Limits...
Ancestry analysis in the 11-M Madrid bomb attack investigation.

Phillips C, Prieto L, Fondevila M, Salas A, Gómez-Tato A, Alvarez-Dios J, Alonso A, Blanco-Verea A, Brión M, Montesino M, Carracedo A, Lareu MV - PLoS ONE (2009)

Bottom Line: Although mitochondrial DNA (mtDNA) and Y-chromosome markers routinely demonstrate informative geographic differentiation, the populations compared in this analysis were known to show a proportion of shared mtDNA and Y haplotypes as a result of recent gene-flow across the western Mediterranean, while any two loci can be unrepresentative of the ancestry of an individual as a whole.Once misclassification error was found to be sufficiently low, genotyping yielded seven near-complete profiles (33 of 34 AIM-SNPs) that in four cases gave probabilities providing a clear assignment of ancestry.The results achieved illustrate the benefit of adding specialized marker sets to provide enhanced scope and power to an already highly effective system of DNA analysis for forensic identification.

View Article: PubMed Central - PubMed

Affiliation: Forensic Genetics Unit, Institute of Legal Medicine, University of Santiago de Compostela, Santiago de Compostela, Galicia, Spain. c.phillips@mac.com

ABSTRACT
The 11-M Madrid commuter train bombings of 2004 constituted the second biggest terrorist attack to occur in Europe after Lockerbie, while the subsequent investigation became the most complex and wide-ranging forensic case in Spain. Standard short tandem repeat (STR) profiling of 600 exhibits left certain key incriminatory samples unmatched to any of the apprehended suspects. A judicial order to perform analyses of unmatched samples to differentiate European and North African ancestry became a critical part of the investigation and was instigated to help refine the search for further suspects. Although mitochondrial DNA (mtDNA) and Y-chromosome markers routinely demonstrate informative geographic differentiation, the populations compared in this analysis were known to show a proportion of shared mtDNA and Y haplotypes as a result of recent gene-flow across the western Mediterranean, while any two loci can be unrepresentative of the ancestry of an individual as a whole. We based our principal analysis on a validated 34plex autosomal ancestry-informative-marker single nucleotide polymorphism (AIM-SNP) assay to make an assignment of ancestry for DNA from seven unmatched case samples including a handprint from a bag containing undetonated explosives together with personal items recovered from various locations in Madrid associated with the suspects. To assess marker informativeness before genotyping, we predicted the probable classification success for the 34plex assay with standard error estimators for a naïve Bayesian classifier using Moroccan and Spanish training sets (each n = 48). Once misclassification error was found to be sufficiently low, genotyping yielded seven near-complete profiles (33 of 34 AIM-SNPs) that in four cases gave probabilities providing a clear assignment of ancestry. One of the suspects predicted to be North African by AIM-SNP analysis of DNA from a toothbrush was identified late in the investigation as Algerian in origin. The results achieved illustrate the benefit of adding specialized marker sets to provide enhanced scope and power to an already highly effective system of DNA analysis for forensic identification.

Show MeSH
Principal component analysis of case samples and the extended Spanish (black points, n = 127) and North African (red points, n = 240) training sets.PC1 and PC2 % contributions shown in brackets.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2719087&req=5

pone-0006583-g002: Principal component analysis of case samples and the extended Spanish (black points, n = 127) and North African (red points, n = 240) training sets.PC1 and PC2 % contributions shown in brackets.

Mentions: Finally, the clearest principal component analysis was obtained using the expanded training sets with improved patterns compared to original attempts using 48-sample sets. The plot of the two principal components is shown in Figure 2 and these accounted for 12.5%, and 5.3% of the total variation.


Ancestry analysis in the 11-M Madrid bomb attack investigation.

Phillips C, Prieto L, Fondevila M, Salas A, Gómez-Tato A, Alvarez-Dios J, Alonso A, Blanco-Verea A, Brión M, Montesino M, Carracedo A, Lareu MV - PLoS ONE (2009)

Principal component analysis of case samples and the extended Spanish (black points, n = 127) and North African (red points, n = 240) training sets.PC1 and PC2 % contributions shown in brackets.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2719087&req=5

pone-0006583-g002: Principal component analysis of case samples and the extended Spanish (black points, n = 127) and North African (red points, n = 240) training sets.PC1 and PC2 % contributions shown in brackets.
Mentions: Finally, the clearest principal component analysis was obtained using the expanded training sets with improved patterns compared to original attempts using 48-sample sets. The plot of the two principal components is shown in Figure 2 and these accounted for 12.5%, and 5.3% of the total variation.

Bottom Line: Although mitochondrial DNA (mtDNA) and Y-chromosome markers routinely demonstrate informative geographic differentiation, the populations compared in this analysis were known to show a proportion of shared mtDNA and Y haplotypes as a result of recent gene-flow across the western Mediterranean, while any two loci can be unrepresentative of the ancestry of an individual as a whole.Once misclassification error was found to be sufficiently low, genotyping yielded seven near-complete profiles (33 of 34 AIM-SNPs) that in four cases gave probabilities providing a clear assignment of ancestry.The results achieved illustrate the benefit of adding specialized marker sets to provide enhanced scope and power to an already highly effective system of DNA analysis for forensic identification.

View Article: PubMed Central - PubMed

Affiliation: Forensic Genetics Unit, Institute of Legal Medicine, University of Santiago de Compostela, Santiago de Compostela, Galicia, Spain. c.phillips@mac.com

ABSTRACT
The 11-M Madrid commuter train bombings of 2004 constituted the second biggest terrorist attack to occur in Europe after Lockerbie, while the subsequent investigation became the most complex and wide-ranging forensic case in Spain. Standard short tandem repeat (STR) profiling of 600 exhibits left certain key incriminatory samples unmatched to any of the apprehended suspects. A judicial order to perform analyses of unmatched samples to differentiate European and North African ancestry became a critical part of the investigation and was instigated to help refine the search for further suspects. Although mitochondrial DNA (mtDNA) and Y-chromosome markers routinely demonstrate informative geographic differentiation, the populations compared in this analysis were known to show a proportion of shared mtDNA and Y haplotypes as a result of recent gene-flow across the western Mediterranean, while any two loci can be unrepresentative of the ancestry of an individual as a whole. We based our principal analysis on a validated 34plex autosomal ancestry-informative-marker single nucleotide polymorphism (AIM-SNP) assay to make an assignment of ancestry for DNA from seven unmatched case samples including a handprint from a bag containing undetonated explosives together with personal items recovered from various locations in Madrid associated with the suspects. To assess marker informativeness before genotyping, we predicted the probable classification success for the 34plex assay with standard error estimators for a naïve Bayesian classifier using Moroccan and Spanish training sets (each n = 48). Once misclassification error was found to be sufficiently low, genotyping yielded seven near-complete profiles (33 of 34 AIM-SNPs) that in four cases gave probabilities providing a clear assignment of ancestry. One of the suspects predicted to be North African by AIM-SNP analysis of DNA from a toothbrush was identified late in the investigation as Algerian in origin. The results achieved illustrate the benefit of adding specialized marker sets to provide enhanced scope and power to an already highly effective system of DNA analysis for forensic identification.

Show MeSH