Advertisement

Genotypically determined ancestry across an infertile population: ovarian reserve and response parameters are not influenced by continental origin

      Objective

      To evaluate the relationship between genetic ethnicity using ancestry informative markers (AIMs) and ovarian reserve and response parameters as evidenced by FSH, antimüllerian hormone (AMH), basal antral follicle count (BAFC), and total oocyte yield in IVF.

      Design

      Retrospective.

      Setting

      Academic medical center.

      Patients(s)

      A total of 2,508 infertile patients undergoing IVF at a single center.

      Intervention(s)

      Patients were genotyped for 32 AIMs and analyzed for differences in allele frequencies. A validated Bayesian clustering algorithm was then used to assign individuals into one of four ethnic populations: European, African, Central/South Asian, or East Asian.

      Main Outcome Measure(s)

      FSH, AMH, BAFC, and oocyte yield variation.

      Result(s)

      After controlling for age and body mass index, genetic ethnicity had no impact on AMH, BAFC, and oocyte yield. FSH was found to be lower in patients of Central/South Asian ancestry (6.46 ng/mL vs. 6.97 ng/mL); however, the absolute difference is of little clinical significance. Subgroup analyses of 1,327 patients restricted to those with limited genetic admixture as determined by AIMs indicated that FSH, AMH, BAFC, and oocyte yield were equivalent.

      Conclusion(s)

      When determining ethnicity using AIMs, ethnic background does not have an impact on markers of ovarian reserve or ovarian response. Specifically, no differences were found in AMH, BAFC, or oocyte yield relative to genotypic ethnicity. Using AIMs rather than self-reported ethnicity allows for elimination of reporting biases and nonreporting of ethnicity, which can confound data. Based upon these data, specific recommendations for ovarian reserve testing should thus be made based on other factors besides ethnic background.

      Key Words

      Discuss: You can discuss this article with its authors and with other ASRM members at http://fertstertforum.com/olcham-ancestry-ovarian-reserve/
      The influence of ethnicity on assisted reproduction is an area of active research, with some investigators finding meaningful differences in ovarian performance between ethnic groups (
      • Bleil M.E.
      • Gregorich S.E.
      • Adler N.E.
      • Sternfeld B.
      • Rosen M.P.
      • Cedars M.I.
      Race/ethnic disparities in reproductive age: an examination of ovarian reserve estimates across four race/ethnic groups of healthy, regularly cycling women.
      ,
      • Seifer D.B.
      • Golub E.T.
      • Lambert-Messerlian G.
      • Benning L.
      • Anastos K.
      • Watts D.H.
      • et al.
      Variations in serum mullerian inhibiting substance between white, black, and Hispanic women.
      ,
      • Purcell K.
      • Schembri M.
      • Frazier L.M.
      • Rall M.J.
      • Shen S.
      • Croughan M.
      • et al.
      Asian ethnicity is associated with reduced pregnancy outcomes after assisted reproductive technology.
      ) and others failing to detect ethnic differences (
      • Bhide P.
      • Gudi A.
      • Shah A.
      • Homburg R.
      Serum anti-Mullerian hormone levels across different ethnic groups: a cross-sectional study.
      ,
      • Iglesias C.
      • Banker M.
      • Mahajan N.
      • Herrero L.
      • Meseguer M.
      • Garcia-Velasco J.A.
      Ethnicity as a determinant of ovarian reserve: differences in ovarian aging between Spanish and Indian women.
      ,
      • Randolph Jr., J.F.
      • Sowers M.
      • Gold E.B.
      • Mohr B.A.
      • Luborsky J.
      • Santoro N.
      • et al.
      Reproductive hormones in the early menopausal transition: relationship to ethnicity, body size, and menopausal status.
      ). A possible explanation for the conflicting results is the reliance on self-reported ethnicity as a means of labeling and comparing populations as this is associated with significant biases (
      • Mersha T.B.
      • Abebe T.
      Self-reported race/ethnicity in the age of genomic research: its potential impact on understanding health disparities.
      ). Others methods of ethnic assignment exist, and of particular interest are ones that make use of genetic analysis to determine ancestral origin. Thus, ancestry determination using genetic markers is a possible solution as it overcomes some challenges associated with self-identified race.
      Self-reported ethnic data, which are often used in population-based studies, are subject to reporting bias specifically and may not provide a precise representation of an individual's origin as determined genotypically (
      • Sucheston L.E.
      • Bensen J.T.
      • Xu Z.
      • Singh P.K.
      • Preus L.
      • Mohler J.L.
      • et al.
      Genetic ancestry, self-reported race and ethnicity in African Americans and European Americans in the PCaP cohort.
      ). Nonbiological factors such as culture, religion, language, and behavior influence the way in which one reports ethnicity (
      • Mersha T.B.
      • Abebe T.
      Self-reported race/ethnicity in the age of genomic research: its potential impact on understanding health disparities.
      ,
      • Kittles R.A.
      • Weiss K.M.
      Race, ancestry, and genes: implications for defining disease risk.
      ). Furthermore, a study of adolescents attending U.S. schools concluded that 4%–5% of multiracial adolescents shift views of self-identification while transitioning to young adulthood, possibly to owing to life events and influences (
      • Hitlin S.
      • Scott Brown J.
      • Elder Jr., G.H.
      Racial self-categorization in adolescence: multiracial development and social pathways.
      ). Over the past decade, the identification of specific single nucleotide polymorphisms (SNPs) termed “ancestry informative markers” (AIMs) has clarified ethnic classification (
      • Price A.L.
      • Butler J.
      • Patterson N.
      • Capelli C.
      • Pascali V.L.
      • Scarnicci F.
      • et al.
      Discerning the ancestry of European Americans in genetic association studies.
      ,
      • Mao X.
      • Bigham A.W.
      • Mei R.
      • Gutierrez G.
      • Weiss K.M.
      • Brutsaert T.D.
      • et al.
      A genomewide admixture mapping panel for Hispanic/Latino populations.
      ). Population-based studies have demonstrated that certain allelic frequencies are markedly different between ethnicities (
      • Rosenberg N.A.
      • Pritchard J.K.
      • Weber J.L.
      • Cann H.M.
      • Kidd K.K.
      • Zhivotovsky L.A.
      • et al.
      Genetic structure of human populations.
      ). Analysis of the differences can provide admixture information and allow for population-based association testing (
      • Kosoy R.
      • Nassir R.
      • Tian C.
      • White P.A.
      • Butler L.M.
      • Silva G.
      • et al.
      Ancestry informative marker sets for determining continental origin and admixture proportions in common populations in America.
      ). As such, AIMs have been used increasingly in genetic association and population studies to overcome the many drawbacks related to self-reported ethnic data (
      • Halder I.
      • Shriver M.
      • Thomas M.
      • Fernandez J.R.
      • Frudakis T.
      A panel of ancestry informative markers for estimating individual biogeographical ancestry and admixture from four continents: utility and applications.
      ,
      • Nassir R.
      • Kosoy R.
      • Tian C.
      • White P.A.
      • Butler L.M.
      • Silva G.
      • et al.
      An ancestry informative marker set for determining continental origin: validation and extension using human genome diversity panels.
      ). Genetic classification of infertile patients using AIMs may help in clarifying some of the inconsistencies in prior literature with respect to the relationship between ethnicity and ovarian performance.
      Parameters of particular interest that have been studied in relation to self-reported ethnicity are markers of ovarian reserve. Antimüllerian hormone (AMH) has been identified as a sensitive marker of ovarian reserve, but its association with ethnicity has been mixed (
      • Seifer D.B.
      • Maclaughlin D.T.
      Mullerian inhibiting substance is an ovarian growth factor of emerging clinical significance.
      ). In early studies investigating ethnic differences in serum AMH, lower levels were found in black women as compared with in white women (
      • Seifer D.B.
      • Golub E.T.
      • Lambert-Messerlian G.
      • Benning L.
      • Anastos K.
      • Watts D.H.
      • et al.
      Variations in serum mullerian inhibiting substance between white, black, and Hispanic women.
      ). Subsequently, other investigators reported no differences between white and black women but noted that white women had significantly lower AMH levels compared with Asians (
      • Gleicher N.
      • Kim A.
      • Weghofer A.
      • Barad D.H.
      Differences in ovarian aging patterns between races are associated with ovarian genotypes and sub-genotypes of the FMR1 gene.
      ). More recently, investigators have found no independent association of ethnicity with AMH (
      • Bhide P.
      • Gudi A.
      • Shah A.
      • Homburg R.
      Serum anti-Mullerian hormone levels across different ethnic groups: a cross-sectional study.
      ). In all studies, ethnicity data were obtained via patient self-identification. Ethnic differences in day 3 FSH levels and basal antral follicle counts (BAFC) have also been studied with similar contradictions. Some studies show differences among ethnicities after controlling for age (
      • Randolph Jr., J.F.
      • Sowers M.
      • Gold E.B.
      • Mohr B.A.
      • Luborsky J.
      • Santoro N.
      • et al.
      Reproductive hormones in the early menopausal transition: relationship to ethnicity, body size, and menopausal status.
      ), while others show no differences in day 3 FSH levels or BAFC (
      • Iglesias C.
      • Banker M.
      • Mahajan N.
      • Herrero L.
      • Meseguer M.
      • Garcia-Velasco J.A.
      Ethnicity as a determinant of ovarian reserve: differences in ovarian aging between Spanish and Indian women.
      ). In sum, the available literature presents contradictory outcomes, is limited by small sample sizes, and is often restricted to only a few ethnic groups. Interpretation of those data is further confounded by the exclusive use of patient self-reported ethnicity, which does not always agree with genetic ancestry, which is more biologically relevant (
      • Mersha T.B.
      • Abebe T.
      Self-reported race/ethnicity in the age of genomic research: its potential impact on understanding health disparities.
      ,
      • Sucheston L.E.
      • Bensen J.T.
      • Xu Z.
      • Singh P.K.
      • Preus L.
      • Mohler J.L.
      • et al.
      Genetic ancestry, self-reported race and ethnicity in African Americans and European Americans in the PCaP cohort.
      ).
      This study aims to reassess the relationship between genotypically determined ethnicity and markers of ovarian reserve. In this large cohort of patients, AIMs are used to isolate an individual's genetic origin from other societal factors that confound self-reported ethnicity.

      Material and methods

       Patients

      This study identified 2,508 infertile patients seeking care at a single center between 2008 and 2014. Inclusion criteria were patients who underwent their first IVF cycle with embryo aneuploidy screening. Patients also had frozen DNA samples available for analysis through a preexisting human DNA repository. Oocyte donation cycles were excluded.

       Study Design

      A retrospective cohort analysis was performed using data from all patients who met the inclusion criteria. Patient characteristics were analyzed including maternal age, height, and weight. Self-reported ethnicity was retrieved from initial intake evaluation. Available categories were white non-Hispanic/Hispanic, black non-Hispanic/Hispanic, Asian non-Hispanic/Hispanic, other, and not reported. Ovarian reserve markers analyzed included BAFC, day 3 FSH level, and AMH level. Oocyte yield was also included as an additional parameter of reproductive function.

       Patient Treatment Protocol

      Standard regimens for controlled ovarian hyperstimulation were employed using purified urinary FSH or recombinant FSH and LH activity in the form of low-dose hCG or hMG along with GnRH agonist (long down-regulation or microdose flare) or GnRH antagonist to prevent a premature LH surge. Monitoring of IVF cycles were per practice routine. Oocyte maturation was induced with recombinant hCG or purified urinary hCG or with GnRH agonist when two or three follicles reached or exceeded 17–18 mm or when the follicular cohort was deemed to be mature by the patient's primary physician. Transvaginal oocyte aspiration was performed approximately 36 hours later.

       Identification of Ethnic Admixture by Genotyping

      Genomic DNA was extracted from whole blood samples using the QIAmp DNA Blood BioRobot MDx Kit (QIAGEN). Isolated DNA was genotyped using a QuantStudio12K Flex Real-Time PCR System with 32 SNP TaqMan OpenArray Real-Time PCR Plates (Thermo Fisher Scientific). Marker panels using as few as 24 AIMs have been shown to be efficacious in ascertaining the origin of subjects from particular continents. As such, the polymerase chain reaction plates were customized with the maximum allotted assays to identify 32 individual markers from a panel previously validated to be informative of continental origin (
      • Kosoy R.
      • Nassir R.
      • Tian C.
      • White P.A.
      • Butler L.M.
      • Silva G.
      • et al.
      Ancestry informative marker sets for determining continental origin and admixture proportions in common populations in America.
      ). SNP profiles were exported using Applied Biosystems TaqMan Genotyper Software (Thermo Fisher Scientific). Patient continental admixture proportions were determined using the Bayesian clustering algorithms implemented in the program STRUCTURE v2.3, as has been described elsewhere (
      • Falush D.
      • Stephens M.
      • Pritchard J.K.
      Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies.
      ,
      • Pritchard J.K.
      • Stephens M.
      • Donnelly P.
      Inference of population structure using multilocus genotype data.
      ). Analysis was performed without prior population assignment and in multiple runs with parameters as follows: 10,000 burn-in cycles, 50,000 replicates, and admixture model, λ = 1. Patients were assigned to a given population (European, African, Central/South Asian, or East Asian) corresponding to their greatest admixture proportion. To further investigate the effects of genetic admixture, a subgroup analysis was performed identifying those patients with very little admixture. Criteria were met when any individual had more than 85% of her genetic profile cluster to a single ancestry.

       Statistical Analysis

      Ethnic populations were analyzed for differences in demographic parameters using Kruskal-Wallis equality-of-populations rank tests and Wilcoxon rank-sum tests. Differences in markers of ovarian reserve were determined using linear regression controlling for age and body mass index (BMI).
      All data collection and analysis was approved by the Institutional Review Board. Subjects provided written consent for use of these samples in research.

      Results

       Ethnicity

      Genotyping for ancestry was successfully accomplished for all patient samples. Self-reported ethnicity was available for 2,053 (81.9%) patients and agreed with genetically determined ancestry 88.9% of the time. The 455 patients (18.1%) who self-reported as other or chose not to self-report ethnicity were genotyped, and a dominant ancestry was assigned. Population characteristics by ancestry are listed in Table 1. Age and BMI differed among ethnic groups (P=.0001, P=.0001), with the East Asian population being oldest (mean, 36.4 years) and Central/South Asian the youngest (mean, 34.0 years). Patients’ genotypic ancestries were composed of 73% European, 8.9% East Asian, 4.5% African, and 13.6% Central/South Asian.
      Table 1Population characteristics by genotyped ethnicity.
      EthnicityAge (y), mean ± SDBMI (kg/m2), mean ± SDn (%)
      European34.8 ± 4.525.0 ± 5.31,832 (73)
      East Asian36.4 ± 3.923.4 ± 4.5223 (8.9)
      African35.1 ± 5.127.8 ± 5.6113 (4.5)
      Central/South Asian34.0 ± 4.224.7 ± 4.8340 (13.6)

       Ovarian Reserve Markers

      Markers of ovarian reserve were analyzed for associations with age (Fig. 1). As expected, AMH, BAFC, and oocyte yield were found to be inversely correlated with age (P=.001, P=.001, P=.001). Day 3 FSH increased with age (P=.001).
      Figure thumbnail gr1
      Figure 1As expected, age alone is strongly correlated with FSH, AMH, BAFC, and oocyte yield among the study population (P=.001, P=.001, P=.001, P=.001). Consequently, all subsequent analyses were controlled for age.
      AMH, BAFC, and oocyte yield were similar among ethnicities after controlling for age and BMI (Fig. 2). Average day 3 FSH levels were 6.97, 6.95, 6.64, and 6.46 mIU/mL for European, East Asian, African, and Central/South Asian patients, respectively. FSH was found to be lower in patients of Central/South Asian ancestry compared with those of European ancestry (P=.04), with an absolute difference of 0.51 mIU/mL. Levels of all analyzed markers categorized by genotypic ancestry can be found in Table 2.
      Figure thumbnail gr2
      Figure 2Cycle parameters stratified by ethnicity after adjusting for age and BMI. Day 3 FSH levels (A) of Central/South Asian patients were found to be lower than those of European patients, 6.46 mIU/mL versus 6.97 mIU/mL (P=.04). AMH levels (B), BAFC (C), and oocyte yield (D) were not statistically different among ethnicities.
      Table 2Ovarian reserve markers by genotyped ethnicity.
      EthnicityFSH (mIU/mL), mean ± SDAMH (ng/mL), mean ± SDBAFC, mean ± SDOocyte yield, mean ± SD
      European6.97 ± 2.93.95 ± 5.617.0 ± 9.215.5 ± 10.2
      East Asian6.95 ± 2.13.18 ± 2.715.3 ± 8.113.3 ± 7.8
      African6.64 ± 2.03.33 ± 4.816.9 ± 11.316.3 ± 11.4
      Central/South Asian6.46 ± 1.93.53 ± 6.417.4 ± 9.115.3 ± 9.6
      The subgroup analysis revealed that 1,327 patients had limited admixture whereby greater than 85% of their AIMs clustered to a single ancestry. Europeans composed the majority of the subgroup, 80.3%, while 11.6% were East Asian, 4.8% were African, and 3.2% were Central/South Asian. Among the four ethnicities, FSH, AMH, BAFC, and oocyte yield were similar after controlling for differences in age and BMI (P=.16, P=.12, P=.22, P=.26).

      Discussion

      The data presented here show no clinically significant differences in major markers of ovarian reserve or oocyte yield by ancestry. Importantly, ancestry was identified using genotyping as determined by validated AIMs to decrease reporting and nonreporting biases associated with self-identified ethnicity.
      Determining continental origin may more accurately reflect true ancestral reproductive physiology, reducing any societal influence that patients may exhibit when asked to self-identify. In multiracial individuals, self-identification may change throughout adolescence, leading to further discrepancy (
      • Hitlin S.
      • Scott Brown J.
      • Elder Jr., G.H.
      Racial self-categorization in adolescence: multiracial development and social pathways.
      ). When identifying with a race, one often takes into consideration qualities that are not necessarily determined by genotype, such as cultural, psychological, language, religious, and behavioral factors (
      • Kittles R.A.
      • Weiss K.M.
      Race, ancestry, and genes: implications for defining disease risk.
      ). Louwers et al. genotyped a large group of patients diagnosed with polycystic ovarian syndrome (PCOS) and reported that self-identified ethnicity and genetic ethnicity differed about 10% of the time. The disparity between these reporting methods, self versus genetic, also varied depending on the ethnicity analyzed. Furthermore, after genotyping, genetic ancestry was found to be a better predictor for phenotypic variability in patients with PCOS than self-identified ethnicity (
      • Louwers Y.V.
      • Lao O.
      • Fauser B.C.
      • Kayser M.
      • Laven J.S.
      The impact of self-reported ethnicity versus genetic ancestry on phenotypic characteristics of polycystic ovary syndrome (PCOS).
      ). The ability to determine ancestral origin with the use of genotyping may prevent false associations that would be otherwise made when self-reported ethnicity is analyzed. Our study uses this principle to better determine whether an individual's true ancestry predicts differences in markers of ovarian reserve while controlling for age and BMI.
      The use of AIMs, however, does have some limitations. Although populations can be classified using purely genetic and ancestral data relevant to graphical origin, certain cultural or environmental factors that could be linked to variations in ovarian reserve are not taken into account. Additionally, the use of smaller marker sets (<64 markers) has been shown to have a reduced performance for discriminating certain subpopulations such as Europeans and South Asians. Additional markers may not only help discriminate these populations but also may allow for further subcontinental stratification. It is possible that differences may be seen with this increased classification precision.
      Other limitations include the fact that interpretation of total oocyte yield may be difficult given the multiple types of stimulation protocols used. However, it is not our practice standard to choose protocols or surgical treatments based on ethnicity, although it is possible that some pathologies dictating care could cluster in certain ethnicities. Additionally, other factors that may affect ovarian reserve, such as prior ovarian surgery or family history of infertility, were not controlled for. Lastly, the study population consisted of primarily Caucasian patients. The limited number of individuals representing African descent is consistent with national data that demonstrate that minorities in general are less likely to seek infertility treatment (
      • Seifer D.B.
      • Zackula R.
      • Grainger D.A.
      Society for Assisted Reproductive Technology Writing Group R
      Trends of racial disparities in assisted reproductive technology outcomes in black women compared with white women: Society for Assisted Reproductive Technology 1999 and 2000 vs. 2004–2006.
      ).
      After assignment of patients in into their respective genotypic origins and adjusting for age and BMI, final analysis indicated that FSH levels were significantly lower in patients of Central/South Asian origin. It is worthwhile noting, however, the absolute difference of 6.46 versus 6.97 is likely of little clinical significance. Additionally, no differences were found in AMH, BAFC, or oocyte yield among various genotypically classified ethnic groups. The subgroup analysis of patients who met the criteria for having little genetic admixture showed no differences in any of the selected parameters. It should be reinforced that our population consisted of infertile patients undergoing IVF, and as such results should be generalized accordingly.
      Although our study indicates that ovarian reserve profiles are similar between individuals of different ancestry, disparities in IVF outcomes, especially between black and white women, have been reportedly increasing since early 2000. In comparison with white women, black women have been found to have poorer clinical outcomes, potentially due to higher rates of tubal factor and uterine factor. Additionally, black women undergoing treatment for infertility were found to be older, leading to greater diminished ovarian reserve (
      • Seifer D.B.
      • Zackula R.
      • Grainger D.A.
      Society for Assisted Reproductive Technology Writing Group R
      Trends of racial disparities in assisted reproductive technology outcomes in black women compared with white women: Society for Assisted Reproductive Technology 1999 and 2000 vs. 2004–2006.
      ). Of note, the disparities found were based on a review of Society for Assisted Reproductive Technology databases and have inherent limitations due to reporting bias. Ethnicity reporting is also not required and is variable among fertility centers. Lastly, ethnicities in the database are self-reported and thus may not be an accurate representation of an individual's origin as discussed previously.
      Aside from differences in clinical outcomes among various ethnicities, disparate results in markers of ovarian reserve have also been described in the literature (
      • Bleil M.E.
      • Gregorich S.E.
      • Adler N.E.
      • Sternfeld B.
      • Rosen M.P.
      • Cedars M.I.
      Race/ethnic disparities in reproductive age: an examination of ovarian reserve estimates across four race/ethnic groups of healthy, regularly cycling women.
      ,
      • Seifer D.B.
      • Golub E.T.
      • Lambert-Messerlian G.
      • Benning L.
      • Anastos K.
      • Watts D.H.
      • et al.
      Variations in serum mullerian inhibiting substance between white, black, and Hispanic women.
      ). Studies evaluating markers such as AMH and FSH use self-reporting of ethnicity as a means of establishing a person's race. It is possible that outside factors, independent of ancestral origin, contribute to an individual's current ovarian reserve status. Certainly environmental factors that may be ethnically/racially dependent may lead to some of the discrepancies other investigators have discovered. This study stratifies individuals based on ancestral origin alone, and thus this methodology may resolve some of the differences seen when self-identified race is used instead.
      In conclusion, ancestry as determined by genotyping did not reveal differences in terms of ovarian response parameters including oocyte yield. These markers should be interpreted similarly for varying ethnic groups in patients seeking fertility treatment, and specific recommendations for ovarian reserve testing should thus be made based on other factors outside of ethnic background.

      References

        • Bleil M.E.
        • Gregorich S.E.
        • Adler N.E.
        • Sternfeld B.
        • Rosen M.P.
        • Cedars M.I.
        Race/ethnic disparities in reproductive age: an examination of ovarian reserve estimates across four race/ethnic groups of healthy, regularly cycling women.
        Fertil Steri. 2014; 101: 199-207
        • Seifer D.B.
        • Golub E.T.
        • Lambert-Messerlian G.
        • Benning L.
        • Anastos K.
        • Watts D.H.
        • et al.
        Variations in serum mullerian inhibiting substance between white, black, and Hispanic women.
        Fertil Steril. 2009; 92: 1674-1678
        • Purcell K.
        • Schembri M.
        • Frazier L.M.
        • Rall M.J.
        • Shen S.
        • Croughan M.
        • et al.
        Asian ethnicity is associated with reduced pregnancy outcomes after assisted reproductive technology.
        Fertil Steril. 2007; 87: 297-302
        • Bhide P.
        • Gudi A.
        • Shah A.
        • Homburg R.
        Serum anti-Mullerian hormone levels across different ethnic groups: a cross-sectional study.
        Br J Obstet Gynaecol. 2015; 122: 1625-1629
        • Iglesias C.
        • Banker M.
        • Mahajan N.
        • Herrero L.
        • Meseguer M.
        • Garcia-Velasco J.A.
        Ethnicity as a determinant of ovarian reserve: differences in ovarian aging between Spanish and Indian women.
        Fertil Steril. 2014; 102: 244-249
        • Randolph Jr., J.F.
        • Sowers M.
        • Gold E.B.
        • Mohr B.A.
        • Luborsky J.
        • Santoro N.
        • et al.
        Reproductive hormones in the early menopausal transition: relationship to ethnicity, body size, and menopausal status.
        J Clin Endocrinol Metab. 2003; 88: 1516-1522
        • Mersha T.B.
        • Abebe T.
        Self-reported race/ethnicity in the age of genomic research: its potential impact on understanding health disparities.
        Hum Genom. 2015; 9: 1
        • Sucheston L.E.
        • Bensen J.T.
        • Xu Z.
        • Singh P.K.
        • Preus L.
        • Mohler J.L.
        • et al.
        Genetic ancestry, self-reported race and ethnicity in African Americans and European Americans in the PCaP cohort.
        PLoS One. 2012; 7: e30950
        • Kittles R.A.
        • Weiss K.M.
        Race, ancestry, and genes: implications for defining disease risk.
        Ann Rev Genom Hum Genet. 2003; 4: 33-67
        • Hitlin S.
        • Scott Brown J.
        • Elder Jr., G.H.
        Racial self-categorization in adolescence: multiracial development and social pathways.
        Child Dev. 2006; 77: 1298-1308
        • Price A.L.
        • Butler J.
        • Patterson N.
        • Capelli C.
        • Pascali V.L.
        • Scarnicci F.
        • et al.
        Discerning the ancestry of European Americans in genetic association studies.
        PLoS Genet. 2008; 4: e236
        • Mao X.
        • Bigham A.W.
        • Mei R.
        • Gutierrez G.
        • Weiss K.M.
        • Brutsaert T.D.
        • et al.
        A genomewide admixture mapping panel for Hispanic/Latino populations.
        Am J Hum Genet. 2007; 80: 1171-1178
        • Rosenberg N.A.
        • Pritchard J.K.
        • Weber J.L.
        • Cann H.M.
        • Kidd K.K.
        • Zhivotovsky L.A.
        • et al.
        Genetic structure of human populations.
        Science. 2002; 298: 2381-2385
        • Kosoy R.
        • Nassir R.
        • Tian C.
        • White P.A.
        • Butler L.M.
        • Silva G.
        • et al.
        Ancestry informative marker sets for determining continental origin and admixture proportions in common populations in America.
        Hum Mutat. 2009; 30: 69-78
        • Halder I.
        • Shriver M.
        • Thomas M.
        • Fernandez J.R.
        • Frudakis T.
        A panel of ancestry informative markers for estimating individual biogeographical ancestry and admixture from four continents: utility and applications.
        Hum Mutat. 2008; 29: 648-658
        • Nassir R.
        • Kosoy R.
        • Tian C.
        • White P.A.
        • Butler L.M.
        • Silva G.
        • et al.
        An ancestry informative marker set for determining continental origin: validation and extension using human genome diversity panels.
        BMC Genet. 2009; 10: 39
        • Seifer D.B.
        • Maclaughlin D.T.
        Mullerian inhibiting substance is an ovarian growth factor of emerging clinical significance.
        Fertil Steril. 2007; 88: 539-546
        • Gleicher N.
        • Kim A.
        • Weghofer A.
        • Barad D.H.
        Differences in ovarian aging patterns between races are associated with ovarian genotypes and sub-genotypes of the FMR1 gene.
        Reprod Biol Endocrinol. 2012; 10: 77
        • Falush D.
        • Stephens M.
        • Pritchard J.K.
        Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies.
        Genetics. 2003; 164: 1567-1587
        • Pritchard J.K.
        • Stephens M.
        • Donnelly P.
        Inference of population structure using multilocus genotype data.
        Genetics. 2000; 155: 945-959
        • Louwers Y.V.
        • Lao O.
        • Fauser B.C.
        • Kayser M.
        • Laven J.S.
        The impact of self-reported ethnicity versus genetic ancestry on phenotypic characteristics of polycystic ovary syndrome (PCOS).
        J Clin Endocrinol Metab. 2014; 99: E2107-E2116
        • Seifer D.B.
        • Zackula R.
        • Grainger D.A.
        • Society for Assisted Reproductive Technology Writing Group R
        Trends of racial disparities in assisted reproductive technology outcomes in black women compared with white women: Society for Assisted Reproductive Technology 1999 and 2000 vs. 2004–2006.
        Fertil Steril. 2010; 93: 626-635