AI-powered multi-scale modeling of microbiome-host interactions

点击翻译按钮获取中文摘要

Prediction of organismal characteristics based on genetic signatures of an individualorganism is a fundamental problem in biology. Recent advancements in the field indicate that microbiome, a community of microorganisms found in all environments on earth from humanbodies to soils, plays an essential role in determining the function of ecosystems. Microbiomegenerates a variety of metabolites that serve as messengers for the microbiome to communicatewith its surroundings. It remains unclear how these messengers communicate. This researchintends to decipher the molecular language of microbiome-environment interactions, therebyresolving a major enigma in life sciences. Understanding the fundamental rules governingmicrobiome-environment interactions will facilitate the solution of pressing problems inagriculture, environment, energy, and human health/well-being. By modulating the function of themicrobiome in plants, for instance, agriculture could become more productive while requiring lesspesticides and fertilizers. Utilizing the microbiome's photosynthetic energy is a promisingalternative clean energy solution. Maintaining gut microbiome stability may improve humanimmunity. This multidisciplinary research involving biology, chemistry, and computerscience will create opportunities for underrepresented students in life sciences to pursuecareers in computer science and bioinformatics, fields in which racial and ethnic minorities arewoefully underrepresented.Metabolite-protein interactions (MPIs) play a major mechanistic role in maintaining microbiomecommunities and mediating microbiome-host interactions. Exploration of the global landscape ofMPIs and the host gene expressions regulated by the MPI would fill critical knowledge gaps incausal environment-genotype-phenotype associations. However, our understanding of MPIs andtheir regulatory pathways is limited due to the transient and low-affinity nature of MPIs. Existingexperimental techniques for determining MPIs and their functional roles are time-consuming,biased to certain molecules, or applicable on a relatively small scale. The rapid growth of multipleomics data provides us with new opportunities to predict MPIs across entire microbiomes and hostgenomes. However, due to noisiness, biasness, incompleteness, and heterogeneity of omicsdata sets, their potential has not been fully appreciated for developing accurate, robust, andinterpretable predictive models for MPIs and their functions. The project will develop aninnovative AI-powered multi-scale modeling framework to predict biological functions ofmicrobiome metabolites by integrating diverse data from chemical genomics, structural genomics,and functional genomics. Specifically, this research will develop a novel end-to-end meta-learningmethod following a sequence-structure-function paradigm to predict genome-wide microbiomemetabolite-host protein interactions, and innovative domain adaption methods to translate cell linescreens to host systemic responses to microbiome metabolites by disentangling intrinsic biologicalsignals from both biological and technical confounders. An immediate outcome will be a bespokeplatform to infer novel biological relationships from noisy, biased, and incomplete data, whichwill have broad applications in addressing many fundamental biological problems. Another far-reachingbenefit will be the discovery of novel genetic, molecular, and cellular mechanisms ofbiological processes. Completing this project will provide a powerful computational framework tocorrelate molecular interactions to physiological processes, and establish causal environment microbiome-genotype-phenotype associations. The results of the project can be found athttps://github.com/XieResearchGroup/.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.