OBJECTIVE: Copy-number variation (CNV) is an important source of genetic diversity in humans. It can cause Mendelian or sporadic traits or be associated with complex diseases by various molecular mechanisms, including gene dosage, gene disruption, gene fusion and position effects. In clinical diagnostics, it is therefore fundamental to be able to identify such variations. The preferred techniques for CNV detection are MLPA, aCGH and qPCR, which have proven to be valuable, and they are complex, costly and require prior knowledge of the region to analyze. CNV calling from NGS data still suffers from data variability. Coverage can vary greatly from one region of the genome to another, depending on many factors like complexity, GC content, repeated regions and many others. In this paper, we describe how we developed a method for CNV detection. MATERIALS AND METHODS: Our method exploits CoNVaDING to detect single- and multiple-exon CNVs in targeted NGS data. RESULTS: We demonstrated that our CNV analysis has 100% specificity and 99.998% sensitivity. We also show how we evaluated the performance of this method based on internal analysis. CONCLUSIONS: The results indicate that the method can be used to screen prior to standard labs technologies, thus reducing the number of analyses, as well as costs, and increasing test conclusiveness.
OBJECTIVE: Copy-number variation (CNV) is an important source of genetic diversity in humans. It can cause Mendelian or sporadic traits or be associated with complex diseases by various molecular mechanisms, including gene dosage, gene disruption, gene fusion and position effects. In clinical diagnostics, it is therefore fundamental to be able to identify such variations. The preferred techniques for CNV detection are MLPA, aCGH and qPCR, which have proven to be valuable, and they are complex, costly and require prior knowledge of the region to analyze. CNV calling from NGS data still suffers from data variability. Coverage can vary greatly from one region of the genome to another, depending on many factors like complexity, GC content, repeated regions and many others. In this paper, we describe how we developed a method for CNV detection. MATERIALS AND METHODS: Our method exploits CoNVaDING to detect single- and multiple-exon CNVs in targeted NGS data. RESULTS: We demonstrated that our CNV analysis has 100% specificity and 99.998% sensitivity. We also show how we evaluated the performance of this method based on internal analysis. CONCLUSIONS: The results indicate that the method can be used to screen prior to standard labs technologies, thus reducing the number of analyses, as well as costs, and increasing test conclusiveness.