Chromatin Organization & Epigenetics
Our lab investigates the role of chromatin organization in cancer. We have identified chromatin hubs as drivers of acute leukemia, uncovered MYB as a master regulator of super-hubs and discovered the presence of epi-clones (Molecular Cell 2025). We have characterized the remodeling of the 3D genome in patients with relapsed leukemia (Nature Communications 2024) and have shown that NSD2 leads to chromatin decompactions in relapse (Genome Biology 2023). We have developed C.Origami, a deep learning model that predicts cell-type-specific chromatin architecture using DNA sequence, chromatin accessibility and CTCF binding, and enables high-throughput screening for the discovery of putative regulatory elements (Nature Biotechnology 2023). Our previous studies revealed extensive remodeling of the 3D chromatin architecture landscape in patients with acute leukemia and demonstrated that small-molecule inhibitors targeting either oncogenic signal transduction or epigenetic regulation can modulate leukemia-specific 3D chromatin interactions (Nature Genetics 2020). We have shown that strong TAD boundaries are found near super-enhancers and are frequently tandemly duplicated in cancer patients (Nature Communications 2018). We have studied the importance of highly connected promoters and enhancers via 3D looping (promoter and enhancer hubs) during reprogramming (Nature Cell Biology 2019)) and differentiation (Nature Communications 2020). In earlier work, we identified LUNAR1, a leukemia-specific lncRNA that regulates IGF1R in the context of a TAD (Cell 2014). We have developed HiC-bench, a computational platform that enables comprehensive multi-tool multi-parameter analyses of Hi-C/HiChIP data and integration with other genomics data (BMC Genomics 2017). We also study alterations of histone modifications in leukemia that lead to the disruption of the normal epigenetic state. We first elucidated the role of the PRC2 complex as a tumor suppressor in acute T cell leukemia (Nature Medicine 2012). In a follow-up study, we delineated the role of the H3K27 demethylases JMJD3 and UTX in T-ALL. We showed that JMJD3 is essential for the initiation and maintenance of T-ALL, as it controls important oncogenic gene targets. By contrast, we found that UTX acts as a tumor suppressor and is frequently genetically inactivated in T-ALL (Nature 2014).
Machine Learning & Precision Medicine
Our team's mission is to revolutionize the understanding, diagnosis, and treatment of diseases through the innovative use of domain-specific expertise, big data and advanced AI/ML methods. Our research is centered around four major data modalities: Electronic Health Records (EHR) data, medical imaging data, multi-omic data, and mobile data from wearable devices. We leverage EHR data to gain comprehensive insights into patient health history. Our use of Medical Imaging data, including Radiology imaging data such as CT scans and MRIs, and Pathology imaging data like H&E slides (Nature Medicine 2018) and multi-plex imaging allows us to visualize and analyze disease patterns. We have developed a pre-trained Pathology models and characterized the derived histomorphological phenotypes using clinical and molecular data (Nature Communications 2024, Clinical Cancer Research 2024, NPJ Digital Medicine 2025, Nature Communications - in press). We have used a multi-channel masked autoencoder on imaging mass cytometry data to identify and characterize local tumor microenvironments (LTMEs) in lung cancer (Nature Biomedical Engineering 2025). We analyze multi-omic data, encompassing genetic, epigenetic, transcriptomic and proteomic data from diseased tissue, adjacent normal, and blood (Nature Communications 2023), as well as single-cell and/or spatial multi-omic data, to understand the complex interplay of various biological factors in disease progression. We develop machine learning models of gene regulation that can enable high-throughput in silico screens for regulatory elements and master regulators of disease (Nature Biotechnology 2023). Lastly, we utilize mobile data from wearable devices to monitor and analyze real-time health metrics. Our goal is to harness these diverse data streams to create a holistic, precise, and dynamic picture of health and disease, ultimately leading to more personalized and effective healthcare solutions. Finally, in collaboration with the NYU Molecular Pathology Lab, we are developing molecular assays for our cancer patients. For example, we have designed a cancer panel of 600+ genes to test for actionable and other common mutations in patients’ tumors. Our computational team has developed methods for validation, analysis and integration of genomic and clinical data. Our diagnostic test has received approval from the New York State Department of Health and by the FDA ("FDA Clears NYU Langone Genome PACT").
Applied Bioinformatics Laboratories
Dr. Tsirigos is the Director of the NYU Applied Bioinformatics Laboratories (ABL). Our mission is to accelerate scientific discoveries by guiding experimental design, performing robust data quality assessment, and carrying out comprehensive computational analyses. ABL offers computational services to the NYU Medical Center scientific community in the context of collaborations that span all departments and include both basic scientists and clinicians. ABL has also designed the computational pipelines and validation of the NYU NGS-580 cancer panel that will be soon offered to cancer patients.
