REFERENCE: Budowle B, Shea B, Niezgoda S, Chakraborty R.CODIS STR loci data from 41 sample populations. J Forensic Sci2001:46;(3):453–489.

ABSTRACT: Allele distributions for 12 or 13 CODIS coretetrameric short tandem repeat (STR) loci CSF1PO, D3S1358,D5S818, D7S820, D8S1179, D13S317, D16S539, D18S51,D21S11, FGA, TH01, TPOX, and vWA were determined in 41 pop-ulation data sets. The major population groups comprise AfricanAmericans, U.S. Caucasians, Hispanics, Far East Asians, and Na-tive Americans. There was little evidence for departures fromHardy-Weinberg expectations (HWE) in any of the populations.The FST estimates over all thirteen STR loci are 0.0006 for AfricanAmericans, �0.0005 for Caucasians, 0.0021 for Hispanics, 0.0039for Asians, and 0.0282 for Native Americans.

KEYWORDS: forensic science, African American, Caucasian,Hispanic, Asian, Native American, population databases, STR,Hardy-Weinberg Expectations, PCR, FST

DNA typing of biological samples provides the ability to elimi-nate individuals who have been falsely associated with forensic ev-idence and to reduce the number of potential contributors to a fewor only one individual. Polymorphic short tandem repeat (STR)loci are the most informative PCR-based genetic markers for at-tempting to individualize biological material (1–4). To exploit thefull potential of STR loci, the FBI Laboratory sponsored a commu-nity-wide scientific effort to select and establish the core STR locifor the national DNA index, CODIS (i.e., Combined DNA IndexSystem). The resultant thirteen CODIS STR loci are CSF1PO,D3S1358, D5S818, D7S820, D8S1179, D13S317, D16S539,D18S51, D21S11, FGA, TH01, TPOX, and vWA (5). One of theoutcomes of the forensic community effort is the generation of 41population data sets that provide a solid foundation for profile fre-quency estimates. This paper presents STR allele distribution dataon 12 or 13 of the CODIS core STR loci in several sampled popu-lations from each of the following major population groups:African Americans, U.S. Caucasians, Hispanics, Far East Asians,and Native Americans.

Materials and Methods

Samples

A total of 20 laboratories generated population data. The Cali-fornia Department of Justice, Orange County Sheriff’s-CoronerLaboratory, and Oregon State Police worked collectively in thestudy and generated the population data on the California-deriveddatabases and the Chinese, Japanese 1, and Korean databases. TheAlabama Department of Forensic Sciences generated the Alabama-derived population data. The Florida Department of Law Enforce-ment, Metro-Dade Police Department/Miami Children’s ResearchInstitute, and Palm Beach County Sheriff’s Office worked collec-tively and generated the Florida-derived population data. The Na-tional Institute of Standards and Technology, North Carolina StateBureau of Investigation, and Virginia Division of Forensic Scienceworked collectively and generated the Virginia-derived populationdata and the general Asian population data. The Armed Forces In-stitute of Pathology, New York City Office of the Chief MedicalExaminer, and Suffolk County Crime Laboratory worked collec-tively and generated the New York-derived population data and theJapanese 2-derived population data. The Illinois State Police,Michigan State Police, and Detroit Police Department worked col-lectively and generated the Michigan and Illinois-derived popula-tion data. The Illinois State Police generated the Vietnamese-de-rived population data. The Minnesota Bureau of CriminalApprehension generated the Minnesota population data. The RoyalCanadian Mounted Police generated the Canadian, Northern On-tario, Salishan, and Saskatchewan population data. The ArizonaDepartment of Public Safety generated the Arizona Hispanic,Apache, and Navajo population data. The FBI Laboratory gener-ated the FBI-labeled population data [which originate from sam-ples from Texas and have been published previously (4)]. The sam-ples from the Bahamas (kindly provided by Chief SuperintendentJames W. Carey, Forensic Science Section, Royal Bahamas PoliceForce), Jamaica (kindly provided by Dr. Yvonne Cruickshank, Di-rector of the Forensic Science Laboratory in Jamaica), andTrinidad (kindly provided by Ms. Yolanda Thompson, Director ofthe Forensic Science Laboratory in Trinidad and Tobago) were an-alyzed by the FBI Laboratory. The Mexican sample, consisting ofcosmopolitan Mestizo individuals from the state of Neuvo Leon,Mexico, was kindly provided by Ricardo Cerda-Flores of IMSS,Neuvo Leon, and typed at the FBI Laboratory. Further details re-garding the source and preparation of the samples can be obtainedby contacting the contributing laboratories.

STR Amplification and Typing: The DNA samples were ampli-fied using kits from either PE Biosystems (Foster City, CA) or the

453

Bruce Budowle,1 Ph.D.; Brendan Shea,2 M.S.; Stephen Niezgoda,2 M.B.A.; andRanajit Chakraborty,3 Ph.D.

CODIS STR Loci Data from 41 SamplePopulations*

1 FBI Laboratory Division, FBI Academy, Quantico, VA.2 FBI Laboratory Division, Scientific Analysis Section, J. Edgar Hoover

Building, Pennsylvania Avenue NW, Washington, DC.3 Human Genetics Center, University of Texas School of Public Health, P.O.

Box 20334, Houston, TX.* This is publication number 00-09 of the Laboratory Division of the Federal

Bureau of Investigation. Names of commercial manufacturers are provided foridentification only, and inclusion does not imply endorsement by the FederalBureau of Investigation.

Received 24 April 2000; and in revised form 1 June 2000; accepted 13 June2000.

Copyright © 2001 by ASTM International

454 JOURNAL OF FORENSIC SCIENCES

TABLE 1—Observed allele frequency distributions for 11 African American sample populations.a

BUDOWLE ET AL. • CODIS STR POPULATION DATA 455

TABLE 1—Continued.

continues

456 JOURNAL OF FORENSIC SCIENCES

TABLE 1—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 457

TABLE 1—Continued.

continues

458 JOURNAL OF FORENSIC SCIENCES

TABLE 1—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 459

TABLE 1—Continued.

Promega Corporation (Madison, WI) and following the manufac-turers’ recommendations. The amplified products were analyzedusing either an ABI Prism™ 310 Genetic Analyzer, ABI Prism®

373 or 377 DNA Sequencer (PE Biosystems, Foster City, CA), oran FMBIO II (MiraiBio/Hitachi Genetic Systems, Alameda, CA)according to the manufacturer’s recommended protocol. Detailsabout the analytical process can be obtained by contacting the con-tributing laboratories.

Statistical Analysis: Allele designations were determined bycomparison of the sample fragments with those of the allelic lad-ders. The frequency of each allele for each locus was calculatedfrom the numbers of each genotype in the sample set (i.e., the genecount method). Unbiased estimates of expected heterozygositywere computed as described by Edwards et al. (2). Possible diver-gence from Hardy-Weinberg expectations (HWE) was tested bycalculating the unbiased estimate of the expected homozygote/het-erozygote frequencies (6–9) and the exact test (10), based on 2000shuffling experiments. The computer program to perform thesetests was developed by R. Chakraborty (University of TexasSchool of Public Health, Houston, Texas). The values for FST weredetermined as described by Weir and Cockerham (11) using theTFPGA program kindly provided by M. Miller (Northern ArizonaUniversity at Flagstaff).

Results and Discussion

The distributions of observed allele frequencies for the STR lociin the 41 data sets are shown in Tables 1–5. The observed and ex-pected homozygosities and exact test results for departures fromHWE are also provided.

All loci are highly polymorphic in all sample populations. Mostalleles (excluding rare variants) are observed in all populations.The sampled populations of African Americans show the highestwithin-population variation and the Native Americans have thelowest within-population variation. While there were some initialindications of departure from HWE, none were significant aftercorrecting for sampling (i.e., Bonferroni correction (12)). The ma-jority of the initially found departures from HWE are due to geno-

types consisting of rare alleles (e.g., those occurring below fivecounts in the data). These alleles are of virtually no consequence,when such rare allele frequencies are replaced by a minimumthreshold allele frequency applicable to the locus/population com-bination (13,14). To determine whether or not the uncorrected de-partures from HWE may have an impact on profile frequency esti-mates, one might suggest that comparisons of allele frequenciesbetween subgroups be assessed by standard contingency table anal-ysis. However, such an analysis is uninformative. For even moder-ately large sample sizes, standard contingency table analysis ex-hibits extreme sensitivity to small perturbations, i.e., it frequentlyrejects the null hypothesis of no difference even if the difference isof little consequence (15). Traditional population genetic ap-proaches that describe the amount of heterogeneity among popula-tions are much more informative than are significance tests.

Therefore, as recommended by the National Research Council(NRC) II Report (14), Wright’s FST estimates (16,17) were calcu-lated (Table 6). The FST estimates over all thirteen STR loci are0.0006 for African Americans, �0.0005 for Caucasians, 0.0021 forHispanics, 0.0039 for Asians, and 0.0282 for Native Americans.The data support that the degree of relatedness in the AfricanAmerican, Caucasian, Hispanic, and Asian groups is low. In fact,the Hispanic value is likely inflated because southeastern andsouthwestern Hispanic data sets were pooled for the FST calcula-tion. Typically, these two groups are not merged for profile fre-quency estimates (although these data suggest that pooling is pos-sible). Budowle, et al. (18) reported an FST estimate ofapproximately 0.0090 for nine of the thirteen core STR loci inChamorros and Filipinos. For Native Americans (in the currentstudy) a larger estimate of 0.0282 was observed. This FST value isconsistent with expectations for more isolated groups, such as Na-tive Americans. Further, the FST values estimated from these datasupport the generalized recommendations of the National ResearchCouncil for taking a value of 0.01 as a conservative threshold formost population groups and the value of 0.03 for Native Americansas reasonable. Further studies are underway for GST estimates us-ing additional measures of within-population variation based onvariance in allele size.

460 JOURNAL OF FORENSIC SCIENCES

TABLE 2—Observed allele frequency distributions for nine U.S. Caucasians sample populations.a

BUDOWLE ET AL. • CODIS STR POPULATION DATA 461

TABLE 2—Continued.

continues

462 JOURNAL OF FORENSIC SCIENCES

TABLE 2—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 463

TABLE 2—Continued.

continues

464 JOURNAL OF FORENSIC SCIENCES

TABLE 2—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 465

TABLE 2—Continued.

466 JOURNAL OF FORENSIC SCIENCES

TABLE 3—Observed allele frequency distributions for eight Hispanic sample populations.a

BUDOWLE ET AL. • CODIS STR POPULATION DATA 467

continues

TABLE 3—Continued.

468 JOURNAL OF FORENSIC SCIENCES

TABLE 3—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 469

TABLE 3—Continued.

continues

470 JOURNAL OF FORENSIC SCIENCES

TABLE 3—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 471

TABLE 3—Continued.

continues

472 JOURNAL OF FORENSIC SCIENCES

TABLE 3—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 473

TABLE 4—Observed allele frequency distributions in six Asian sample populations.

continues

474 JOURNAL OF FORENSIC SCIENCES

TABLE 4—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 475

TABLE 4—Continued.

continues

476 JOURNAL OF FORENSIC SCIENCES

TABLE 4—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 477

TABLE 4—Continued.

continues

478 JOURNAL OF FORENSIC SCIENCES

TABLE 4—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 479

TABLE 4—Continued.

continues

480 JOURNAL OF FORENSIC SCIENCES

TABLE 4—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 481

TABLE 5—Observed allele frequencies for seven Native American populations.

continues

482 JOURNAL OF FORENSIC SCIENCES

TABLE 5—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 483

continues

TABLE 5—Continued.

484 JOURNAL OF FORENSIC SCIENCES

TABLE 5—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 485

TABLE 5—Continued.

continues

486 JOURNAL OF FORENSIC SCIENCES

TABLE 5—Continued.

BUDOWLE ET AL. • CODIS STR POPULATION DATA 487

TABLE 5—Continued.

continues

488 JOURNAL OF FORENSIC SCIENCES

TABLE 5—Continued.

TABLE 6—FST values for the thirteen CODIS core STR loci.

African NativeLocus American Caucasian Hispanic Asian American

CSF1PO �0.0009 �0.0007 �0.0003 �0.0012 0.0244D3S1358 �0.0005 �0.0009 0.0014 0.0035 0.0764D5S818 0.0010 �0.0001 0.0010 0.0028 0.0656D7S820 0.0000 �0.0005 0.0010 0.0039 0.0201D8S1179 �0.0001 0.0000 0.0005 0.0025 0.0125D13S317 0.0029 �0.0008 0.0047 0.0071 0.0157D16S539 �0.0013 �0.0005 0.0067 0.0017 0.0132D18S51 0.0012 0.0001 0.0011 0.0046 0.0268D21S11 0.0005 0.0008 0.0013 0.0056 0.0371FGA 0.0004 �0.0004 0.0008 0.0029 0.0168TH01 0.0015 �0.0012 0.0041 0.0058 0.0356TPOX 0.0021 �0.0015 0.0024 0.0100 0.0164vWA 0.0011 �0.0011 0.0029 0.0027 0.0172

FST over 0.0006 �0.0005 0.0021 0.0039 0.0282all loci

data. Without their efforts, these population data would not beavailable to the scientific community and CODIS.

Part of the data analysis was supported by the U.S. Public HealthService Research grant GM 41399 from the U.S. National Insti-tutes of Health.

References1. Edwards A, Civitello A, Hammond HA, Caskey CT. DNA typing and

genetic mapping with trimeric and tetrameric tandem repeats. Amer JHum Genet 1991;49:746–56.

2. Edwards A, Hammond HA, Jin L, Caskey CT, Chakraborty R. Geneticvariation at five trimeric and tetrameric repeat loci in four human popu-lation groups. Genomics 1992;12:241–53.

3. Lins AM, Micka KA, Sprecher CJ, Taylor JA, Bacher JW, Rabbach DR,Bever RA, Creacy SD, Schumm JW. Development and population studyof an eight-locus short tandem repeat (STR) multiplex system. J Foren-sic Sci 1998;43:1168–80.

4. Budowle B, Moretti TR, Baumstark AL, Defenbaugh DA, Keys KM.Population data on the thirteen CODIS core short tandem repeat loci inAfrican Americans, U.S. Caucasians, Hispanics, Bahamians, Jamaicans,and Trinidadians. J Forensic Sci 1999;44:1277–86.

5. Budowle B, Moretti TR, Niezgoda SJ, Brown BL. CODIS and PCR-based short tandem repeat loci: Law enforcement tools. In: Second Eu-ropean Symposium on Human Identification 1998, Promega Corpora-tion, Madison, WI, 1998;73–88.

6. Chakraborty R, Smouse PE, Neel JV. Population amalgamation and ge-netic variation: observations on artificially agglomerated tribal popula-tions of Central and South America. Amer J Hum Genet 1988;43:709–25.

7. Chakraborty R, Fornage M, Guegue R, Boerwinkle E. Population genet-ics of hypervariable loci: analysis of PCR based VNTR polymorphismwithin a population. In: Burke T, Dolf G, Jeffreys AJ, Wolff R, editors.DNA fingerprinting: approaches and applications, Berlin: BirkhauserVerlag, 1991;127–43.

8. Nei M, Roychoudhury AK. Sampling variances of heterozygosity andgenetic distance. Genetics 1974;76:379–90.

9. Nei M. Estimation of average heterozygosity and genetic distance froma small number of individuals. Genetics 1978;89:583–90.

10. Guo SW, Thompson EA. Performing the exact test of Hardy-Weinbergproportion for multiple alleles. Biometrics 1992;48:361–72.

11. Weir BS, Cockerham CC. Estimating F-statistics for the analysis of pop-ulation structure. Evolution 1984;38:1358–70.

In conclusion, substantial population data are available for thethirteen CODIS STR loci CSF1PO, D3S1358, D5S818, D7S820,D8S1179, D13S317, D16S539, D18S51, D21S11, FGA, TH01,TPOX, and vWA. These data should enable estimation of multipleloci profile frequencies for most scenarios in the United States. Thedata support that the recommended FST value of 0.01 for correctingfor population substructure is conservative for most populationsand that 0.03 should be used for profile frequency calculations forNative Americans (14).

Acknowledgments

We would like to express our deepest gratitude to all those indi-viduals and institutions that contributed their time, resources, and

BUDOWLE ET AL. • CODIS STR POPULATION DATA 489

12. Weir BS. Multiple tests. In: Genetic Data Analysis, Sunderland, MA:Sinauer Associates, Inc., 1990;109–10.

13. Budowle B, Monson KL, Chakraborty R. Estimating minimum allelefrequencies for DNA profile frequency estimates for PCR-based loci. IntJ Leg Med 1996;108:173–6.

14. National Research Council II Report. The evaluation of forensic evi-dence. National Academy Press, Washington, D.C., 1996.

15. Rudas T, Clogg CC, Lindsey BG. A new index of fit based on mixturemethods for the analysis of contingency tables. J Roy Stat Soc Series B1994;56:623–39.

16. Wright S. Coefficients of inbreeding and relationship. Amer Nat (Chem-ical Abstracts) 1922;56:330–8.

17. Wright S. The interpretation of population structure by F-statistics withspecial regard to systems of mating. Evolution 1965;19:395–420.

18. Budowle B, Defenbaugh DA, Keys KM. Genetic variation at nine shorttandem repeat loci in Chamorros and Filipinos from Guam. Leg Med2000:2;26–30.

Additional information and reprint requests:Bruce BudowleForensic ScienceFBI AcademyQuantico, VA 22135