Public articles linked to the same research event.
PLoS computational biology This study performed single-cell RNA sequencing of decidual tissue from normal pregnancies and cytogenetically normal recurrent pregnancy loss (RPL), combined genotype-based origin assignment with a hierarchical machine learning model (devCellPy) and a transformer-based foundation model (scGPT) for cross-architecture validation, found elevated decidual immune activation with maternal T cells carrying the most distinct and generalizable RPL-associated signatures, and converged network centrality analysis with origin-controlled expression filtering to nominate CXCR4 and JUN as candidate druggable targets.
This study performed single-cell RNA sequencing of decidual tissue from normal pregnancies and cytogenetically normal recurrent pregnancy loss (RPL), combined genotype-based origin assignment with a hierarchical machine learning model (devCellPy) and a transformer-based foundation model (scGPT) for cross-architecture validation, found elevated decidual immune activation with maternal T cells carrying the most distinct and generalizable RPL-associated signatures, and converged network centrality analysis with origin-controlled expression filtering to nominate CXCR4 and JUN as candidate druggable targets.
This study performed single-cell RNA sequencing of decidual tissue from normal pregnancies and cytogenetically normal recurrent pregnancy loss (RPL), combined genotype-based origin assignment with a hierarchical machine learning model (devCellPy) and a transformer-based foundation model (scGPT) for cross-architecture validation, found elevated decidual immune activation with maternal T cells carrying the most distinct and generalizable RPL-associated signatures, and converged network centrality analysis with origin-controlled expression filtering to nominate CXCR4 and JUN as candidate druggable targets.
This study performed single-cell RNA sequencing of decidual tissue from normal pregnancies and cytogenetically normal recurrent pregnancy loss (RPL), combined genotype-based origin assignment with a hierarchical machine learning model (devCellPy) and a transformer-based foundation model (scGPT) for cross-architecture validation, found elevated decidual immune activation with maternal T cells carrying the most distinct and generalizable RPL-associated signatures, and converged network centrality analysis with origin-controlled expression filtering to nominate CXCR4 and JUN as candidate druggable targets.