Integrated AI platform for causal discovery and drug validation

This technology is a suite of statistical and deep learning methods for the analysis of large-scale biomedical data with applications in drug testing, biomarker validation, and personalized medicine.

Unmet Need: Accurate, scalable statistical tool to analyze multimodal biomedical data

With the increased use of high-throughput assays, researchers are increasingly relying on advanced statistical and machine learning tools to facilitate the data analysis process. However, current approaches often struggle with accuracy, scalability, and integration of different types of data. Therefore, powerful tools with accurate and scalable statistical inference abilities are required to maximize output from research data.

The Technology: Cloud-based AI software for multimodal high-throughput data analysis

This technology is a cloud-based software platform containing a suite of statistical and deep learning methods for causal inference and translational biomedical applications. The platform includes tools such as MR-SPI & xMR, DeepMed, and ImmuneMirror, and enables the integration of structural biology data with genetic causal inference. This technology can analyze neoantigens for cancer immunotherapy and can enhance the process of drug target validation, biomarker discovery, and precision medicine.

Applications:

• Drug screening analysis tool
• Diagnostic support for cancer and other diseases
• Tool for developing personalized medicine
• Research tool for the study of genetic causes of various diseases
• Research tool for structural biology
• Research tool for drug development

Advantages:

• Better accuracy with less statistical bias
• Capable of integrating structural biology data with genetic causal inference frameworks
• Cloud-based software implemented through R and Python (easy to access)
• Scalable

Lead Inventor:

Zhonghua Liu, Sc.D.

Related Publications:

• Yao M, Miller GW, Vardarajan BN, Baccarelli AA, Guo Z, Liu Z. “Deciphering proteins in Alzheimer's disease: A new Mendelian randomization method integrated with AlphaFold3 for 3D structure prediction.” Cell Genom. 2024 Dec 11;4(12):100700.

• Chuwdhury GS, Guo Y, Chiang CL, Lam KO, Kam NW, Liu Z, Dai W. “ImmuneMirror: A machine learning-based integrative pipeline and web server for neoantigen prediction.” Brief Bioinform. 2024 Jan 22;25(2):bbae024.

• Xu S, Wang P, Fung WK, Liu Z. “A novel penalized inverse-variance weighted estimator for Mendelian randomization with applications to COVID-19 outcomes.” Biometrics. 2023 Sep;79(3):2184-2195.

• Yang J, Xu Y, Yao M, Wang G, Liu Z. “ERStruct: a fast Python package for inferring the number of top principal components from whole genome sequencing data.” BMC Bioinformatics. 2023 May 2;24(1):180.

• Tian P, Yao M, Huang T, Liu Z. “CoxMKF: a knockoff filter for high-dimensional mediation analysis with a survival outcome in epigenetic studies.” Bioinformatics. 2022 Nov 30;38(23):5229-5235.

Tech Ventures Reference:

• IR CU25378

• Licensing Contact: Joan Martinez