCAACAACCGCTATGTATTTCGTACATTACTGCCAGCCACCATGAATAT
TGTACGGTACCATAAATACTTGACCACCTGTAGTACATAAAAACCCAATC
CACATCAAAACCCTCCCCCCATGCTTACAAGCAAGTACAGCAATCAACCT
TCAACTATCACACATCAACTGCAACTCCAAAGCCACCCCTTACCCATTAG
GATATCAACAAACCTACCCGCCCTTAACAGTACATAGCACATAAAGCCAT
TTACCGTACATAGCACATTACAGTCAAATCCCTTCTCGCCCCCATGGATG
ACCCCCCTCATTCTTTCATGGGGAAGCAGATTTGGGTACCACCCAAGTAT
TGACTCACCCATCAACAACCGCTATGTATTTCGTACATTACTGCCAGCCA
CCATGAATATTGTACGGTACCATAAATACTTGACCACCTGTAGTACATAA
AAACCCAATCCACATCAAAACCCTCCCCCCATGCTTACAAGCAAGTACAG
CAATCAACCTTCAACTATCACACA
- Rick Kittles, Ph.D.
- Section of Genetic Medicine
2Genomes
- Human
- 20,000 genes
- 3.0 billion nucleotides
- 10 million SNPs
- 23 pairs of chromosomes
- Dog
- 19,000 genes
- 2.4 billion nucleotides
- Currently 600,000 SNPs
- 39 pairs of chromosomes
3(No Transcript)
4(No Transcript)
5(No Transcript)
6Race as metapopulation is a complex composite
variable
- Race as generally understood and as used in
biomedical research refers to both cultural and
biological features of metapopulation groups. - Race is composed of
- Ethnic heritage social component
- Biogeographical ancestry biological component
- Interaction social and biological components
may affect each other in non-additive ways
7Why is Race Problematic?
- Does not explain human biological variation.
- Socio-cultural meanings and the colloquial use of
the term. - Lack of discourse between disciplines.
- When used as a causal genetic variable it
perpetuates biological determinism.
8Is Ancestry better?
- Useful index for human biological variation.
- No historical baggage.
- Increased discourse between disciplines.
- When used as an index for genetic background it
allows for a better investigation of biological
risk factors.
9- Era of Genomic Ancestry and challenges
- related to Health.
- Group definition and membership.
- Can we accurately assess genomic ancestry?
- How does genomic ancestry relate to skin color
and possibly SES? - How useful is genomic ancestry for informing us
about disease risk? - Health Disparities are they due to biological
differences? - How do we prevent repeating the negative past
abuses of race.
10- Promises in the post- genome sequencing era.
- Advances in microarray and DNA chip technology.
- Increased research in diverse populations.
- Integration of epidemiological, cultural, and
population history.
11(No Transcript)
12Divergence of human populations
Mountain et al. 2002
13Genetic variation among the major continents
CD4 haplotype variation Tishkoff
2003 ARGHG
14(No Transcript)
15(No Transcript)
16- Genetic variation is not randomly distributed.
- Frequencies vary with age of allele.
- Old common across many populations
- Recent low frequency and localized
- Population demographic history usually affects
all loci similarly. - Natural selection acts upon specific loci.
17FST across the genome by chromosome position
Akey et al. (2002) Genome Research
18Selection (natural and mate) may have driven
this modular evolution.
- Infectious disease (high pathogen load).
- Immune response MSR1, RnaseL, COX-2
- Optimal phenotype for traditional diet/
lifestyle - Thrifty-genotype hypothesis..
- Fat storage? IGF and LPL related genes
- Salt sensitivity? CYP3A, ICAM gene clusters
- Higher testosterone? CYP3A, AR, SRD5AR, CYP17
- Darker skin color? Vitamin D related genes
19Several important diseases showing differences
among populations likely have some biological
component
- Obesity (gt in African-American women and Native
American, and Hispanic populations) - Type 2 diabetes (gt in Native American, Hispanic,
and African-American populations) - Hypertension (gt in African-American populations)
- End Stage Renal Disease (gt in Native American,
Hispanic, and African-American populations) - Dementia (gt in European-American)
- Osteoporosis (gt in European-American)
- Autoimmune diseases (SLE, PPROM, MS, asthma)
- Particular drug responsiveness or adverse effects
- Skin Pigmentation (varies with latitude across
the world) - Cancer
- Skin cancer (gt in European-American)
- Lung Cancer (gt in African-American)
- Prostate Cancer (gt in African-American)
20Prevalence of Hypertension by Mean Body Mass
Index Among Populations of the African Diaspora
North America
Caribbean
West Africa
Cooper R, Rotimi C. et al. AJPH. 1997
21Alzheimer's Disease and APOE e4 gene
Farrer LA, JAMA 1997
22Alzheimer's Disease and APOE e4 gene
Farrer LA, JAMA 1997
23Differences in genetic risk for CVD across
populations suggest that many genes are
modulated by environmental factors
ALOX5AP gene and risk for myocardial infarction
(Helgadottir et al. 2006) TCF7L2 gene and risk
for diabetes (Helgason et al. 2007)
24Influencing factors
- Gene/gene and gene/environment interactions
- Discrimination and perceived racism (stress
process) - Accumulated stress (weathering, allostatic load,
etc.) - Life course selection
- Cultural factors
- Behavioral differences
- SES and institutional arrangement
25(No Transcript)
26The experience of Discrimination and Black-White
Health Disparities in Medical Care. L.A. Penner
et al. (2009) Journal of Black Psychology 352
pp.180-203.
27Self and Other Race Perceptions
James S. Jackson, Institute of Social Research,
University of Michigan
28Race (social)
Disease
Ancestry (genetic)
29(No Transcript)
30(No Transcript)
31Genetic features of African Americans
- High genetic heterogeneity due to African
ancestry and admixture w/ non-African
populations. - Increased LD due to admixture.
- Pattern of variation differs geographically.
- High levels of population stratification may
confound association studies.
32(No Transcript)
33(No Transcript)
34(No Transcript)
35Black Map
African Americans 2000 U.S. Census
36Hispanic Map
Hispanics 2000 U.S. Census
37MIXED RACE 2000 U.S. Census
38Ancestry Informative Markers (AIMs)
- Genetic markers with large allele frequency
difference (d) between parental populations. - For biallelic markers d ?p1-p2 ?
- Single Nucleotide Polymorphisms (SNPs) or
Deletion/ Insertion Polymorphisms (DIPs).
39d of AIM SNPs for West African/ Europeans
Tian et al. (2006) AJHG
40Biogeographical ancestry
- Usefully accurate individual ancestry estimates
possible using ancestry informative markers - Many populations show variation in individual
ancestry levels - Ancestry estimates can be used to control for
heterogeneity in admixed populations - Conditioning variable for regression models
- Matching cases and controls
- Bayesian Admixture Mapping
41STRUCTURE plot of individual ancestry estimates
using 112 AIMs in 3 populations
Bamileke (Cameroon) European Americans (MD)
African Americans (DC)
(23.2 European ancestry)
42Individual Ancestry Beyond Black and White
African-American sample from Washington, D.C.
(n221) European-American sample from State
College, PA (n193)
43Individual AncestrySan Luis Valley, CO
Hispanics
Bonilla et al. (2004) Annals of Human Genetics
68139-153.
44Individual AncestryPuerto Rican Women from New
York City
Bonilla et al. (2004) Human Genetics 11557-68.
45European genetic contribution in African-American
populations living in different geographical
areas of the US.
Parra et al. AJHG 1998 Parra et al. AJPA 2002
Kittles et al. unpublished
46Genetic Ancestry in Caribbean Populations
Jamaica Closed circles Barbados Open
circles St. Thomas Triangles
47(No Transcript)
48When is AM appropriate?
- There is evidence that the trait is genetic.
- Admixture is recent (lt 20 generations).
- High prevalence difference of the trait in
founding populations.
49Candidate Disease for Admixture Mapping in
African American
Disease Relative Risk (95 CI) Multiple
sclerosis 0.45 (0.35-0.55) Lung
Cancer 1.48 (1.38-1.67) Stroke 1.57
(1.27-1.94) End-stage renal disease 1.87
(1.47-2.39) Prostate cancer 2.73
(2.03-3.86) Hypertensive heart disease 2.80
(2.00-4.93)
-Relative risks are for African Americans versus
European Americans
50Ancestry can be estimated across chromosomal
regions.
Smith et al. AJHG 741001-1013, 2004
Seldin et al. Genome Res. 141076 -1084, 2004
51Admixture mapping for Pca genes
- Disease gene identification may be facilitated
if we know which parts of the genome the cases
and controls have inherited at a disproportionate
rate from one of the parental populations.
Patterson et al. AJHG 74, 2004
52Genetic Ancestry in Ipira, BrazilPaschoalin et
al. 2003. Int Braz J Urol 29(4)300-305
53Genetic Ancestry of AA men from DC
210 AIMs 2-way ANOVA P0.0001
54Race (social)
Disease
Ancestry (genetic)
55(No Transcript)
56(No Transcript)
57(No Transcript)
58Race (social)
Disease
Ancestry (genetic)
59(No Transcript)
60(No Transcript)
61Revolution in genetics
Colon
Prostate
Breast
62(No Transcript)
63Genome-wide association studies for Pca in AAs
- None published to date
- Four published for EAs
- About 20 SNPs identified
64Replication of Pca GWAS SNPs in African Americans
Hooker et al. (In Press) The Prostate
65Genome-wide association studies in AAs
- Why were only 5 of 20 SNPs replicated in AAs?
- Identified SNPs are not causal, but linked to
causal SNPs in EAs. - Rare variants in genes.
- Low power to detect effects.
- Clear need to perform GWAS in AAs.
- Two are currently being performed
66(No Transcript)
67(No Transcript)
68(No Transcript)
69(No Transcript)
70(No Transcript)
71Race (social)
Disease
Ancestry (genetic)
72Studies have suggested that persons who identify
themselves as black may have, on average, a less
active renin-angiotension system and a lower
bioavailability of nitric oxide than those
self-identified as white. p. 2050 Our trial
represents a departure from the recent approach
to the design of cardiovascular trials. Rather
than studying a large heterogeneous population,
we examined a specific population. A
heterogeneous population may have substantial
variations in genetic and environmental factors
that influence disease progression and the
response to therapy. p. 2055
73- BiDil Issues
- Who is Black?
- Race is a crude biological proxy
- Trial did not test other groups.
- Did not determine if BiDil works better in Blacks
than others. - 43 improvement in survival.
- What are the important functional genetic
variants?
74Significance of identifying susceptibility genes
- Improved diagnostics
- better phenotype
- early detection and prevention
- pharmacogenomics
- efficient designer drugs
- avoid complications
75(No Transcript)
76(No Transcript)
77Summary
- African and Hispanic Americans are a socially,
culturally, and genetically heterogeneous
macro-ethnic group with diverse ancestries and
continental U.S. experiences. - Exhibit significant population substructure.
- Need to estimate genetic ancestral background to
fully assess genetic and non-genetic confounders
and predictors of complex diseases.
78Summary
- Equitable benefits from genetic medicine depends
on population genetics, market economics, and
most importantly on historical and cultural
identities. - Inclusion means paying attention to context.
(social, cultural and historical)
79Specifics
- Take genetic ancestry into account in biomedical
study designs (self-identified Race alone may
be a confounder). - Figure out effects (interaction) of Race/ skin
color, racism and poverty on health disparities. - Engage social scientists more to gain better
appreciation and understanding of critical
non-genetic (and genetic) variables which should
also be explored.
80COLLABORATORS
- Howard University
- Chiledum Ahaghotu, M.D.
- Aaron Jackson, M.D.
- Georgia Dunston, Ph.D.
- Medical College of Georgia
- Sally Weinreich, Ph.D.
- St. Thomas, Virgin Islands
- Neil Garbutt, M.D.
- University of Louisville
- La Creis Kidd, Ph.D.
- Penn State University
- Mark Shriver, Ph.D.
- Translational Genomics
- John Carpten, Ph.D.
- Johns Hopkins University
- William Isaacs, Ph.D.
- University of Illinois, Chicago
- Vincent Freeman, M.D.
- Liberia
- Linda Sanvee, M.D.
- Cameroon
- Fru Angwafo III, M.D.
- Nigeria
- Usifo Osime, M.D.
- Clement Adebamowo, M.D.
- Meharry Medical College
- Flora Ukoli, M.B.B.S., MPH
- University of Washington
- Layron Long, M.D.
FUNDING NIGMS, NCRR, NCI, ORMH, and NHGRI of the
National Institutes of Health. Department of
Defense