Senior Bioinformatics Engineer — GATK Variant Calling & Benchmark Optimization

Less than 30 hrs/week We are seeking a senior bioinformatics engineer for a focused, approximately two-week project to optimize short-read whole-genome variant calling against a custom benchmark. The work combines practical genomics, rigorous experimentation, and Python engineering. We will provide a local starter environment, benchmark samples, a scoring specification, supported parameter ranges, and a clear input/output interface. Commercial deployment details are confidential. What you will do - Set up the Docker environment and reproduce baseline GATK HaplotypeCaller results. - Investigate SNP/indel precision–recall trade-offs, with particular attention to indels and false positives. - Build a reproducible parameter-search and experiment-tracking workflow. - Validate selected configurations on untouched samples and report runtime and failure rates. - Deliver maintainable Python code, configurations, experiment results, documentation, and a handover session. Required experience - Hands-on GATK HaplotypeCaller and short-read WGS variant calling. - BAM/VCF/FASTA, reference builds, indexing, and variant representation. - hap.py or equivalent rigorous variant benchmarking; GIAB experience is strongly preferred. - Python, Linux, Docker, and reproducible scientific pipelines. - Experience preventing overfitting and making fair comparisons across genomic samples/regions. DeepVariant, BCFtools, and automated hyperparameter optimization are useful additional skills. Please include in your proposal - One relevant example of improving a variant-calling pipeline, ideally indel performance. - Your GATK and hap.py/GIAB experience. - How you would compare configurations without overfitting benchmark samples. - A relevant code sample or a brief walkthrough of a reproducible pipeline you built. This is a pipeline optimization and engineering project; it does not involve wet-lab work or clinical reporting.