The inference of associations between environmental factors and microbes and among microbes is critical to interpreting metagenomic data, but compositional bias, indirect associations resulting from common factors, and variance within metagenomic sequencing data limit the discovery of associations. To account for these problems, we propose metagenomic Lognormal-Dirichlet-Multinomial (mLDM), a hierarchical Bayesian model with sparsity constraints, to estimate absolute microbial abundance and simultaneously infer both conditionally dependent associations among microbes and direct associations between microbes and environmental factors. We empirically show the effectiveness of the mLDM model using synthetic data, data from the TARA Oceans project, and a colorectal cancer dataset. Finally, we apply mLDM to 16S sequencing data from the western English Channel and report several associations. Our model can be used on both natural environmental and human metagenomic datasets, promoting the understanding of associations in the microbial community.
Alopecia areata (AA) is a chronic inflammatory disease mediated by an array of cells and cytokines. Immunohistochemistry (IHC) of histological sections with antibodies to mast cell tryptase, CD4, CD8, CD1a and semi-quantitative real-time PCR analysis of Th1- and Th2-type cytokines were performed in 55 patients to investigate the infiltration features of mast cells (MCs), T lymphocytes and Langerhans cells (LCs) in scalp lesions of patients with AA. In AA patients, increased MCs mainly infiltrated the peri-follicular and peri-vascular areas, and correlated positively with numbers of CD8(+) T lymphocytes in deep peri-follicular areas (P = 0.04), but negatively with CD4(+) T lymphocytes in deep peri-vascular areas (P = 0.031). In patients with active hair loss, LCs in epidermis, deep dermis and peri-vascular were elevated (Ps < 0.05). Infiltration of LCs in upper peri-vascular areas and CD8(+) T cell infiltration in deep peri-follicular areas were positively correlated (R = 0.618, P = 0.011), as well as LCs in deep peri-vascular areas with CD8(+) T cells in upper peri-follicular areas (R = 0.570, P = 0.017). In patients with active hair loss, Th1-type cytokine (IL-2, IL-8, TNF-α) mRNA expression in deep dermis were higher than in upper dermis (Ps < 0.05). However, in patients with non-active hair loss, Th2-type cytokine (IL-5, IL-10) mRNA expression in deep dermis was higher than that in the upper dermis (Ps < 0.05). Positive correlations were found existing between MCs and CD8(+) T cells, as well as between LCs and CD8(+) T cells. In conclusion, findings in this study allow us to propose a close relationship between mast cells and CD8(+) T cells, as well as between LCs and CD8(+) T cells in AA, as well as allergy may interfere with infiltrating T lymphocytes in AA lesional regions. Also, Th1-type cytokine are related to disease activity of alopecia areata, whereas Th2-type cytokines may be associated with persistence of AA.
Objective: Venous thromboembolism (VTE) is a fatal complication and the most common preventable cause of death in hospitals. The risk-to-benefit ratio of thromboprophylaxis depends on the performance of the risk assessment model. A linear model, the Padua model, is recommended for medical inpatients in the United States but is not suitable for Chinese inpatients due to differences in race and disease spectrum. Currently, machine learning (ML) methods show advantages in modeling complex data patterns and have been applied to clinical data analysis. This study aimed to build VTE risk assessment ML models among Chinese inpatients and compare the predictive validity of the ML models with that of the Padua model. Methods: We used 376 patients, including 188 patients with VTE, to build a model and then evaluate the predictive validity of the model in a consecutive clinical dataset from Peking Union Medical College Hospital. Nine widely used ML methods were trained on the model derivation set and then compared with the Padua model. Results: Among the nine ML methods, random forest (RF), boosting-based methods, and logistic regression achieved a higher specificity, Youden index, positive predictive value, and area under the receiver operating characteristic curve than the Padua model on both the test and clinical validation sets. However, their sensitivities were inferior to that of the Padua model. Combined with the receiver operating characteristic curve, RF, as the best performing model, maintained high specificity with relatively better sensitivity and captured VTE patients' patterns more precisely. Conclusions: Advances in ML technology provide powerful tools for medical data analysis, and choosing models conforming to the disease pattern would achieve good performance. Popular ML models do not surpass the Padua model on all indicators of validity, and the drawback of low sensitivity should be improved upon in the future.
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.