Simulated test sets and evaluation results for in silico validation of a clinical genome sequencing pipeline

Data record of the article “An end-to-end in silico framework for truth-based validation and regression testing of clinical genome sequencing pipelines, from simulated reads to reportable findings” (Dvorský, Komárková, Klempíř, Kolinová, Krupička; Frontiers in Genetics, submitted). The record contains the truth and expected outcomes of ten simulated test sets (expert-curated ClinGen variants and synthetic novel variants planted in the GIAB HG002 background), the comparison of four configurations of nf-core/sarek with BIAS-2015 against this truth, a repeated baseline run, metrics with Wilson intervals, regression diffs, the fidelity comparison with real HG002 data, DeepVariant calls, and the ClinGen exports used for variant selection. The data were produced with insilico-clinval 1.0.0 (https://doi.org/10.5281/zenodo.23859473). Simulated reads and alignments are not included; they are regenerated byte-identically from the configuration and seed of each test set. README.md describes the structure and file formats, manifest.tsv lists every file with its SHA-256, and LICENSE.txt gives the terms of third-party content.