Automating the identification of DNA variations using quality-based fluorescence re-sequencing: analysis of the human mitochondrial genome

Diagnostic re-sequencing plays a central role in medical and evolutionary genetics. In this report we describe a process that applies fluorescence-based re-sequencing and an integrated set of analysis tools to automate and simplify the identification of DNA variations using the human mitochondrial genome as a model system. Two programs used in genome sequence analysis (Phred, a base-caller, and Phrap, a sequence assembler) are applied to assess the quality of each base call across the sequence. Potential DNA variants are automatically identified and 'tagged' by comparing the assembled sequence with a reference sequence. We also show that employing the Consed program to display a set of highly annotated reference sequences greatly simplifies data analysis by providing a visual database containing information on the location of the PCR primers, coding and regulatory sequences and previously known DNA variants. Among the 12 genomes sequenced 378 variants including 29 new variants were identified along with two heteroplasmic sites, automatically detected by the PolyPhred program. Overall we document the ease and speed of performing high quality and accurate fluorescence-based re-sequencing on long tracts of DNA as well as the application of new approaches to automatically find and view DNA variants among these sequences.