The growth and applications that Forensic Investigative Genetic Genealogy (FIGG) have experienced during the last years have been rapid, but more validation studies using realistic casework forensic samples are required to integrate this technology into forensic workflows. The main goal of this study was to test the performance of the AxiomTM Spain Biobank Array (TFS), which consists of 756,834 SNPs and it is commonly applied to biomedical samples, in typical forensic samples, where the quantity and quality of DNA represent a challenge. Therefore, considering the typical incomplete profiles and the high number of SNPs needed to solve distant kinship cases, we explored the performance of genotype imputation in simulated low-quality samples and its impact on the kinship analysis using a method based on calling genomic segments and a method based on the widely used likelihood framework. 

A total of 89 samples were genotyped with the AxiomTM Array, including buccal swabs from individuals with diverse kinship relationships at different concentrations and bone samples. Forensic low-quality samples were simulated by randomly selecting 50,000 SNPs or deleting from 10,000 to 500,000 SNPs. Genotype imputation of simulated low-quality samples was performed, and its accuracy was calculated comparing the imputed genotypes to the original ones. Genotype refinement was performed in real low-quality samples to determine the reliability of the obtained genotypes.

Our results showed that for the segment approach more than 100,000 SNPs were required, although with less SNPs some real shared segments could be detected, with the risk of finding false positives. With the likelihood approach we were able to solve distant cases up to third cousins with only ≈6,000 SNPs. Genotype imputation error rate, despite being low, increased when the input SNPs decreased in simulated low-quality samples. However, the imputation helped to recover partial profiles and to detect real segments shared by the individuals. Conversely, the likelihood approach did not greatly benefit from imputations, and no false positives were detected for the unrelated individuals. Following genotype refinement inconclusive results were obtained for both segment and likelihood approach.

Finally, we conclude that, despite the analysis of low-quality samples such as bone remains might present a limitation, the AxiomTM Array can be used as a tool for distant kinship cases where reference samples or genotypes of shared SNPs obtained with other technologies are available.

Authors

• Amaia Cabrejas-Olalla (Forensic Genetics Unit, Institute of Forensic Sciences, Universidade de Santiago de Compostela, Spain)
• Daniel Kling (Department of Forensic Genetics and Forensic Toxicology, National Board of Forensic Medicine / Department of Forensic Sciences, Oslo University Hospital, Oslo, Norway, Sweden)
• María de la Puente (Forensic Genetics Unit, Institute of Forensic Sciences, Universidade de Santiago de Compostela, Spain)
• Inés Quintela (Fundación Pública Galega de Medicina Xenómica, SERGAS / Grupo de Medicina Xenómica, Universidade de Santiago de Compostela, Spain)
• José Javier Suárez-Rama (Fundación Pública Galega de Medicina Xenómica, SERGAS / Grupo de Medicina Xenómica, Universidade de Santiago de Compostela, Spain)
• Amelia Rodríguez-López (Forensic Genetics Unit, Institute of Forensic Sciences, Universidade de Santiago de Compostela, Spain)
• Javier González-Bao (Forensic Genetics Unit, Institute of Forensic Sciences, Universidade de Santiago de Compostela, Spain)
• Ángel Carracedo (Forensic Genetics Unit, Institute of Forensic Sciences, Universidade de Santiago de Compostela (USC) / Fundación Pública Galega de Medicina Xenómica, SERGAS / Grupo de Medicina Xenómica, USC, Spain)
• Christopher Phillips (Forensic Genetics Unit, Institute of Forensic Sciences, Universidade de Santiago de Compostela / King’s Forensics, Faculty of Life Sciences and Medicine, King’s College, London, UK, Spain)
• María Victoria Lareu (Forensic Genetics Unit, Institute of Forensic Sciences, Universidade de Santiago de Compostela, Spain)
• Andreas Tillmar (Department of Forensic Genetics and Forensic Toxicology, National Board of Forensic Medicine / Department of Biomedical and Clinical Sciences, Faculty of Medicine and Health Sciences, LIU, Spain)
• Ana Mosquera-Miguel (Forensic Genetics Unit, Institute of Forensic Sciences, Universidade de Santiago de Compostela, Spain)