Ilya Shmulevich

This study identifies NF-κB-driven chromatin remodeling as a mechanism for drug-induced dedifferentiation in cancer, revealing its role in promoting drug-tolerant persister states.

This paper characterizes IDH+ glioma stem cell cultures to model GBM biology and therapeutic response, showing reproducible survival curves and methods to track cellular heterogeneity.

This study predicts microsatellite instability from whole slide images using texture features, achieving high AUC scores on both tile and sample levels.

The paper develops a multiscale mathematical model of immune surveillance in micrometastases to simulate patient trajectories and identify key parameters affecting immunosurveillance outcomes.

The paper describes the transformation of multiomics and EHR data into privacy-preserving biomedical knowledge graphs within the Biomedical Data Translator ecosystem.

This paper introduces diffusion-based multicellular reservoir computing as a framework for complex temporal signal processing, demonstrating its feasibility and computational advantages over single-cell systems.

SL-Cloud is a cloud-based resource that facilitates the prediction and comparison of context-specific synthetic lethal interactions using multi-omic datasets.

The paper develops a method to generate patient-specific cell communication networks from gene expression data and shows that specific communication edges are associated with disease progression in cancer.

The paper develops a Functional Module States framework to predict drug sensitivity by analyzing transcriptional states from gene expression data in breast cancer cell lines.

The paper identifies a six-state genetic program in Mycobacterium tuberculosis that enables entry into and exit from hypoxia-induced dormancy through high-resolution transcriptional time-course analysis.

ADAPTS improves cell type deconvolution in single-cell RNAseq data by customizing signature matrices and using hierarchical clustering to refine cell type identification.

The paper introduces LOBICO, a logic modeling approach for binary inputs predicting continuous outputs, showing it outperforms single-gene predictors in cancer drug response.

The paper introduces an empirical adaptation of Brown’s method to combine dependent P-values, showing it outperforms Fisher’s method on simulated and TCGA data.