Improved Variant Detection in Clinical Myeloid NGS Testing by Supplementing a Commercial Myeloid NGS Assay with Custom or Extended Data Filtering and Accessory Fragment Analysis
| Autor: | Peter Nørgaard, Daniel El Fassi, Lone Schejbel, Guy Wayne Novotny, Marie Fredslund Breinholt, Estrid Høgdall, Claudia Schöllkopf |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
0301 basic medicine
Male Myeloid Computer science Computational biology 03 medical and health sciences Data filtering 0302 clinical medicine Unknown Significance Fragment (logic) Genetics medicine Humans Myeloid Cells Pharmacology Diagnostic Tests Routine Computational Biology Genetic Variation High-Throughput Nucleotide Sequencing General Medicine Filter (signal processing) Ion semiconductor sequencing Sequence Analysis DNA Human genetics Repressor Proteins 030104 developmental biology medicine.anatomical_structure 030220 oncology & carcinogenesis Mutation (genetic algorithm) Mutation Molecular Medicine Female Algorithms |
| Zdroj: | Molecular diagnosistherapy. 25(2) |
| ISSN: | 1179-2000 |
| Popis: | Commercial myeloid next-generation sequencing (NGS) panels may facilitate uniform generation of raw data between laboratories. However, different strategies for data filtering and variant annotation may contribute to differences in variant detection and reporting. Here, we present how custom data filtering or the use of Oncomine extended data filtering improve detection of clinically relevant mutations with the Oncomine Myeloid Research Assay. The study included all patient samples (n = 264) analyzed during the first-year, single-site, clinical use of the Ion Torrent Oncomine Myeloid Research Assay. In data analysis, the default analysis filter was supplemented with our own data filtering algorithm in order to detect additional clinically relevant mutations. In addition, we developed a sensitive supplementary test for the ASXL1 c.1934dupG p.Gly646fs mutation by fragment analysis. Using our custom filter chain, we found 96 different reportable variants that were not detected by the default filter chain. Twenty-six of these were classified as variants of strong or potential clinical significance (tier I/tier II variants), and the custom filtering discovered otherwise undetected tier I/tier II variants in 25 of 132 patients with clinically relevant mutations (19%). The remaining 70 variants not detected by the default filter chain were classified as variants of unknown significance. Among these were several unique variants with possible pathogenic potential judged by bioinformatic predictions. The recently launched Oncomine 5.14 extended filter algorithm detects most but not all of the tier I/tier II variants that were not detected by the default filter. The supplementary fragment analysis for the ASXL1 c.1934dupG p.Gly646fs confidently detected a variant allele frequency of down to 4.8% (SD 0.83%). The assay also detected the ASXL1 c.1900_1922del23 mutation. Detection of clinically relevant variants with the Oncomine Myeloid Research NGS assay can be significantly improved by supplementing the default filter chain with custom data filtering or the recently launched Oncomine 5.14 extended filter algorithm. Our accessory fragment analysis facilitates easy testing for frequent ASXL1 mutations that are poorly or not covered by the NGS assay. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink