A5090004915- Cancer Genomics and Diagnostics
- Cancer Immunotherapy and Biomarkers
- Computational Drug Discovery Methods
- Single-cell and spatial transcriptomics
- Ferroptosis and cancer prognosis
- Radiomics and Machine Learning in Medical Imaging
- Gene expression and cancer classification
- CAR-T cell therapy research
- Molecular Biology Techniques and Applications
- Immune cells in cancer
- Gaussian Processes and Bayesian Inference
- RNA modifications and cancer
- Optimal Experimental Design Methods
- AI in cancer detection
- Cancer Cells and Metastasis
- Machine Learning in Bioinformatics
- Machine Learning and Algorithms
- Machine Learning and Data Classification
- Colorectal Cancer Treatments and Studies
- 3D Printing in Biomedical Research
- Pharmacogenetics and Drug Metabolism
- Gene Regulatory Network Analysis
- Non-Invasive Vital Sign Monitoring
- Cell Image Analysis Techniques
- HER2/EGFR in Cancer Research
National Institutes of Health
2022-2025
National Cancer Institute
2022-2025
Center for Cancer Research
2023-2024
University of Cambridge
2024
Discovery Institute
2024
Sanford Burnham Prebys Medical Discovery Institute
2024
Royal Marsden Hospital
2024
Royal Ottawa Mental Health Centre
2024
Weatherford College
2024
Samsung Medical Center
2023
Deep learning with Convolutional Neural Networks has shown great promise in various areas of image-based classification and enhancement but is often unsuitable for predictive modeling involving non-image based features or without spatial correlations. We present a novel approach representation high dimensional feature vector compact image form, termed REFINED (REpresentation Features as Images NEighborhood Dependencies), that conducible convolutional neural network deep learning. consider...
Abstract Drug sensitivity prediction for individual tumors is a significant challenge in personalized medicine. Current modeling approaches consider of single metric the drug response curve such as AUC or IC 50 . However, summary dose-response fails to provide entire profile which can be used design optimal dose patient. In this article, we assess problem predicting complete based on genetic characterizations. We propose an enhancement popular ensemble-based Random Forests approach that...
Precision oncology is increasingly becoming integral to clinical practice, showing notable improvements in treatment outcomes. While molecular data such as gene expression and methylation profiles offer comprehensive insights, obtaining this costly slow. Addressing challenge, here we developed Path2Omics, a deep learning model that predicts from histopathology. Path2Omics was trained on 20,497 slides (a combination of 9,456 formalin fixed (FFPE) 11,041 fresh frozen (FF)) 8,007 patients...
Abstract Introduction: Chimeric antigen receptor (CAR) T-cell therapy has transformed the treatment of hematological malignancies. However, its application to solid tumors remains limited by heterogeneity and on-target, off-tumor toxicities. One critical emerging approach combat this challenge is logic-gated multi-antigen targeting, for both enhanced safety efficacy. Methods: To address these challenges, we developed a novel genetic algorithm, termed LogiCAR, applied it analyze more than...
Chimeric antigen receptor (CAR) T-cell therapy has revolutionized the treatment of hematological malignancies. However, its application in solid tumors remains limited because single targets are unlikely to suffice due tumor heterogeneity and off-tumor toxicities. To overcome these obstacles, we developed LogiCAR designer, a computational approach that utilizes single-cell transcriptomics data from patient systematically identify cancer-specific circuits with logic gates ("AND," "OR," "NOT")...
Abstract Motivation: The immune-stromal microenvironment plays a crucial role in breast cancer precision oncology, offering insights that could guide targeted therapies. However, high-resolution methods like single-cell RNA sequencing remain prohibitively expensive and are rarely feasible clinical settings. While recent deconvolution can estimate cell type-specific expression from bulk (RNA-seq) data, even assays costly have long turnaround times, limiting their widespread use. To address...
In precision medicine, scarcity of suitable biological data often hinders the design an appropriate predictive model. this regard, large scale pharmacogenomics studies, like CCLE and GDSC hold promise to mitigate issue. However, one cannot directly employ from multiple sources together due existing distribution shift in data. One way solve problem is utilize transfer learning methodologies tailored fit specific context. paper, we present two novel approaches for incorporating information a...
Immune checkpoint blockade (ICB) is a promising cancer therapy; however, resistance frequently develops. To explore ICB mechanisms, we develop Immunotherapy Resistance cell-cell Interaction Scanner (IRIS), machine learning model aimed at identifying cell-type-specific tumor microenvironment ligand-receptor interactions relevant to resistance. Applying IRIS deconvolved transcriptomics data of the five largest melanoma cohorts, identify specific downregulated interactions, termed (RDI), as...
Ammonia is a cytotoxic molecule generated during normal cellular functions. Dysregulated ammonia metabolism, which evident in many chronic diseases such as liver cirrhosis, heart failure, and obstructive pulmonary disease, initiates hyperammonemic stress response tissues including skeletal muscle myotubes. Perturbations levels of specific regulatory molecules have been reported, but the global responses to hyperammonemia are unclear. In this study, we used multiomics approach vertically...
Emerging applications of radio frequency (RF) vision sensors for security and gesture recognition primarily target single individual scenarios which restricts potential applications. In this article, we present the design a cyber-physical framework that analyzes RF micro-Doppler signatures anomaly detection, such as hidden rifle among multiple individuals. avoids certain limitations video surveillance, recognizing concealed objects privacy concerns. Current RF-based approaches human activity...
Immune checkpoint blockade (ICB) is a promising cancer therapy; however, resistance often develops. To learn more about ICB mechanisms, we developed IRIS (
Recent years have observed a number of Pharmacogenomics databases being published that enable testing various predictive modeling techniques for personalized therapy applications. However, the consistencies between are usually limited in spite having significant common cell lines and drugs. In this article, we consider problem whether can use model learned from one secondary database to improve prediction other target database. We illustrate using two pharmacogenomics representing basis...
Abstract The aging of the immune system has profound implications for individual responses, yet precise quantification age remains a challenge. Analyzing single-cell peripheral blood mononuclear cells (PBMC) transcriptomics data from 981 healthy individuals, we developed IMMClock (IMMune Clock), human clock derived single cell transcriptomics. is first to accurately assess three major types, CD8+ T cells, CD4+ and NK at level. Testing ability these clocks capture dynamic process, show that...
Abstract The advancement of chimeric antigen receptor (CAR) T-cell therapy has been groundbreaking in the treatment hematological malignancies. However, its application solid tumors is constrained by variability and unintended off-tumor, on-target toxicities. Addressing these issues, we introduce a genetic algorithm to analyze single-cell transcriptomics data from patient tumors. This method pinpoints sets surface antigens exclusive cancer cells, utilizing logical operators "AND," "OR,"...
Abstract Introduction: The tumor microenvironment (TME) is a complex and dynamic ecosystem that plays critical roles in development clinical outcome. While the multifarious cellular interactions within TME have been extensively studied with significant emphasis on immunotherapy, understanding its role chemotherapy outcome remains less explored. To this end, we present novel generic computational framework named DECODEM (DEcoupling Cell-type-specific Outcomes using DEconvolution Machine...
Abstract Immune checkpoint blockade (ICB) is a promising cancer therapy; however, resistance often develops. To learn more about ICB mechanisms, we developed IRIS (Immunotherapy Resistance cell-cell Interaction Scanner), machine learning model aimed at identifying candidate ligand-receptor interactions (LRI) that are likely to mediate in the tumor microenvironment (TME). We and applied identify resistance-mediating cell-type-specific by analyzing deconvolved transcriptomics data of five...
The tumor microenvironment (TME) is a complex ecosystem of diverse cell types whose interactions govern growth and clinical outcome. While the TME's impact on immunotherapy has been extensively studied, its role in chemotherapy response remains less explored. To address this, we developed DECODEM ( DE coupling C ell-type-specific O utcomes using convolution M achine learning), generic computational framework leveraging cellular deconvolution
Abstract The aging of the immune system substantially impacts individual responses, yet accurately quantifying age remains a complex challenge. Here we developed IMMClock , novel clock that uses gene expression data to predict biological CD8⁺ T cells, CD4⁺ and NK cells. accuracy is first validated across multiple independent datasets, demonstrating its robustness. Second, utilizing IMMClock, find intrinsic cellular processes are more strongly altered during than differentiation processes....
ABSTRACT Circular extrachromosomal DNA (ecDNA) can drive tumor initiation, progression and resistance in some of the most aggressive cancers is emerging as a promising anti-cancer target. However, detection currently requires costly whole-genome sequencing (WGS) or labor-intensive cytogenetic FISH imaging, limiting its application routine clinical diagnosis. To overcome this, we developed ecPath ( ec from histo path ology ) , computational method for predicting ecDNA status routinely...
The majority of cancer drug sensitivity models are built utilizing genomic data measured before application to predict the steady state an applied drug. Restricting this type is limiting and can only explain one small piece puzzle. Better characterization cells be accomplished through use proteomic as more directly corresponds cellular activity. We have implemented that cell viability protein expression post application. These Random Forest, Elastic Net, Partial Least Square Regression...
<title>Abstract</title> The tumor microenvironment (TME) is a complex ecosystem of diverse cell types whose interactions govern growth and clinical outcome. While the TME’s impact on immunotherapy has been extensively studied, its role in chemotherapy response remains less explored. To address this, we developed DECODEM (DEcoupling Cell-type-specific Outcomes using DEconvolution Machine learning), generic computational framework leveraging cellular deconvolution <italic>bulk...