Tumorized precision dataset for ONCOLINER: A new solution for monitoring, improving, and harmonizing somatic variant calling across genomic oncology centers

The identification and characterization of the somatic genomic variation associated with the biology of tumors is one of the central pillars of Cancer Research and Personalized Medicine. The quality and the scope of the somatic variant calling determine the reliability and the impact of Cancer Genomic studies, as well as their potential downstream clinical applicability. But the overall quality, scope and consistency of the analysis of the somatic genome across different centers and studies remain significantly limited, affecting not only the overall outcome and the reach of cancer studies, but also the possibilities of improving discovery through the sharing and integration of datasets and results across centers. With the aim of providing users with actionable recommendations for the overall improvement and standardization of somatic variant identification strategies across research environments, we have developed ONCOLINER, an integrated platform with benchmarking data and tools for the detailed assessment, improvement and quality-based harmonization of analysis pipelines across centers. This will not only improve the overall efficiency of somatic variant identification globally, but it will also enable and guide emerging multi-center environments to share and integrate cancer datasets and results. ONCOLINER has been developed in the context of EUCANCan (https://eucancan.com/). These functionalities rely on an extensive catalog of somatic variants identified and validated in previous studies. To evaluate and improve the precision (specificity), we have generated two tumorized tumor-normal genome pairs, which consist of a controlled collection of in-silico-generated true somatic variants from the PCAWG Consensus Callsets inserted into real WGS samples from the GIAB project (https://www.nature.com/articles/sdata201625) (NA12878:HG001 and AshkenazimTrio:HG002). All CRAM files are aligned to GRCh37 (https://ftp.broadinstitute.org/pub/seq/references/Homo_sapiens_assembly19.fasta without scaffolds and supercontigs, only 1-22-X-Y-MT) using BWA-0.7.17 MEM. Reference: R. Martín et al., “ONCOLINER: A new solution for monitoring, improving, and harmonizing somatic variant calling across genomic oncology centers,” Cell Genomics, vol. 4, no. 9. Elsevier BV, p. 100639, Sep. 2024. doi: 10.1016/j.xgen.2024.100639