Caroline E. Kostrzewa

This study shows that chromatin architecture and physical constriction together drive phenotype switching and metastasis in melanoma by altering CTCF positioning and heterochromatin structure.

This paper introduces BayesMAGPIE, a Bayesian method for identifying driver mutations in oncogenic pathways using mutual exclusivity, showing improved accuracy over MAGPIE in simulation and TCGA data.

The paper improves somatic mutation detection in tumor samples without matched normal samples by estimating tumor purity and variant allele frequency, reducing false positive and negative rates.

The paper introduces SpatialTopic, a Bayesian topic model for multiplexed imaging analysis that identifies spatial features and tracks their dynamics across disease progression.

CellGate automates spatial immunophenotyping of multiplexed tissue images with high accuracy, enabling scalable analysis of tumor-immune microenvironments.