SFB1551 Project: R11

Project Title: Physically Interpretable Reconstruction of 3D Genome Architecture from Sparse Chromatin Contact Data

Summary: My project focuses on improving the quantitative interpretation of chromatin contact maps by identifying and correcting fundamental preprocessing biases in sparse 3D genome data. We further developed CCUT, a deep-learning framework that enables reconstruction of chromatin architecture while at the same time allowing to quantify the uncertainty of reconstruction quality, and facilitates direct comparison with polymer physics models.