Tandem repeats are hard to assess but important regions on the human genome. There are around 200 pathogenic and phenotypic tandem repeats, which are often ignored by standard analysis. Furthermore, these tandem repeats are utilized in crime scene investigations. Given the impact of these regions, it is important that we learn more about these regions of the human genome and their role in diseases and other traits.

In this presentation, Fritz will highlight our work to comprehensively characterize 1.7 million tandem repeat regions across the entire human genome. Utilizing unsupervised machine learning we identified interesting characteristics and clustering of these regions with implications into regulation and thus medical implications. To promote the usage of tandem repeats we further generated a novel GIAB benchmark and comparison methods. These enable for the first time a comparison across different technologies or samples at scale. To promote these activities further Fritz will present their tandem repeat caller that identifies variations together with epigenetic changes in tandem repeats leveraging PacBio sequencing data. All together this presentation brings the field of human genomic and genetics forward enabling the routine assessment of benign and pathogenic tandem repeats.

Learning Objectives: 

1. Review the complexity and variety of tandem repeats for human health.

2. Review new concepts on how to identify variants and methylation simultaneously in tandem repeats.

3. Summarize the benefits of long reads for the identification of variants.