Artificial Immune Cell, AI-cell, a New Tool to Predict Interferon Production by Peripheral Blood Monocytes in Response to Nucleic Acid Nanoparticles

Abstract

Nucleic acid nanoparticles, or NANPs, rationally designed to communicate with the human immune system, can offer innovative therapeutic strategies to overcome the limitations of traditional nucleic acid therapies. Each set of NANPs is unique in their architectural parameters and physicochemical properties, which together with the type of delivery vehicles determine the kind and the magnitude of their immune response. Currently, there are no predictive tools that would reliably guide the design of NANPs to the desired immunological outcome, a step crucial for the success of personalized therapies. Through a systematic approach investigating physicochemical and immunological profiles of a comprehensive panel of various NANPs, the research team developes and experimentally validates a computational model based on the transformer architecture able to predict the immune activities of NANPs. It is anticipated that the freely accessible computational tool that is called an "artificial immune cell," or AI-cell, will aid in addressing the current critical public health challenges related to safety criteria of nucleic acid therapies in a timely manner and promote the development of novel biomedical tools.

Keywords: RNA nanotechnology; artificial intelligence; immunology; immunorecognition; machine learning; nucleic acid nanoparticles (NANPs).

Figure 1.

Conceptual representation of artificial immune cell (or AI-cell) tool. (A) The initial design and synthesis of nucleic acid nanoparticles (NANPs) is followed by their physicochemical characterization and assessment of immunostimulatory potential to then be applied for predictive computational analysis of the NANPs immune responses. (B) The experimental workflow used for the development of AI-cell.

Figure 2.

Representative NANPs chosen to collectively address the influence of their physicochemical properties and architectural parameters on their immunorecognition in human PBMC to further the development of AI-cell.

Figure 3.

Schematic representation of the quantitative structure–activity relationship (QSAR) methodology used in this project. (A) Modeling workflow: three machine learning approaches are evaluated using five-fold cross-validation (5-CV) repeated 10 times. (B) Overall workflow and the training procedure for prediction of nanoparticle sequence using transformer-based approach: tokenization, embedding followed by transformer modeling and prediction.

Figure 4.

Average performance (A) R² and (B) RMSE for different modeling approaches over 5-fold cross-validation and repeated 10-times. The error bar represents the standard deviation of the average performance over 5-folds cross-validation repeated 10-times (n=50). Detailed statistical analysis is shown in SI Table S5.