Lauren Cole Peres

The paper introduces multiScaleAC, a method to analyze cell-cell interactions across spatial scales using a modified Moran's I with varying kernel bandwidths, showing improved statistical power and reduced bias.

The paper develops a functional data analysis framework to model spatial clustering in ovarian cancer, identifying prognostic T cell patterns beyond immune cell abundance.

The paper shows that bulk transcriptomic subtypes in HGSC reflect tumor composition rather than intrinsic biology, with composition predicting subtype and prognosis.

The paper deconvolves adipocyte proportions from bulk RNA-seq data of HGSOC tumors and finds higher adipocyte content is associated with worse survival.

The paper provides exact analytical solutions for the mean and variance of the Nearest Neighbor G(r) spatial metric under complete spatial randomness, avoiding computationally expensive permutations.

scSpatialSIM simulates spatial single-cell molecular data to evaluate spatial clustering methods, finding Ripleys K(r) best detects clustering across scenarios.

This study identifies molecular subtypes of high-grade serous ovarian cancer across racial groups and gene expression platforms, finding similar subtype-specific survival rates despite differences in subtype prevalence.