P039 – Evaluating the Accuracy of Simulated Likelihood Ratio Distributions Used in Kinship Analysis

11:01

Wednesday August 19^th

Station 08

Duration: 12 minutes

03. Human identification

Lucinda Davenport

Using genetic information to investigate the way in which two individuals may be related is an important application of forensic DNA typing. Pairwise kinship assessments will often compare the probability of observing the individuals’ genotype data under two competing relationship hypotheses, for example they are related as full-siblings or they are unrelated (i.e. any alleles that may be shared, are shared by chance). The result of such a kinship evaluation is presented as a likelihood ratio (LR), whereby the greater the LR value, the stronger the statistical support for a chosen relationship hypothesis.

It can be helpful to understand the expected range and distribution of LR values based on the specifics of a given case. In many forensic disciplines, such evaluations can be made based on “ground truth” data, however, this is often hard to come by in a kinship context. Therefore, a commonly used method for estimating the expected likelihood ratios for kinship scenarios involves simulating genotype data. These simulations can be conditioned on the allele frequencies of typed loci, the pedigree hypotheses compared and other relevant parameters such as recombination rates between linked markers. The distribution of the expected LR values can demonstrate the general capability of targeted genetic loci and/or analysis methodologies for resolving such relationship cases. Overall, this can provide important information, either to aid in deciding what markers to use in a given case or giving context to an LR value when reporting results. A simulation-based approach, however, relies upon the assumption that the simulated case genotypes and subsequently calculated LRs are representative of real cases.

This work aimed to explore the factors that influence the accuracy of simulated likelihood ratios, by comparing the distribution of theoretical LRs with the actual distribution of LRs generated from a large number of known sibling pairs. The impact of multiple calculation parameters was assessed. In particular, the results highlighted that the concordance between the simulated and real-life data sets was greatly influenced by the population-of-origin of the sibling pairs, and the corresponding population substructure. Additionally, accounting for genetic linkage when using expanded marker sets was found to be notably important.

Overall, understanding how representative simulated LR distributions are of genuine case LRs is an important and under-studied aspect of kinship analysis. Given that these simulated datasets are frequently used for evaluations in multiple contexts, it is crucial to understand the reliability of using this approach.

Authors

Lucinda Davenport (King's College London, United Kingdom)
Somruetai Satmun (King's College London, United Kingdom)
Denise Syndercombe Court (King's College London, United Kingdom)
David Ballard (King's College London, United Kingdom)

Contact:

Follow us

P039 – Evaluating the Accuracy of Simulated Likelihood Ratio Distributions Used in Kinship Analysis

Authors

On the same topic

P041 – A Sample-Specific Kinship Analysis Strategy with Whole Genome Sequencing Data for Highly Degraded Samples

P036 – Forensic Intelligence Production through Dynamic Analysis of DNA Match Data in Co-offending Networks Combined with Police Information

P031 – Systematic Evaluation of RNA-DNA Differences in Forensic Body Fluids: Toward Reliable RNA-based Forensic Applications

P026 – What’s Left Behind? A Case‑Specific Investigation of possible Human DNA Detection in Soil

P046 – Enhancing the Performance of the ForenSeq® Kintelligence HT Workflow for Poor-Quality Samples

P042 – ClusIBD: a Promising Tool for IBD Detection and Kinship Inference from Challenging Human Remains

P047 – Between Full and Half Siblings: Robust Kinship Analysis Using the MIcrohaplotype Panel for Extended Kinship Analyses (MIEKA)

P037 – The Evolution of Rapid DNA Technology and its Applications

P032 – Comparisons of DNA and STR Recovery from Burned Human Tissue and Bone

P027 – DNA Quantification Thresholds for MPS Analysis of Highly Degraded Skeletal Remains

P028 – Performance of the ForenSeq™ Kintelligence HT System on Environmentally Exposed Post-Mortem Bone Samples

P033 – Accidental Brothers: The Case of the Exchanged Twins in Colombia that has Captured Global Attention

P038 – Collaborative Capacity Building in Vietnam: Transferring Hybridization Capture Competence Across Laboratories

P043 – HyperMatch: A 91-plex Hyperplex PCR Assay for Human Identification with Single-digit Picogram Sensitivity and High Resistance to PCR Inhibition

P048 – Comparison of Kinship Inference Methods for Low-coverage Whole-genome Sequencing Data Before and After Imputation

P044 – Rapid Direct STR Profiling from Toothbrushes Using microFLOQ® Direct Swabs: A DVI-Oriented Workflow Study

P049 – CRISPR Diversity in the Human Salivary Microbiome as a Novel Forensic Biomarker for Discriminating Monozygotic Twins

P029 – Impact of Temperature on Dental DNA Recovery: Capillary Electrophoresis versus Next-Generation Sequencing

P034 – A Simplified Single-cell DNA Typing Workflow for Forensic Mixture Deconvolution Using the DispenCell System

P030 – ID-Lead: a cost-effective SNP panel for human identification and forensic leads

P035 – Tooth Image Colorimetry Combined with Machine Learning Enables Rapid Pre-screening for DNA Analysis

P040 – A Genetic Investigation of a Medieval Slovenian Cemetery Using Next-Generation Sequencing

P045 – Characterization of Microhaplotypes Variation from full FORENSEQ SNP Amplicons

P050 – The Use of the RapidINTEL™ Plus Kit and the RapidHIT™ ID System to Analyse Contact Traces Secured on SceneSafe™ Adhesive Tape.

P116 – Improvement of DNA extraction from FFPE tissue blocks and histological slides for forensic molecular analysis

P119 – Evaluating the RapidHIT™ ID System for Sexual Assault and Crime Evidence

P122 – When Tooth Type Matters: Distinct Methylation Signatures for Forensic Age Estimation

P125 – DNA Databases for Separated Children: Infrastructure, Ethics, and Global Reunification Strategies

P117 – Identification of a Missing Woman 57 Years After a Motor Vehicle Accident

P120 – Epigenomic Profiling in Human Cadavers in Southern Quebec (Dfb)

P123 – Harnessing the Power of Microhaplotypes for Forensic Identification in the Democratic Memory Context in Spain

P126 – Validation of the New EZ2 DNA Investigator Sep&Prep Kit for Automated Processing of Sexual Assault Samples

P129 – A Portable LAMP-microfluidic Point-of-care System for Rapid Human DNA Detection and Sex Identification

P127 – Growth Beyond STRs: Development and Optimization of a Microhaplotype Analytic Workflow for Enhanced Mixture Deconvolution

P130 – A Comparative Study of Chip and NGS-based Approaches for the Expanded Application of Kinship Analysis based on Genomic Segment Sharing

P121 – Pushing the Boundaries of Kinship Inference: High-Density SNP Profiling of 400-Year-Old Skeletal Remains.

P118 – Towards selective male DNA profiling: A microfluidic strategy for epithelial cell mixtures

P124 – Making the Hair Testify: Shotgun Sequencing of Short Telogen Hairs for Human Identification and Forensic Investigative Genetic Genealogy

P186 – Optimising SNP Panels for Forensic Kinship Analysis: From High-Density Arrays to Practical Implementation

P196 – Identifying Seminal Stains on Human Hair and Skin using STK(R) SpermTracker Skin

P201 – Degradation Mechanism Matters: Interpretation of the Degradation Index in STR and Y-STR Profiling

P206 – Internal Validation of the Precision ID mtDNA Whole Genome Panel and Ion Torrent Sequencing for Forensic Casework Applications

P187 – Improving Validation Studies, Quality Data Evaluation and Consistency

P192 – Development and Validation of a Targeted Capture-Based mRNA-SNP Panel for Body Fluid Identification and Contributor Assignment

P197 – Skeletal DNA Mapping of the Post-cranial Skeleton for the Identification of Human Remains in a Modern Forensic Context

P202 – Performance Evaluation of an Infinium HTS iSelect Custom Microarray Panel Under Low Template and Inhibitory Conditions for Forensic Applications

P207 – RNA-STR/SNP: A Mirror of Forensic DNA-STR/SNP

P203 – Assessing the Micro- and Macroscopic Changes of Chemically Altered Human Bone and Teeth

P208 – A Novel Standard for Identifying Half Siblings Developed Based on Likelihood Ratio

P198 – Persistence, Recovery, and Biological Context of Human DNA in Burial Soils for Forensic Applications

P193 – A High-resolution HLA-A Genotyping Workflow Based on Nanopore Sequencing and its Forensic Applicability

P188 – DNA Identification of Unknown Remains from the Battle of Hürtgen Forest

P199 – When DNA Needs a Story: How Archival Evidence Shapes Identification Outcomes

P194 – A New Forensic Genetics Open-Access Textbook

P189 – PCR artefacts in forensic STR typing using GlobalFiler™ IQC

P190 – Comparison of algorithms for kinship inference using the Verogen ForenSeq® Kintelligence Kit

P195 – Assessing DNA Yield in Three Bones Sites Using a Less Destructive Powdering Method

P200 – Pro-Búsqueda and the Long-term Use of DNA for Family Reunification in El Salvador

P285 – Occupational Environments are Associated with Distinct Oral and Gut Microbiome–metabolome Profiles

P282 – State DNA Database of Victims of the War and the Dictatorship in Spain. Validation of the Bonaparte Software by the INTCF.

P279 – Non-parametric Simulation of DNA Mixture Profiles from One-person Trace Profiles

P277 – Evaluating single cell whole genome amplification approaches for STR profiling to resolve DNA mixtures

P280 – Using DNA Methylation Markers in Bones and Cartilage to Facilitate Identification of Human Remains

P283 – Extended Microhaplotypes Enabled by Long-read Sequencing: A High-Informativeness Marker System for Complex DNA Mixture Deconvolution

P286 – Identification of Age-related DNA Methylation Markers and Application in Forensic age Prediction

P275 – Donor Identification From Single Telogen Hairs Through Transcriptome Sequencing

P278 – Recovery and SNP Genotyping of Nuclear DNA from Rootless Hair Shafts Using Hybridization Based Target Enrichment

P281 – Simultaneous Capture Enrichment of mtDNA and Nuclear SNP Targets Using a Baits Spike-in Approach

P284 – From Mixture Detection to Haplotype Reconstruction: Read-level Approaches for Degraded and Mixed DNA

P287 – Shared Genomic Segment Analysis for Inferring Distant and Complex Genetic Relationships

Contact:

Follow us