The American College of Medical Genetics and Genomics (ACMG) has published recommended standards and guidelines for clinical case review for copy number variations (CNVs). These guidelines provide means to classify variants into different well-defined classes (e.g. Pathogenic, Benign, etc.). Integrated into the NxClinical software system is the Variant Interpretation Assistant (VIA), which uses a decision tree machine learning approach to set up rules for automatic pre-classification of events to be reviewed and interpreted by laboratory personnel. The system allows users to embed guidelines for efficiency and consistency in reporting. Benign, likely benign, pathogenic and likely pathogenic regions are pre-classified based on public copy number polymorphism databases and publications, including ClinGen, OMIM, and DECIPHER, and on previous in-house case classification results. VIA tracks results in real time using previously annotated samples to allow for interpretation of newly uploaded samples. Additionally, the system allows for multiple review and edits, as needed, with extensive user-based audit trailing.
As defined by ACMG reporting guidelines, the final report results include a description of the position, size, amino acid affected (if applicable), type of alteration detected, designation of the genes within the interval, and the clinical significance as defined by final review. Examples of both individual and familial case reviews will be shown.
