Accurate prediction of antigen presentation by human leukocyte antigen (HLA) class II molecules would be valuable for vaccine development and cancer immunotherapies. Current computational methods trained on in vitro binding data are limited by insufficient training data and algorithmic constraints. Here we describe MARIA (major histocompatibility complex analysis with recurrent integrated architecture; https://maria.stanford.edu/), a multimodal recurrent neural network for predicting the likelihood of antigen presentation from a gene of interest in the context of specific HLA class II alleles. In addition to in vitro binding measurements, MARIA is trained on peptide HLA ligand sequences identified by mass spectrometry, expression levels of antigen genes and protease cleavage signatures. Because it leverages these diverse training data and our improved machine learning framework, MARIA (area under the curve = 0.89-0.92) outperformed existing methods in validation datasets. Across independent cancer neoantigen studies, peptides with high MARIA scores are more likely to elicit strong CD4 + T cell responses. MARIA allows identification of immunogenic epitopes in diverse cancers and autoimmune disease.
Highlights d Large-scale profiling of cell states & cellular ecosystems in hematologic malignancies d Atlas of malignant B cell states and 12 cell types in the DLBCL tumor microenvironment d Nine DLBCL cellular ecosystems & their relationships to molecular subtypes and survival d Candidate cellular biomarkers of response to bortezomib in DLBCL
Superparamagnetic iron oxide nanoparticles (SPION) are increasingly used to label human bone marrow stromal cells (BMSCs, also called “mesenchymal stem cells”) to monitor their fate by in vivo MRI, and by histology after Prussian blue (PB) staining. SPION-labeling appears to be safe as assessed by in vitro differentiation of BMSCs, however, we chose to resolve the question of the effect of labeling on maintaining the “stemness” of cells within the BMSC population in vivo. Assays performed include colony forming efficiency, CD146 expression, gene expression profiling, and the “gold standard” of evaluating bone and myelosupportive stroma formation in vivo in immuncompromised recipients. SPION-labeling did not alter these assays. Comparable abundant bone with adjoining host hematopoietic cells were seen in cohorts of mice that were implanted with SPION-labeled or unlabeled BMSCs. PB+ adipocytes were noted, demonstrating their donor origin, as well as PB+ pericytes, indicative of self-renewal of the stem cell in the BMSC population. This study confirms that SPION labeling does not alter the differentiation potential of the subset of stem cells within BMSCs.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.