Semi-Automated Unidirectional Sequence Analysis for Mutation Detection in a Clinical Diagnostic Setting

Background: The past 10 years have seen an improvement in sequence data quality due to the introduction of capillary sequencers and new sequencing chemistries. In parallel, new software programs for automated mutation detection have been developed. We evaluated the sensitivity of semiautomated unidirectional sequence analysis for the detection of heterozygous base substitutions using the Mutation Surveyor software package. Methods: Detection rates for heterozygous base substitutions in 29 genes by automated and visual inspection were compared. Examples of heterozygous bases not detected in one direction during bidirectional analysis were also sought through a national survey of United Kingdom (UK) genetics laboratories. Sequence quality was assessed in a consecutive cohort of 50 patients for whom the 39 exons of the ABCC8 gene had been sequenced in one direction. Results: A total of 701 different heterozygous base substitutions were detected by the software with no false negatives ( sensitivity >= 99.57%). Four examples of heterozygous bases missed in one direction during bidirectional analysis were reported. Two were detected using unidirectional analysis settings, and the other two bases had low-quality scores. Of the 1950 amplicons examined, 97.2% had a quality score >= 30 and an average PHRED-like score >= 50 for the defined region of interest, and 98.1% of the 323,650 bases had a PHRED score >40. Conclusions: We found no evidence to support a requirement for bidirectional sequencing. Semi-automated analysis of good quality unidirectional sequence data has high sensitivity and is suitable for heterozygote mutation scanning in clinical diagnostic laboratories. Further work is required to determine minimum quality parameters for semiautomated analysis.