Julia Rymuza

The paper identifies acromegaly-related PIT-1/SF-1 positive pituitary tumors as a subtype of PIT-1 lineage tumors through transcriptomic analysis, showing they differ from gonadotroph tumors despite SF-1 activity.

The paper identifies copy number alterations in highly aneuploid tumors using genome-wide methods and highlights the bias from improper normal ploidy level setting in LPWGS and methylation arrays.

The paper introduces methods to construct and evaluate consensus genomic interval sets, showing how to minimize loss of resolution while enabling interoperability of interval data.

The paper introduces scEmbed, a framework for fast clustering and cell-type annotation of scATAC-seq data using pre-trained embeddings, showing improved performance and transferability.

The paper introduces four metrics to evaluate unsupervised embeddings of genomic regions and provides guidelines for reliable embedding learning.