Unlike existing methods for this, which operate on the number of peaks in the signal and/or the rarity of the alleles, NOCIt uses both the rarity of the alleles and the quantitative data in the signal, i.e., the heights of the peaks. NOCIt was calibrated using single-source samples amplified from various low-level DNA amounts (0.007 - 0.25 ng) and three different times of injection (5, 10, and 20s). The peak heights were modeled using the Gaussian distribution, and appropriate parameters (mean and variance) were obtained from the calibration data. Dropout rates were also computed at each DNA mass. The performance of NOCIt was compared with the Maximum Allele Count (MAC) and the Maximum Likelihood Estimator (MLE) methods. The performance of NOCIt was consistently better than MAC and MLE across all DNA amounts and all times of injection for the 1-, 2-, and 3-person samples. In cases where NOCIt failed to pick the correct number of contributors, it was able to identify the region in which the NOC is most likely to lie. NOCIt has been implemented using the Java programming language and is currently available on www.bu.edu/dnamixtures. 8 tables, 26 figures, and 36 references
Downloads
Similar Publications
- Victim Participation in Intimate Partner Violence Prosecution: Implications for Safety
- Cognitive and contextual influences in determination of latent fingerprint suitability for identification judgments
- Two-Stage Approach for the Inference of the Source of High-Dimension and Complex Chemical Data in Forensic Science