PhD position: Relevance-based Supervised Detection of Somatic Variants
Next-generation sequencing has become essential in molecular pathology. Panel sequencing is increasingly replaced by whole exome (WES) or whole genome (WGS) sequencing, often paired with RNAseq-based transcriptome analysis. While these advancements provide more data for personalized treatment, false positives and negatives in somatic lesion detection remain a challenge in precision oncology. Conventional analysis relies on unbiased genome-wide variant calling with stringent filtering to reduce false positives yet increasing the risk of false negatives. However, detecting clinically actionable (i.e. predictive, prognostic, diagnostic) variants, crucial for treatment recommendations, remains difficult due to technical (e.g. low coverage or low complexity regions) and biological (e.g. clonality or low expression levels) challenges.
The PhD student will develop novel methods to enhance the sensitivity of variant calling for the targeted detection of clinically relevant variants, complementing standard unbiased approaches. They will design a model to characterize variants or variant groups comprehensively, considering factors such as class, genomic position, size and create algorithms that leverage these characterizations to generate variant-specific workflows, directly analyzing WGS, WES, or RNAseq data. These workflows will integrate sequence analysis tools with optimized parameters, machine learning models trained on appropriately annotated datasets, and new statistics-based filtering strategies. The goal is to automate the creation of workflows for highly sensitive variant detection, a process currently requiring manual effort and advanced bioinformatics expertise.
The PhD position will be located in Beule Lab (BIH) and part of a collaboration with Leser Lab (HU) and Sers Lab (Charité), funded BIH PhD Program in collaboration with the compCancer graduate school. Please apply here by Feb 10th:
https://www.bihealth.org/en/notices/bih-phd-program-call-for-phd-candidates-2025
Last modified: Jan 10, 2025