Theodore Sakellaropoulos

This study shows that DNA damage in glioblastoma cells leads to long-term chromatin changes and gene expression reprogramming, affecting cellular function and tumor outcomes.

This study identifies HOXD13 as a driver of a developmental program in melanoma that promotes angiogenesis and suppresses T-cell infiltration through a gene regulatory axis involving VEGFA, SEMA3A, and CD73.

The paper investigates how CTCF mutations affect chromatin accessibility, cohesin binding, and gene regulation through multivariate modeling and mutant perturbations.

This paper uses self-supervised learning to identify histomorphological patterns in colon cancer WSIs, revealing clinically relevant clusters associated with survival and treatment responses.

This study identifies inflammation in tumor-adjacent lung tissue as a predictor of clinical outcome in lung adenocarcinoma using multi-omic analysis of 143 stage I cases.

This study introduces C.Origami, a deep learning model that predicts cell type-specific 3D chromatin organization, enabling in silico genetic screening to identify genomic regions affecting chromatin structure.

The paper predicts combination therapies for MS by modeling patient-specific signaling networks using phosphoproteomics data and identifies TAK1 as a co-druggable target.

This study shows that H3K27 acetylation retention during mitosis accelerates transcriptional and topological resetting of stem cell genes in mouse pluripotent stem cells upon G1 entry.

EpiMethylTag combines ATAC-seq or ChIP-seq with bisulfite conversion to simultaneously assess chromatin accessibility/TF binding and DNA methylation.

The paper shows NSD2 overexpression alters chromatin structure and gene expression by affecting CTCF and H3K27ac in specific chromatin domains.

The paper presents a method to qualitatively model signaling networks by integrating experimental data and protein connectivity, showing that senescent cells are less responsive to insulin and activate senescence pathways.

The paper improves Hi-C contact matrix reproducibility using fused two-dimensional lasso, identifies TAD boundaries with varying insulation strength, and shows super-enhancers are preferentially insulated by strong boundaries, which are frequently co-duplicated in cancer.