Growing insights in biomedical research indicates a substantial role of copy number variants (CNVs) in various diseases. CNVs are represented by duplicated or deleted parts of various lengths and can affect multiple genes or change gene dosage, lead to different pathologies. CNVs can be detected by advanced bioinformatics approaches that are currently based on aligning reads to static reference genome and investigating read distributions in windowed fractions of the genome. Particularly interesting for the fields of prenatal and cancer genomics is the accurate quantification of CNVs in mixed samples, such as mixed maternal and fetal DNA in the blood of pregnant women or mixed cancer and normal DNA which is also present in blood of a patient. Graph-based genome reference will lead to lower ambiguity and higher precision of CNV detection in such samples.
New algorithms for CNV detection and quantification using graph-based reference.