Abstract
Purpose: :
Sanger sequencing is a common method used to test patients for mutations within the retinoblastoma gene (RB1). Obtaining high quality data from Sanger sequencing of RB1 is technically challenging due to homopolymer tracts bordering many exons and other sequence-related issues. We found these problems could be greatly improved by manipulating the size and position of sequencing primers.
Methods: :
Position specific error probability estimates of individual base calls (QV) and length of read (LOR) were used to estimate DNA sequence quality for each primer/template combination. The performance of 64 different primers in 825 robotically assembled sequencing reactions was systematically quantified in the context of a highly automated diagnostic assay.
Results: :
Sequence data quality from different sequencing primers under carefully controlled reaction conditions with the same DNA template varied widely and differences were reproducible and often highly significant. The mean probability of error was reduced by more than 99% for several regions of interest in RB1 by supplementing the PCR primer with a custom internal primer for sequencing and we are able to obtain bidirectional data from all RB1 exons, regardless of sequence context.
Conclusions: :
We conclude that sequencing primer choice can be a critical determinant of success and quality for resequencing projects involving difficult templates and can provide more reliable results to patients and families with heritable retinoblastoma.
Keywords: retinoblastoma • genetics • mutations