Immunogram: comprehensive assessment of immune system through omics and causal inference 4181

Abstract Description How is my immune system? Answering this question is challenging because current methods assess only isolated components rather than providing a holistic view. This project introduces an innovative blood test for evaluating the immune system comprehensively through causal omics inference. Our approach utilizes whole-blood transcriptomics data primarily from public databases such as the Gene Expression Omnibus. To generate new data, we collect blood samples using PAXgene Blood RNA Tubes, extract RNA using the PAXgene Blood RNA Kit, deplete globin mRNAs with the GLOBINclear-Human Kit, and sequence the RNA on a NovaSeq X Plus platform, achieving 50 million paired-end reads of 150 base pairs. Alternatively, and preferably, we can use the Omics One transcriptomics platform. To interpret this data, we develop a causal model represented as a network of nodes that correspond to immunologically relevant elements, including cell types and proteins. By integrating omics data into this network, we perform causal inferences to refine node probabilities and compress the network into interpretable factors. These factors are synthesized into an Immunogram, which summarizes an individual’s immune status. We present Immunograms for various conditions, including autoimmune diseases, infections, and cancers, highlighting variations in immune responses to treatments. This innovative approach enhances our understanding of individual immune profiles and supports personalized therapeutic strategies. Funding Sources Kwat Medetgul-Ernar was supported by Stanford University Medical Scientist Training Program grant T32-GM007365. Topic Categories Computational and Systems Immunology (COMP)