Abstract

Despite the variety in sequencing platforms, mappers, and variant callers, no single pipeline is optimal across the entire human genome. Therefore, developers, clinicians, and researchers need to make tradeoffs when designing pipelines for their application. Currently, assessing such tradeoffs relies on intuition about how a certain pipeline will perform in a given genomic context. We present Stratomod, which addresses this problem using an interpretable machine-learning classifier to predict variant calling errors in a data-driven manner. We showed Stratomod can precisely quantify the likelihood of missing variants using Hifi or Illumina, and leveraged Stratomods interpretability to measure contributions from difficult-to-map and homopolymer regions for each respective outcome. Furthermore, we used Statomod to assess the likelihood of missing variants due to mismapping using linear vs. graph-based references, and identified the hard-to-map regions where graph-based methods excelled and by how much. For these we utilized our new benchmark based on the Q100 HG002 assembly, which contains previously-inaccessible difficult regions. Furthermore, Stratomod presents a new method of finding likely false negatives, which is an improvement over current pipelines which only filter false positives. We anticipate this being useful for performing precise risk-reward analyses when designing variant calling pipelines.