Machine learning methods accurately predict host specificity of coronaviruses based on spike sequences alone.

Coronaviruses infect many animals, including humans, due to interspecies transmission. Three of the known human coronaviruses: MERS, SARS-CoV-1, and SARS-CoV-2, the pathogen for the COVID-19 pandemic, cause severe disease. Improved methods to predict host specificity of coronaviruses will be valuable for identifying and controlling future outbreaks. The coronavirus S protein plays a key role in host specificity by attaching the virus to receptors on the cell membrane. We analyzed 1238 spike sequences for their host specificity. Spike sequences readily segregate in t-SNE embeddings into clusters of similar hosts and/or virus species. Machine learning with SVM, Logistic Regression, Decision Tree, Random Forest gave high average accuracies, F1 scores, sensitivities and specificities of 0.95-0.99. Importantly, sites identified by Decision Tree correspond to protein regions with known biological importance. These results demonstrate that spike sequences alone can be used to predict host specificity.

Arginase-1 is neither constitutively expressed in nor required for myeloid-derived suppressor cell-mediated inhibition of T-cell proliferation.

Although previous reports suggest that tumor-induced myeloid-derived suppressor cells (MDSC) inhibit T cells by L-arginine depletion through arginase-1 activity, we herein show that arginase-1 is neither inherently expressed in MDSC nor required for MDSC-mediated inhibition. Employing Percoll density gradients, large expansions of MDSC in the bone marrow of tumor-bearing mice were isolated and demonstrated potent inhibition in T-cell proliferation activated by TCR-ligation, Concanavalin A, PMA plus ionomycin, or IL-2. Despite demonstrating characteristic immunosuppressive capacity, these MDSC exhibit no arginase-1 expression and/or exert their inhibitory effects independent of arginase-1 activity. However, arginase-1 expression in MDSC can be induced by exposure to TCR-activated T cells or their culture medium, but not T cells activated by other means or growing tumor cells. Further investigation reveals multiple cytokines secreted by TCR-activated T cells as orchestrating two signaling-relay axes, IL-6-to-IL-4 and GM-CSF/IL-4-to-IL-10, leading to arginase-1 expression in MDSC. Specifically, IL-6 signaling increases IL-4R, enabling IL-4 to induce arginase-1 expression; similarly, GM-CSF in concert with IL-4 induces IL-10R, allowing IL-10-mediated induction. Surprisingly, our study indicates that induction of arginase-1 expression is not conducive to the critical MDSC-mediated inhibition toward T cells, which is rather dependent on direct cell contacts undiminished by PD-L1 blockade or SIRPα deficiency.