Biomedical Data Research Lab
at Oregon Health & Science University
Research Interests
Understanding the mathematical representations of biological processes, relations, and functions in omics data and developing underlying analysis principles and mathematical theories.
Developing new systems biology models and AI frameworks to maximize the understanding of biological mechanisms in omics and multi-omics data.
Representation learning of high dimensional data, with a focus on linear/non-linear low rank and local low-rank representation of matrix and high order tensor.
Understanding biochemical variations in the microenvironment cancer and inflammatory diseases.
Developing explainable graphical/network models for biomedical data and transfer learning.
Study single cell and spatial multi-omics data to infer sample-wise/spatial dependent activity of transcriptional regulation, metabolism, and signaling pathways.
Biomarker prediction and development of personal wearable sensors and other smart health-related AI and biotechnologies.
Natural language processing-based mining of biological literature data.
Nutrient and food recommendations.
Our column brings you news and SOTA papers about GenAI, Computational Biology, AI4S, Quantum Computing, and many more!
Editor: Xin Lu, Pengtao Dang
Associated Editor: Chi Zhang, Sha Cao, Rosalie C. Sears, Yunlong Liu
News at a glance
Chi was invited by the NIH Innovation Lab to participate in a five-day close door discussions with other experts to explore innovative use of quantum computing in real world biomedical research problems. https://apply.knowinnovation.com/quantum-biomed/ 10/2024
Sha has been awarded the NIGMS R35 grant, to develop systems biology framework for dissecting the crosstalk of different cellular components in disease tissue microenvironment. 10/2024
Sha has been awarded the American Cancer Society Research Scholar Grant, which will support her research focusing on breast and pancreatic cancer. 10/2024
Congratulations to Chi on being awarded an R01 that uses AI-empowered wearable multimodal sensors for noninvasive monitoring of Parkinson’s disease. 09/2024
Sha presented a poster entitled “Computational methods to study metabolic variations in PDAC” at the AACR Pancreatic Cancer Meeting 2024 in Boston. 09/2024.
A collaborator work with Dr. Xinna Zhang’s group on “Inhibition of Glutamate-to-Glutathione Flux Promotes Tumor Antigen Presentation in Colorectal Cancer Cells” was accepted by Advanced Science. 09/2024.
The BDRL group officially launched at OHSU on July 1, 2024. We’re thrilled to embark on this next chapter in such a vibrant and supportive research environment.
Sha gave a talk on “Integrating multi-omics data for sparse latent space detection” at the ICSA China in Wuhan, as well as the ISGTM conference in Xi’An. 06/2024.
We announced that BDRL will move to Oregon Health and Science University (OHSU) in the fall semester of 2024. Chi Zhang and Sha Cao will be Associate Professors of Biomedical Engineering at OHSU. We will also join the Brenden-Colson Center for Pancreatic Care (BCC) group and the Center for Biomedical Data Science at OHSU Knight Cancer Institute. Multiple positions at the level of postdocs and data scientist will be opened soon. 05/2024
Jia Wang received a 10k Training Grant entitled "Computational Modeling of MHC-I antigen presentation flow in cancer cells" from IUSCCC. We appreciate the support from IUSCCC. 05/2024
Collaborative work "A highly reproducible and efficient method for retinal organoid differentiation from human pluripotent stem cells" has been accepted by PNAS, Congratulations! 05/2024
Changlin's paper "Ambiguities in neural-network-based hyperedge prediction" has been published by the Journal of Applied and Computational Topology, Congratulations! 05/2024
Xiao's paper "Bias-aware Boolean Matrix Factorization Using Disentangled Representation Learning" has been accepted by UAI 2024, Congratulations! 05/2024
BDRL Computational Biology Tools
Try our latest method to estimate metabolic flux or perform deconvolution via the Webservers
A webserver to estimate cell-/sample-wise metabolic fluxome by using scRNA-seq or general transcriptomics data
Predict the cell type PROPORTION and data specific cell type MARKER GENES
A semi-supervised approach for a robust identification of cell types and deconvolution of mouse transcriptomics data