Dana Raed Al-Hindi uses High-Performance Computing (HPC) resources to analyze large-scale genomic datasets, leveraging SLURM to efficiently manage computationally intensive tasks. Her work involves local ancestry deconvolution, merging data from multiple genotyping arrays post-imputation, applying rigorous quality control pipelines to minimize batch effects, and increasing power to detect novel variants. She also applies advanced tools such as GTeX, and FUMA to prioritize candidate genes for functional validation.

One of her key achievements enabled by HPC is deconvoluting local ancestry within a genetically diverse South African cohort (figure of karyogram below), and using association study methodologies to uncover genetic variants related to normal pigmentation variation – offering a deeper insights into the genetic architecture of skin pigmentation. Having conducted fieldwork in South Africa, she plans to return in the future to share her findings with local communities.

HPC has significantly enhanced her research and encourages other students to take full advantage of these resources on campus. She notes that analyses that would normally take a week on a local system, are completed in just two days with HPC resources. However, Dana also points out that HPC technologies would benefit from more user-friendly debugging tools to help researchers optimize job submissions and troubleshoot errors more efficiently. Academia often develops software that sometimes lack clear debugging outputs, which can make it difficult for researchers to resolve issues. When faced with challenges, Dana turns to her colleagues or consults HPC staff for assistance.

Highlighted Publication

Al-Hindi, D.R., Reynolds, A.W., & Henn, B.M. (2023). Genetic divergence within southern Africa during the Later Stone Age. In The Late Pleistocene Human Skull from Hofmeyr, South Africa (Vertebrate Paleobiology and Paleoanthropology, Stringer Series). https://doi.org/10.1007/978-3-031-07426-4_3