Search

Additional file 2: Table S2. EuroGentest and the European Society of Human Genetics, guidelines for the evaluation and validation of NGS applications.

Additional file 1: Table S1. TruSight Cancer (Illumina) target genes in alphabetical order.

Additional file 3: Figure S1. Sequencing and Mapping Evaluation. a. Number of reads and clusters generated per run. b Number of unmapped reads per sample and alignment method.

Additional file 9. Detailed commands and parameters used for data pre-processing, sequence alignment, post-alignment processing and variant discovery.

Additional file 7: Figure S3. Box plot of Matthew Correlation coefficient (MCC) comparisons per alignment, variant calling or padding method. Each dot represents one observation and horizontal bold lines denote median MCC values. Boxes extend from the 25th to the 75th percentile of each group’s distribution of values. Vertical extending lines (whiskers) denote the upper and lower adjacent values. Statistical analyses were performed using the non-parametric Kruskal-Wallis or one-way ANOVA tests.

Additional file 8: Figure S4. Box plot comparisons of False Positive Rates (FPR). Only statistically significant differences are shown. Each dot represents one observation and horizontal bold lines denote median FPR values. Boxes extend from the 25th to the 75th percentile of each group’s distribution of values. Vertical extending lines (whiskers) denote the upper and lower adjacent values. Statistical analysis was performed using the non-parametric Kruskal-–Wallis test.

Additional file 4: Table S3. Number of reads per sample and mapping tool.

Additional file 5: Table S4. Minimum, mean and maximum depth of coverage, per gene and alignment algorithm. Depth of coverage results upon NextSeq 500 High-Output kit 2x75 cycles and 2x150 cycles sequencing.

Additional file 6: Figure S2. Hierarchical clustering of the tools. The d, MCC, p, r and F1 values were used to perform hierarchical clustering analysis based on the Lance–Williams agglomerative hierarchical clustering algorithm, which at each stage recomputes dissimilarities between clusters.

BESTCILIA Diagnostic Algorithm for PCD. (PDF 188 kb)

Technical Appendix. (DOCX 20 kb)

Secondary Analysis. (DOCX 169 kb)