학술논문

Pitfalls in variant annotation for hereditary cancer diagnostics: The example of Illumina® VariantStudio®.
Document Type
Article
Source
Genomics. Jan2021:Part 2, Vol. 113 Issue 1, p748-754. 7p.
Subject
*HUMAN chromosome abnormality diagnosis
*ANNOTATIONS
*DATA analysis
Language
ISSN
0888-7543
Abstract
Next Generation Sequencing (NGS), and specifically targeted panel sequencing is the state-of-the-art in clinical genetic diagnosis of Mendelian diseases. However, the bioinformatics analysis and interpretation of the generated data can be challenging. A spotlight on the default transcript selection of a user-friendly, commercially available software that is widely used by genetics professionals, i.e. Illumina® VariantStudio®, is presented. For the sake of comparison, we employed Ensembl VEP, an open-source command-line tool, as it provides flexibility regarding transcript selection. The analysis of NGS data deriving from sequencing of 857 germline DNA samples of cancer patients indicated a concordance of 82.82% between the two software programs. Significantly, using the default transcript configuration of VariantStudio®, we failed to annotate correctly 11.45% of the identified loss-of-function variants. Our results underline the importance of cautious software and transcript selection and the need for reliable, white-box data analysis, along with bioinformatics expertise in clinical diagnostics. • A detailed comparison between two annotation methods highlights the pitfalls in variant annotation in a clinical setting. • Using the default parameters in VariantStudio®, 11.45% of the identified loss-of-function variants were misannotated. • The use of black-box software in a clinical setting can give misleading results. [ABSTRACT FROM AUTHOR]