Search bioRxiv⌕ Search

Biology subjects

Caterer, Z.

Publications and source records attributed to Caterer, Z..

2 recordsLinked to original sources
bioRxiv2025

A graphical user interface for wrmXpress 2.0 streamlines helminth phenotypic screening

Image-based phenotypic screening is a fundamental technique used to better understand the basic biology of helminths and advance discovery of new anthelmintics. Miniaturization of screening platforms and automated microscopy have led to a surge in imaging data and necessitated software to organize and analyze these data. Traditionally, these analyses are performed remotely on high-performance computers, often requiring an understanding of a command-line interface (CLI) and the ability to write scripts to control the software or job scheduler. Requiring access to specialized computing equipment and advanced computational skills raises the barrier to entry for these sorts of studies. The development of efficient, performant computer and graphical processing units for personal computers and cheaper imaging solutions has made the requirement of remote servers superfluous for many small to medium-scale screens, but most analytical software still require interaction with a CLI. To democratize the analysis of image-based phenotypic screens, we have developed a graphical user interface (GUI) for wrmXpress, a tool that integrates many popular computational pipelines for analyzing imaging data of parasitic and free-living worms. The GUI operates on any personal computer using the operating systems native web browser, allowing users to configure and run analyses using a point-and-click approach. Containerization of the application eliminates the need to install specialized programming libraries and dependencies, further increasing the ease of use. GUI development required a substantial reorganization of the wrmXpress backend codebase, which allowed for the addition a new pipeline for high-resolution tracking of worm behavior, and we demonstrate its functionality by showing that praziquantel modulates the behavior of Schistosoma mansoni miracidia. These advances make cutting-edge analyses of image-based phenotyping of worms more equitable and accessible.

bioinformatics↗
bioRxiv2025

CellPhenoX: An eXplainable Cell-specific machine learning method to predict clinical Phenotypes using single-cell multi-omics

Single-cell technologies have enhanced our knowledge of molecular and cellular heterogeneity underlying disease. As the scale of single-cell datasets expands, linking cell-level phenotypic alterations with clinical outcomes becomes increasingly challenging. To address this, we introduce CellPhenoX, an eXplainable machine learning method to identify cell-specific phenotypes that influence clinical outcomes. CellPhenoX integrates classification models, explainable AI techniques, and a statistical framework to generate interpretable, cell-specific scores that uncover cell populations associated with relevant clinical phenotypes and interaction effects. We demonstrated the performance of CellPhenoX across diverse single-cell designs, including simulations, binary disease-control comparisons, and multi-class studies. Notably, CellPhenoX identified an activated monocyte phenotype in COVID-19, with expansion correlated with disease severity after adjusting for covariates and interactive effects. It also uncovered an inflammation-associated gradient in fibroblasts from ulcerative colitis. We anticipate that CellPhenoX holds the potential to detect clinically relevant phenotypic changes in single-cell data with multiple sources of variation, paving the way for translating single-cell findings into clinical impact.

bioinformatics↗