NCJ Number
206374
Journal
Journal of Forensic Sciences Volume: 49 Issue: 3 Dated: May 2004 Pages: 492-499
Date Published
May 2004
Length
8 pages
Annotation
This paper reports on the design, development, and validation of an automated allele concordance analysis system (CompareCalls sm, patent pending), which compares two independent DNA analysis platforms to ensure the highest accuracy for allele calls.
Abstract
The automated STR concordance analysis system is based on using two software analysis packages that identify the concordant and discordant STR allele calls. The concordant STR allele calls require no further human review, but the discordant STR allele calls do require further human review. As long as the discordant STR allele calls are a small portion of the overall data, the automated STR concordance analysis system will alleviate much of the human STR profile interpretation currently used. The automated STR concordance analysis system described in this paper meets these requirements and can alleviate the majority of the human time required to review STR data generated from outsourced CODIS samples. These findings have been supported by the review of over 600,000 lanes of STR data. Because automated allele concordance analysis systems based on independent allele calling algorithms are less prone to error than the current technical review process, the NDIS board of CODIS is justified in setting criteria for implementing expert systems. The use of such a validated expert system will produce high-quality data for CODIS while freeing the human DNA data analyst to perform other critical duties. 5 figures, 4 tables, and 21 references