Assessing Ethnicity from Human Mitochondrial DNA Types Determined by Hybridization with Sequence-Specific Oligonucleotides

Abstract

A logistic regression model was developed to predict ethnic group from mitochondrial DNA (mtDNA) types determined by hybridization with sequence-specific oligonucleotide (SSO) probes of the two hypervariable segments of the mtDNA control region. The model was developed with, and tested against, a previously reported data set of 525 individuals from five ethnic groups (African-American, Southeast Asian, Caucasian, Japanese, and Mexican) involving 23 probes at nine regions within the two hypervariable segments [1]. The model correctly predicted the ethnic group of 65.3% of the overall sample; however, the success rate varied substantially among ethnic groups, with the most success obtained with Caucasians (81% correctly classified). A discriminant analysis yielded similar results. An example is given of using the model to predict the ethnic group of an SSO-type from a forensic case. Such models provide alternatives to traditional skeletal-based methods of predicting ethnicity, especially in cases where skeletal material is absent or incomplete.

Self-reported race/ethnicity in the age of genomic research: its potential impact on understanding health disparities

“…In fact, studies found that mitochondrial DNA can be used to infer the probable assignment of coarse ethnicity with almost 90% accuracy [ 46 ]. This level of accuracy in predicting investigator-assigned ethnicity could be very useful in forensic investigations [ 47 ]. X chromosome (X-DNA testing) markers: an X chromosome DNA test looks at markers on X chromosome(s).…”

Inferring the Most Likely Geographical Origin of mtDNA Sequence Profiles

“…As mentioned previously, logistic regression may be used when there are only two groups (Connor & Stoneking, 1994). We have constructed a procedure based on logistic regression which resembles the PCA‐QDA approach.…”

“…Of the mentioned papers, Connor & Stoneking (1994) and Röhl et al (2001) are the only ones to use data similar to ours. As far as methods go, the approach we propose combines tools from multivariate statistics in a way that differs from those in the above‐mentioned publications.…”

“…Several authors have discussed variations of the problem addressed in the title of this paper. To indicate some related topics we quote a few publications, including some titles: ‘Assessing ethnicity from human mitochondrial … ’ (Connor & Stoneking, 1994), ‘Probable Race of a Stain Donor ’ (Brenner, 1997), ‘Inference of population structure using multilocus genotype data' (Pritchard et al 2000), ‘Inferring ethnic origin by means of an STR profile' (Lowe et al 2001), ‘An annotated mtDNA database', (Röhl et al 2001), ‘Patterns of human diversity, within and among continents, inferred from biallelic DNA polymorphisms' (Romualdi et al 2002), ‘Informativeness of genetic markers for inference of ancestry' (Rosenberg et al 2003). There are also computer programs available, for instance Röhl et al (2001) describe MTRADIUS while Pritchard et al (2000) explain STRUCTURE.…”

“…In the case of only two groups, can be seen as a special case of logistic regression. The use of logistic regression in the two‐class situation has been demonstrated by Connor & Stoneking (1994).…”