An XGBoost-Based Fitted Q Iteration for Finding the Optimal STI Strategies for HIV Patients.

The computational algorithm proposed in this article is an important step toward the development of computational tools that could help guide clinicians to personalize the management of human immunodeficiency virus (HIV) infection. In this article, an XGBoost-based fitted Q iteration algorithm is proposed for finding the optimal structured treatment interruption (STI) strategies for HIV patients. Using the XGBoost-based fitted Q iteration algorithm, we can obtain acceptable and optimal STI strategies with fewer training data, when compared with the extra-tree-based fitted Q iteration algorithm, deep Q-networks (DQNs), and proximal policy optimization (PPO) algorithm. In addition, the XGBoost-based fitted Q iteration algorithm is computationally more efficient than the extra-tree-based fitted Q iteration algorithm.

Non-diphtheriae Corynebacterium species are associated with decreased risk of pneumococcal colonization during infancy.

Streptococcus pneumoniae (pneumococcus) is a leading cause of severe infections among children and adults. Interactions between commensal microbes in the upper respiratory tract and S. pneumoniae are poorly described. In this study, we sought to identify interspecies interactions that modify the risk of S. pneumoniae colonization during infancy and to describe development of the upper respiratory microbiome during infancy in a sub-Saharan African setting. We collected nasopharyngeal swabs monthly (0-6 months of age) or bimonthly (6-12 months of age) from 179 mother-infant dyads in Botswana. We used 16S ribosomal RNA gene sequencing to characterize the nasopharyngeal microbiome and identified S. pneumoniae colonization using a species-specific PCR assay. We detect S. pneumoniae colonization in 144 (80%) infants at a median age of 71 days and identify a strong negative association between the relative abundance of the bacterial genera Corynebacterium within the infant nasopharyngeal microbiome and the risk of S. pneumoniae colonization. Using in vitro cultivation experiments, we demonstrate growth inhibition of S. pneumoniae by secreted factors from strains of several Corynebacterium species isolated from these infants. Finally, we demonstrate that antibiotic exposures and the winter season are associated with a decline in the relative abundance of Corynebacterium within the nasopharyngeal microbiome, while breastfeeding is associated with an increase in the Corynebacterium relative abundance. Our findings provide novel insights into the interspecies interactions that contribute to colonization resistance to S. pneumoniae and suggest that the nasopharyngeal microbiome may be a previously unrecognized mechanism by which environmental factors influence the risk of pneumococcal infections during childhood. Moreover, this work lays the foundation for future studies seeking to use targeted manipulation of the nasopharyngeal microbiome to prevent infections caused by S. pneumoniae.