Abstract
The adaptive immune system of vertebrates can detect, respond to, and memorize diverse pathogens from past experience. While the clonal selection of T helper (Th) cells is the simple and established mechanism to better recognize new pathogens, the question that still remains unexplored is how the Th cells can acquire better ways to bias the responses of immune cells for eliminating pathogens more efficiently by translating the recognized antigen information into regulatory signals. In this work, we address this problem by associating the adaptive immune network organized by the Th cells with reinforcement learning (RL). By employing recent advancements of network-based RL, we show that the Th immune network can acquire the association between antigen patterns of and the effective responses to pathogens. Moreover, the clonal selection as well as other inter-cellular interactions are derived as a learning rule of the network. We also demonstrate that the stationary clone-size distribution after learning shares characteristic features with those observed experimentally. Our theoretical framework may contribute to revising and renewing our understanding of adaptive immunity as a learning system.
Highlights
The adaptive immunity of vertebrates is a complex adaptive system
We address this problem by associating the adaptive immune network organized by the T helper (Th) cells with reinforcement learning (RL)
Each T helper cell can be characterized by its T cell receptor (TCR) and the types of cytokines secreted, which roughly correspond to the phenotypic subtypes of the Th cells, e.g., Th1 and Th2 [2,33]
Summary
The adaptive immunity of vertebrates is a complex adaptive system. The system constantly adapts to intruding pathogens by orchestrating the populations and responses of diverse immune cells, each type of which can have distinct roles [1,2,3]. The problem that still remains unsolved is how Th cells are modulated by these different signals to recognize antigens and to induce and bias the activities of the groups of effector cells for evicting pathogens more efficiently by translating the recognized antigen information To this end, we revised the concept of immunological learning and bestowed it with a modern mathematical basis by focusing on the computational and algorithmic levels. The goal of the system may be to learn better ways from past experiences to bias the activities of the effector cells in response to infections, so as to evict the infected pathogens more promptly and We formulated this process as a reinforcement learning (RL) problem described using a Markov decision process (MDP) [27,28,29]. Our approach can complement more mechanistic investigations of the dynamics and regulation of immune responses by suggesting the functional roles of such dynamics at the computational and algorithmic levels
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
