ancestry informative markers: inference of ancestry in aged bone samples using an autosomal...

Forensic Science International: Genetics 16 (2015) 58–63

Short Communication

Ancestry informative markers: Inference of ancestry in aged bonesamples using an autosomal AIM-Indel multiplex

Carola Romanini a, Magdalena Romero a, Mercedes Salado Puerto a, Laura Catelli a,Christopher Phillips b, Rui Pereira c, Leonor Gusmao c,d, Carlos Vullo a,*a Forensic DNA Laboratory, Argentinean Forensic Anthropology Team (EAAF) Independencia 644,3A, 5000 Cordoba, Argentinab Forensic Genetics Unit, Institute of Legal Medicine, Faculty of Medicine, University of Santiago de Compostela, ES-15705 Santiago de Compostela, Galicia,

Spainc Institute of Molecular Pathology and Immunology of the University of Porto (IPATIMUP), Porto, Portugald DNA Diagnostic Laboratory (LDD), State University of Rio de Janeiro (UERJ), Rio de Janeiro, Brazil

A R T I C L E I N F O

Article history:

Received 29 May 2014

Received in revised form 25 October 2014

Accepted 28 November 2014

Keywords:

Indels

Ancestry informative markers (AIMs)

Genetic ancestry

HGDP-CEPH

Skeletal remains

A B S T R A C T

Ancestry informative markers (AIMs) can be useful to infer ancestry proportions of the donors of forensic

evidence. The probability of success typing degraded samples, such as human skeletal remains, is

strongly influenced by the DNA fragment lengths that can be amplified and the presence of PCR

inhibitors. Several AIM panels are available amongst the many forensic marker sets developed for

genotyping degraded DNA. Using a 46 AIM Insertion Deletion (Indel) multiplex, we analyzed human

skeletal remains of post mortem time ranging from 35 to 60 years from four different continents (Sub-

Saharan Africa, South and Central America, East Asia and Europe) to ascertain the genetic ancestry

components. Samples belonging to non-admixed individuals could be assigned to their corresponding

continental group. For the remaining samples with admixed ancestry, it was possible to estimate the

proportion of co-ancestry components from the four reference population groups. The 46 AIM Indel set

was informative enough to efficiently estimate the proportion of ancestry even in samples yielding

partial profiles, a frequent occurrence when analyzing inhibited and/or degraded DNA extracts.

� 2014 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at ScienceDirect

Forensic Science International: Genetics

jou r nal h o mep ag e: w ww .e lsev ier . co m / loc ate / fs ig

1. Introduction

The ancestry of a forensic evidence contributor, based onancestry informative marker (AIM) analysis, can provide valuableinformation [1–7]. Although lineage markers from mitochondrialDNA (mtDNA) and Y chromosome give information on the maternaland paternal ancestries, respectively, they are haploid non-recombinant genetic markers that represent just a very smallproportion of the genome. Therefore, the ancestry of admixedor non-admixed individuals is more comprehensively estimatedwhen a selection of recombining autosomal genetic markers isused. It has been shown that ancestry informative single nucleotidepolymorphisms (AIM-SNPs) are able to distinguish the majorcontinental population groups relatively efficiently [5,7–10]. Morerecently, AIM-Indels [11,12] that share many of the characteristicsseen in SNPs [2] were also found to reliably distinguish ancestry

* Corresponding author. Tel.: +54 351 4240434; fax: +54 351 4240418.

E-mail address: cvullo@fibercorp.com.ar (C. Vullo).

http://dx.doi.org/10.1016/j.fsigen.2014.11.025

1872-4973/� 2014 Elsevier Ireland Ltd. All rights reserved.

amongst the main continental populations of Africa, Europe, EastAsia and America. In addition, Indels bring increased capability todetect mixed-source DNA, since they produce much more balancedpeak patterns (within locus, the two alleles for a heterozygote showsimilar peak heights) than those of single base extension SNP tests.Of equal relevance to forensic analyses is the fact that SNP and Indelmultiplexes used for identification purposes [13,14] or ancestryanalysis [2–4,12,15] have been developed specifically to analyzevery short amplicons to enable increased sensitivity in thegenotyping of highly degraded biological material.

When dealing with challenging samples such as human skeletalremains, DNA degradation and/or the presence of PCR inhibitorsfrom the soil may constrain the correct typing of the sample[16,17]. Typing mini-STRs or binary markers like SNPs or Indelsthrough the PCR of short amplicon fragments (less than 200 bp)can overcome this problem in degraded DNA extracts [18,19]. Theability of the 46 AIM-Indel multiplex developed by Pereira et al.[12] to obtain reliable ancestry information, in combination withthe relatively short fragment amplification strategy, can help toaddress the challenge of achieving useful ancestry information

C. Romanini et al. / Forensic Science International: Genetics 16 (2015) 58–63 59

from DNA extracts of extremely degraded samples. In order toevaluate the efficiency of the 46plex AIM-Indel set to ascertainancestry on bones with variable degrees of DNA degradation, in thepresent study, DNA extracts from aged bones from routinecasework were analyzed, from four different continental regions:Sub-Saharan Africa, Native America, Europe and East Asia. Thisreport presents the results obtained when applying the AIM-Indelgenotyping protocol to this skeletal material, as part of theidentification process undertaken.

2. Materials and methods

2.1. DNA samples from aged bones

Bone samples included in this study were obtained from routinework with human skeletal remains for human identificationpurposes in different countries, as presented in Table 1. A total of20 skeletal samples from different geographical locations wereanalyzed. Three remains belonged to victims of the Apartheidsystem in South Africa (1960–1994) of sub-Saharan African origin(AFR). The South African National Prosecuting Authority (NPA) setup the Missing Persons Task Team (MPTT) to identify victims whodisappeared in the political oppression of that period and continuetracing their fate. The EAAF DNA laboratory has now beenintegrated into this identification project. Two East Asian originremains (EAS) belong to victims of the Vietnam War (1959–1975).According to ante mortem information the victims were buried in1968. In October 2012, the EAAF exhumed twelve remains thatwere submitted to anthropological and genetic analysis foridentification purposes. Four East Timor remains (EAS) belong tothe period of the Indonesian invasion of East Timor (1975–1999).Nine remains were collected from different Latin Americancountries to represent Native American origin (NAM): threebelong to victims of the Argentinean military dictatorships(1976–1983), two are from victims of the Bolivian dictatorships(1964–1982) and four from the El Salvador War (1980–1992). Two

Table 1Ancestry analysis data from STRUCTURE (cluster membership proportions) and Snippe

Bone sample Geographical

origin

STRUCTURE cluster proportions

AFR EUR EAS

BS01 Bolivia 0.010 0.017 0.022

BS02 Bolivia 0.035 0.338 0.007

BS03 East Timor 0.003 0.003 0.990

BS04 East Timor 0.273 0.124 0.593

BS05 East Timor 0.005 0.021 0.967

BS06 East Timor 0.053 0.421 0.503

BS07 El Salvador 0.012 0.305 0.010

BS08 El Salvador 0.026 0.564 0.048

BS09 El Salvador 0.275 0.716 0.004

BS10 El Salvador 0.025 0.579 0.009

BS11 South Africa 0.963 0.004 0.020

BS12 South Africa 0.980 0.006 0.006

BS13 South Africa 0.993 0.002 0.003

BS14 Argentina 0.174 0.158 0.005

BS15 Argentina 0.007 0.279 0.025

BS16 Argentina 0.093 0.668 0.038

BS17 Argentina

(Welsh)

0.006 0.963 0.010

BS18 Argentina

(Austrian)

0.002 0.918 0.022

BS19 Vietnam 0.009 0.007 0.978

BS20 Vietnam 0.003 0.019 0.968

NAM, Native America; EAS, East Asia; EUR, Europe; AFR, Africa. BS17 and BS18 from A

samples from European individuals (EUR) were also analyzed;these two European origin samples were collected and identified inArgentina within the framework of the Latin American Initiativefor the Identification of the ‘‘Disappeared’’ (LIID) conducted byEAAF [20] : both identified remains belong to a Welsh and anAustrian individual who disappeared in Argentina.

2.2. DNA extraction and AIM-Indel genotyping

The bone samples analyzed were different anatomical elementsof femur, tooth or temporal bones. The post-mortem time rangedfrom 35 to 60 years, according to the anthropological and historicalinformation obtained.

DNA extraction and quantification were performed according toRomanini et al. [19]. All DNA protocols were carried out inlaboratories especially designed for DNA extraction from agedbone samples with facilities to minimize risk of contamination,following ISFG recommendations [21]. Plastic-ware used wasDNA-free, autoclaved and UV irradiated as an additional precau-tion. Bone fragments were cleaned and sanded to remove outerand inner layers and then 0.5 cm disks were cut with a rotary toolin a Biosafe cabinet. All the following steps were made in sterilelaminar flow cabinets. Reagent blanks accompanying the extrac-tion procedure were processed and checked for contamination.After cleaning and sanding, samples were washed (2�) with steriledistilled water and decontaminated 2 min with 10% bleach, thenwashed (4�) with sterile distilled water, two times in 95% alcoholand left to dry. Bone pieces were then ground with a Spex6750 cryogenic grinder freezer-mill (Spex Centriprep Inc.). Twograms of bone powder was decalcified with 20 ml EDTA 0.5 M pH8 for 2 days at 4 8C with continuous rotation. Decalcifiedbone tissue was centrifuged and the supernatant discarded.Decalcified pellets and reagent blanks were then extracted usingthe QIAamp DNA Blood Maxi Kit (Qiagen) according to manufac-turer’s recommendations. The eluted DNA was concentrated toapproximately 100 ml using Vivacon 2-100 MWCO columns

r Bayes analysis likelihood ratios.

Snipper analysis (lowest likelihood ratio)

NAM

0.951 20,777 times more likely American

than East Asian

0.620 148 times more likely American

than European

0.004 65,842,476,070 times more likely East Asian than American

0.010 45,951 times more likely East Asian than European

0.007 93,056,370 times more likely East Asian than American

0.023 5.5 times more likely East Asian than European

0.673 17,413 times more likely American than European

0.362 44 times more likely European than American

0.005 6,446,150,707 times more likely European than African

0.387 6497 times more likely European than American

0.013 150,334,685,781,751,083,565,056 times more likely African

than East Asian

0.008 163,899,108,611,827,657,867,264 times more likely African

than European

0.003 19,202,072,389,817,456,943,513,703,809,024 times more

likely African than E Asian

0.663 231,280 times more likely American than European

0.689 17,489 times more likely American than European

0.201 524,535 times more likely European than American

0.021 5,238,127,897 times more likely European than East Asian

0.058 469,647,983 times more likely European than East Asian

0.006 59,864,948,269 times more likely East Asian than American

0.009 4172 times more likely East Asian than European

rgentina have known European origin, as indicated.

C. Romanini et al. / Forensic Science International: Genetics 16 (2015) 58–6360

(Sartorius). All DNA extracts were quantified using the ABQuantifilerTM Human DNA Quantification kit with the 7500 real-time PCR system according to the manufacturer’s protocol. Basedon the quantification results, inhibited samples were dilutedallowing comparable target DNA conditions for all the samplesprior to PCR amplification.

Indel genotyping of the 46 autosomal loci was made in a singlePCR multiplex reaction following the original protocol describedby Pereira et al. [12]. All PCR reactions included a negative controlas well as 9947A control DNA. A minimum of two amplificationswere performed on the bone extracts. An allele was reported only ifit was called in replicate analysis of the same extract to provideconsensus profiles. There was no evidence of dropin in any reagentblanks or negative PCR controls.

Capillary electrophoresis was performed in an ABI Prism3130 Genetic Analyzer (Life Technologies, CA, USA) usingGeneMapper 3.2 software.

2.3. Statistical analysis

In order to estimate the ancestry of the 20 bone samples,STRUCTURE v2.3.3 [22,23] analyses were carried out with aburnin length of 100,000 followed by 100,000 MCMC steps.HGDP-CEPH diversity panel genetic data [24,25] from African,European, East Asian and Native American population groupswere used as reference samples, alongside the tested boneprofiles. No prior information on the origin of the samples wasprovided in the parameter settings. The Admixture model, withcorrelated allele frequencies and advanced option to update allelefrequencies using only ancestral samples (with POPFLAG = 1),were used for K = 4 genetic clusters, following previous discus-sions on the optimum number of clusters for this panel andgenotype dataset [12].

As a supplementary study to the STRUCTURE analyses of the46 Indels, naıve Bayes ancestry inferences plus principal compo-nent analysis (PCA) were made by uploading the same referencegenotypes for African, European, East Asian and Native Americanpopulations (genotypes available as supplementary data in Pereiraet al. [12]) along with the bone Indel profiles, to the multiple-profile module of the Snipper online forensic classifier: (http://mathgene.usc.es/snipper/analysismultipleprofiles.html). Custom-ized PCA plots were obtained in R using adaptations of the basescript implemented in Snipper (scripts developed by Carla Santos,USC, available on request).

3. Results

3.1. Inference of genetic ancestry

Using prior information provided by the anthropologists(exhumation place and country, historical investigations, identifi-cation results), bone samples were selected as representative ofcollections from different geographic locations, comprising Sub-Saharan Africa, South or Central America, Europe and East Asia.

In accordance with the prior data and the population history ofeach country, a high membership proportion in a single populationgroup was expected for some bone samples, and variableadmixture patterns in others, namely those from South andCentral America.

The three samples from South Africa showed high proportionof membership in the African reference group (BS11, BS12, andBS13). Interestingly, the allele 3 for marker MID360 described byPereira et al. [12] and only detected in Africans in their study wasfound in samples BS11 and BS13. Samples originally from

Vietnam (BS19, BS20) and East Timor (BS03, BS05) exhibitedthe highest proportion of East Asian origin, although samplesBS04 and BS06 from East Timor showed admixture withEuropeans and Africans, in accordance with the colonialhistory of this country, when Europeans dominated the islandfor centuries.

The highest Native American membership proportion wasfound in three samples from Bolivia, El Salvador and Argentina(BS01, BS07, and BS15), as expected. Bone samples collected inArgentina, known to be from Welsh and Austrian individuals(BS17 and BS18, respectively), showed mainly European mem-bership proportions. Finally, variable levels of admixture werefound in the remaining samples from Bolivia (BS02), El Salvador(BS8, BS10) and Argentina (BS14, BS16), as outlined in Table1. These countries were colonized by Europeans over manycenturies and African slaves were brought to the regions at thesame period.

The likelihood ratios obtained in the Snipper Bayes ancestryanalysis of the 20 bone profiles are shown in Table 1. Theassignments based on the Snipper likelihoods matched thecluster membership proportions well, although in the case ofadmixed individuals, Snipper assigned ancestry to the maincontributor at markedly lower probabilities. In such cases, PCAplots can provide a more intuitive way to infer the admixturecontributors, as well as to gauge the co-ancestries in admixedsamples. PCA plots, outlining the first three principal components(PCs) of variation of all 20 bone profiles, are shown in Fig. 1. PC1and PC2 (Fig. 1A) capture approximately 39% of the variance in thedata and show three main clusters corresponding to Africans,Europeans, and a composed East Asian plus Native Americancluster, which is only clearly resolved by PC3 (Fig. 1B). Taking intoaccount the global PCA and STRUCTURE results, for an improvedreview of the population clusters in PCA we produced dedicatedthree-way PCA plots (Fig. 1C) including all bone profiles withnegligible East Asian co-ancestry (as seen in Table 1) whileaccordingly removing the East Asian reference data from theanalysis; conversely Fig. 1D shows the analysis of Vietnamese andEast Timorese profiles ignoring a negligible Native Americancomponent, hence removing Native American reference datafrom the analysis. PCA plot 1C, comparing African, European andAmerican Indel population data indicates that BS09 from ElSalvador has a main European co-ancestry but detectable minorAfrican co-ancestry (positioned on the left side of the Europeancluster), matching the STRUCTURE cluster proportions. Likewise,BS08 and BS10 indicate a main European co-ancestry and a minorAmerican co-ancestry, with STRUCTURE data and PCA positionsmatching well. PCA plot 1D, comparing African, European andEast Asian Indel population data indicates that East TimoreseBS06 has a minor European co-ancestry and BS04 a minor Africanco-ancestry. A supplementary four-way analysis in which theEast Timorese samples were compared to Oceanian instead ofNative American reference population data was made and plotsare shown in Supplementary Fig. S1. The PCA clusters shows adistribution and pattern of co-ancestry more typical of the EastTimor region positioned between South-East Asia and Oceania,despite European and African minor co-ancestries evident in BS06and BS04, respectively. Nevertheless, it should be noted that theIndels used were not originally selected to differentiate Ocea-nians from other groups and this remains the least well separatedpopulation group when using Indels alone. Furthermore, patternsof admixture in this part of the world are complex, therefore theresults obtained for the East Timor samples must be interpretedwith caution.

Supplementary material related to this article can be found, inthe online version, at doi:10.1016/j.fsigen.2014.11.025.

Fig. 1. Principal component analysis plots showing the first three components of variation for 20 bone samples compared to HGDP-CEPH reference data of four population

groups. (A) PC1 vs. PC2, and (B) PC1 vs. PC3 for all bone profiles. (C) South African and American profiles compared to African, European and Native American data only. (D)

Vietnamese and East Timorese compared to African, European and East Asian data only.

C. Romanini et al. / Forensic Science International: Genetics 16 (2015) 58–63 61

3.2. Indel genotyping performance

All 46-AIM-Indels were successfully genotyped in replicatesfrom 14 out of the 20 bone samples. DNA quantification for thesefourteen full profile samples ranged from 0.02 to 0.85 ng/ml. Forthe remaining six samples, although with allele dropout, afterperforming replicated reactions it was possible to report 30, 3336, 43, 44 and 45 markers as shown in Fig. 2. It is worthmentioning that DNA quantification values in these six sampleswere below 0.03 ng/mL or it was not possible to determine DNAconcentration after 40 qPCR cycles (BS01, BS18, BS19). From the45 total allele dropouts observed in these six samples, 19 (42%)were present in the ten Indel markers amplifying the largestamplicon sizes (200–230 bp), as expected. Samples with inde-terminate DNA quantification results, and evidence of PCRinhibition, were diluted for the PCR amplification, that is: BS02in which a total of 33 out of the 46 markers were successfullytyped, indicating that PCR inhibition effects could not becompletely overcome by serial dilutions or that very lowlevel DNA was present. The sample with the least markerrecovery (30 out of 46) also gave the lowest DNA concentration(0.015 ng/ml).

4. Discussion

Using African, American, European and East Asian HGDP-CEPHdiversity panel samples as reference population data, it waspossible to cluster samples from non-admixed individuals withone of the four ancestral populations, with high membershipproportions in one of the four reference group genetic clusters.Also, by obtaining variable membership proportions in more thanone group, we were able to detect the presence of admixture,mainly in samples from South and Central American countries andEast Timor, with both regions having widely recorded Europeanadmixture from colonial times. Other authors have found variabledegrees of admixture in South American countries among NativeAmericans, Africans and Europeans (e.g. [26–28]).

Our results show that the 46 AIM-Indel multiplex is robustenough to type and ascertain genetic ancestry information fromhuman bone remains. This additional information may be helpfulfrom the forensic perspective, in the identification of humanremains especially for identification programmes analyzingindividuals with different ancestries, for example, frontierinvestigations of illegal immigration or DVI, among others.Knowledge of the ancestry associated with skeletal remains opens

Fig. 2. Correlation between number of reportable AIM-Indel loci and DNA concentration in sample extracts. Round light grey dots: number of reportable markers, square dark

grey dots: DNA quantification values; zero value means no amplification after 40 qPCR cycles (samples BS01, BS19 and BS18).

C. Romanini et al. / Forensic Science International: Genetics 16 (2015) 58–6362

the possibility of redirecting investigations towards new informa-tion on alternative scenarios involving missing persons in thesame region, and the accurate collection of relevant familyreference materials. As this information may be very sensitive indifferent scenarios, ancestry data should be used carefully andrestricted to the purpose of identification of the victims,and ethical and confidentiality constraints should always beconsidered.

Applying the short amplicon 46 AIMs-Indel multiplex we wereable to obtain genetic profiles of all the bone remains analyzed,even in highly degraded DNA extracts showing allele dropout, thatis highly degraded and/or inhibited samples such as BS19 and BS20from Vietnam, giving 43 and 30 reportable markers respectively,were classified as East Asian. From the 20 tested bones, 95% of the46 markers were successfully typed, 14 of which showed 100%genotyping success. Informative estimates of ancestry wereobtained in each case, for DNA extracts with a wide range ofDNA quantification values (0.02–0.85 ng/ml); more than one half(11/20) presenting values below 0.2 ng/mL and some displayinginhibition characteristics.

In conclusion, the 46 AIM-Indel multiplex can efficientlyprovide informative ancestry estimations on aged bone samplesfrom four different continental origins, while being able to dealwith the usual problems presented by forensic samples, in thepresence of low DNA concentration, PCR inhibitors or variablelevels of severe DNA degradation.

Acknowledgements

RP is the recipient of a postdoctoral fellowship (SFRH/BPD/81986/2011) granted by the Portuguese Foundation for Scienceand Technology (FCT) and co-financed by the European Social Fund(Human Potential Thematic Operational Programme). IPATIMUP isan associate laboratory of the Portuguese Ministry of Science,Technology and Higher Education, and is partially supported byFCT. The authors thank Carla Santos, Forensic Genetics Unit,University of Santiago de Compostela for allowing use of PCAscripts implemented in Snipper.

References

[1] P. Kersbergen, K. van Duijn, A.D. Kloosterman, J.T. den Dunnen, M. Kayser, P. deKnijff, Developing a set of ancestry-sensitive DNA markers reflecting continentalorigins of humans, BMC Genet. 10 (2009) 69.

[2] M. Fondevila, C. Phillips, C. Santos, A. Freire Aradas, P.M. Vallone, J.M. Butler, M.V.Lareu, A. Carracedo, Revision of the SNPforID 34-plex forensic ancestry test: assayenhancements, standard reference sample genotypes and extended populationstudies, Forensic Sci. Int. Genet. 7 (2013) 63–74.

[3] M.D. Shriver, R. Mei, E.J. Parra, V. Sonpar, I. Halder, S.A. Tishkoff, T.G. Schurr, S.I.Zhadanov, L.P. Osipova, T.D. Brutsaert, et al., Large-scale SNP analysis revealsclustered and continuous patterns of human genetic variation, Hum. Genomics 2(2005) 81–89.

[4] O. Lao, K. van Duijn, P. Kersbergen, P. de Knijff, M. Kayser, Proportioning wholegenome single-nucleotide-polymorphism diversity for the identification of geo-graphic population structure and genetic ancestry, Am. J. Hum. Genet. 78 (2006)680–690.

[5] P. Paschou, E. Ziv, E.G. Burchard, S. Choudhry, W. Rodriguez-Cintron, M.W.Mahoney, P. Drineas, PCA-correlated SNPs for structure identification in world-wide human populations, PLoS Genet. 3 (2007) 1672–1686.

[6] M.A. Enoch, P.H. Shen, K. Xu, C. Hodgkinson, D. Goldman, Using ancestry-infor-mative markers to define populations and detect population stratification,J. Psychol. Pharmacol. 20 (2006) 19–26.

[7] C. Phillips, A. Salas, J.J. Sanchez, M. Fondevila, A. Gomez-Tato, J. Alvarez-Dios, M.Calaza, M.C. de Cal, et al., Inferring ancestral origin using a single multiplexassay of ancestry-informative marker SNPs, Forensic Sci. Int. Genet. 1 (2007)273–280.

[8] J.R. Kidd, F.R. Friedlaender, W.C. Speed, A.J. Pakstis, F.M. De La Vega, K.K. Kidd,Analyses of a set of 128 ancestry informative single-nucleotide polymorphisms ina global set of 119 population samples, Invest. Genet. 2 (2011) 1.

[9] C. Phillips, W. Parson, B. Lundsberg, C. Santos, A. Freire-Aradas, M. Torres, M.Eduardoff, C. Børsting, et al., Building a forensic ancestry panel from the groundup: the EUROFORGEN Global AIM-SNP set, Forensic Sci. Int. Genet. 11 (2014)13–25.

[10] C. Phillips, A. Freire Aradas, A.K. Kriegel, M. Fondevila, O. Bulbul, C. Santos, F.Serrulla Rech, M.D. Perez Carceles, et al., Eurasiaplex: a forensic SNP assay fordifferentiating European and South Asian ancestries, Forensic Sci. Int. Genet. 7(2013) 359–366.

[11] J.M. Mullaney, R.E. Mills, W.S. Pittard, S.E. Devine, Small insertions and deletions(INDELs) in human genomes, Hum. Mol. Genet. 19 (2010) R131–R136.

[12] R. Pereira, C. Phillips, N. Pinto, C. Santos, S.E. de Santos, et al., Straightforwardinference of ancestry and admixture proportions through ancestry-informativeinsertion deletion multiplexing, PLoS ONE 7 (2012) e29684.

[13] J.J. Sanchez, C. Phillips, C. Børsting, K. Balogh, M. Bogus, M. Fondevila,C.D. Harrison, E. Musgrave-Brown, et al., A multiplex assay with 52 singlenucleotide polymorphisms for human identification, Electrophoresis 27 (2006)1713–1724.

[14] R. Pereira, C. Phillips, C. Alves, A. Amorim, A. Carracedo, et al., A new multiplex forhuman identification using insertion/deletion polymorphisms, Electrophoresis30 (2009) 3682–3690.

C. Romanini et al. / Forensic Science International: Genetics 16 (2015) 58–63 63

[15] C. Phillips, M. Fondevila, M.V. Lareu, A 34-plex autosomal SNP single baseextension assay for ancestry investigations, Methods Mol. Biol. 830 (2012) 109–126.

[16] R. Alaeddini, Forensic implications of PCR inhibition: a review, Forensic Sci. Int.Genet. 6 (2012) 297–305.

[17] R. Alaeddini, S.J. Walsh, A. Abbas, Forensic implications of genetic analysis fromdegraded DNA: a review, Forensic Sci. Int. Genet. 4 (2010) 149–157.

[18] M. Fondevila, C. Phillips, N. Naveran, L. Fernandez, M. Cerezo, A. Salas, A.Carracedo, M.V. Lareu, Case report: identification of skeletal remains using short-amplicon marker analysis of severely degraded DNA extracted from a decomposedand charred femur, Forensic Sci. Int. Genet. 2 (2008) 212–218.

[19] C. Romanini, M.L. Catelli, A. Borosky, R. Pereira, M. Romero, M. Salado Puerto, C.Phillips, et al., Typing short amplicon binary polymorphisms: supplementary SNPand Indel genetic information in the analysis of highly degraded skeletal remains,Forensic Sci. Int. Genet. 6 (2012) 469–476.

[20] Latin American Initiative for the Identification of the ‘‘Disappeared’’ (LIID),Argentinean Forensic Anthropology Team (EAAF). http://www.eaaf.org/iniciativa_en/.

[21] A. Carracedo, W. Bar, P. Lincoln, W. Mayr, N. Morling, B. Olaisen, P. Schneider, B.Budowle, et al., DNA Commission of the International Society for ForensicGenetics: guidelines for mitochondrial DNA typing, Forensic Sci. Int. Genet.110 (2000) 79–85.

[22] J.K. Pritchard, M. Stephens, P. Donnelly, Inference of population structure usingmultilocus genotype data, Genetics 155 (2000) 945–959.

[23] D. Falush, M. Stephens, J.K. Pritchard, Inference of population structure usingmultilocus genotype data: linked loci and correlated allele frequencies, Genetics164 (2003) 1567–1587.

[24] H.M. Cann, C. de Toma, L. Cazes, M.F. Legrand, V. Morel, et al., A human genomediversity cell line panel, Science 296 (2002) 261–262.

[25] N.A. Rosenberg, Standardized subsets of the HGDP-CEPH Human Genome Diver-sity Cell Line Panel, accounting for atypical and duplicated samples and pairs ofclose relatives, Ann. Hum. Genet. 70 (2006) 841–847.

[26] N.P. Santos, E.M. Ribeiro-Rodrigues, A.K. Ribeiro-dos-Santos, R. Pereira,L. Gusmao, et al., Assessing individual interethnic admixture and populationsubstructure using a 48-insertion-deletion (INDEL) ancestry informative marker(AIM) panel, Hum. Mut. 31 (2010) 184–190.

[27] U. Toscanini, L. Gusmao, G. Berardi, A. Gomez, R. Pereira, E. Raimondi, Ancestryproportions in urban populations of Argentina, Forensic Sci. Int. Genet. Suppl. Ser.3 (2011) 387–388.

[28] T. Heinz, V. Alvarez-Iglesias, J. Pardo-Seco, P. Taboada-Echalar, A. Gomez-Carballa,A. Torres-Balanza, O. Rocabado, A. Carracedo, C. Vullo, A. Salas, Ancestry analysisreveals a predominant Native American component with moderate Europeanadmixture in Bolivians, Forensic Sci. Int. Genet. 7 (2013) 537–542.