Oana Carja

The paper introduces PULLI, a computer vision platform that extracts phenotypes from bacterial growth timelapses, revealing genotype-phenotype relationships in Vibrio cholerae mutants.

This study shows that elevated cfDNA and fragment shortening in cancer are due to saturated hepatic clearance, not tumor burden alone.

This paper maps gene drive dynamics onto Mendelian models to enable classical population genetics tools for predicting gene-drive outcomes.

This study shows that local negative frequency-dependent selection can reduce global coexistence in fragmented populations, contradicting its role in maintaining biodiversity in continuous landscapes.

This paper uses reinforcement learning to design adaptive antibiotic treatment strategies that overcome phenotypic heterogeneity in bacterial populations.

This study shows that heterogeneous population structures can increase evolutionary rates and alter genetic variation dynamics, challenging prior assumptions about structured populations slowing evolution.

This study shows that host population social stratification increases pathogen strain diversity by creating balancing selection pressures that support stable viral coexistence.

This study combines resource competition models with evolutionary graph theory to show that ecological competition can reverse the role of amplifier and suppressor networks for deleterious mutants.

This study models how spatially variable cell death in solid tumors affects ctDNA composition and clonal representation, showing potential biases in liquid biopsy profiles.

The paper introduces riboviz, a tool for analyzing and visualizing ribosome profiling datasets, providing both a backend pipeline and a web-based interface for comparison with public datasets.