SPLDExtraTrees: robust machine learning approach for predicting kinase inhibitor resistance.

Drug resistance is a major threat to the global health and a significant concern throughout the clinical treatment of diseases and drug development. The mutation in proteins that is related to drug binding is a common cause for adaptive drug resistance. Therefore, quantitative estimations of how mutations would affect the interaction between a drug and the target protein would be of vital significance for the drug development and the clinical practice. Computational methods that rely on molecular dynamics simulations, Rosetta protocols, as well as machine learning methods have been proven to be capable of predicting ligand affinity changes upon protein mutation. However, the severely limited sample size and heavy noise induced overfitting and generalization issues have impeded wide adoption of machine learning for studying drug resistance. In this paper, we propose a robust machine learning method, termed SPLDExtraTrees, which can accurately predict ligand binding affinity changes upon protein mutation and identify resistance-causing mutations. Especially, the proposed method ranks training data following a specific scheme that starts with easy-to-learn samples and gradually incorporates harder and diverse samples into the training, and then iterates between sample weight recalculations and model updates. In addition, we calculate additional physics-based structural features to provide the machine learning model with the valuable domain knowledge on proteins for these data-limited predictive tasks. The experiments substantiate the capability of the proposed method for predicting kinase inhibitor resistance under three scenarios and achieve predictive accuracy comparable with that of molecular dynamics and Rosetta methods with much less computational costs.

Spatial analysis and evaluation of medical resource allocation in China based on geographic big data.

BACKGROUND: Spatial allocation of medical resources is closely related to people's health. Thus, it is important to evaluate the abundance of medical resources regionally and explore the spatial heterogeneity of medical resource allocation. METHODS: Using medical geographic big data, this study analyzed 369 Chinese cities and constructed a medical resource evaluation model based on the grading of medical institutions using the Delphi method. It evaluated China's medical resources at three levels (economic sectors, economic zones, and provinces) and discussed their spatial clustering patterns. Geographically weighted regression was used to explore the correlations between the evaluation results and population and gross domestic product (GDP). RESULTS: The spatial heterogeneity of medical resource allocation in China was significant, and the following general regularities were observed: 1) The abundance and balance of medical resources were typically better in the east than in the west, and in coastal areas compared to inland ones. 2) The average primacy ratio of medical resources in Chinese cities by province was 2.30. The spatial distribution of medical resources in the provinces was unbalanced, showing high concentrations in the primate cities. 3) The allocation of medical resources at the provincial level in China was summarized as following a single-growth pole pattern supplemented by bipolar circular allocation and balanced allocation patterns. The agglomeration patterns of medical resources in typical cities were categorized into single-center and balanced development patterns. GDP was highly correlated to the medical evaluation results, while demographic factors showed, low correlations. Large cities and their surrounding areas exhibited obvious response characteristics. CONCLUSIONS: These findings provide policy-relevant guidance for improving the spatial imbalance of medical resources, strengthening regional public health systems, and promoting government coordination efforts for medical resource allocation at different levels to improve the overall functioning of the medical and health service system and bolster its balanced and synergistic development.

Bioinformatics prediction of differential miRNAs in non-small cell lung cancer.

BACKGROUND: Non-small cell lung cancer (NSCLC) accounts for 85% of all lung cancers. The drug resistance of NSCLC has clinically increased. This study aimed to screen miRNAs associated with NSCLC using bioinformatics analysis. We hope that the screened miRNA can provide a research direction for the subsequent treatment of NSCLC. METHODS: We screened out the common miRNAs after compared the NSCLC-related genes in the TCGA database and GEO database. Selected miRNA was performed ROC analysis, survival analysis, and enrichment analysis (GO term and KEGG pathway). RESULTS: A total of 21 miRNAs were screened in the two databases. And they were all highly expressed in normal and low in cancerous tissues. Hsa-mir-30a was selected by ROC analysis and survival analysis. Enrichment analysis showed that the function of hsa-mir-30a is mainly related to cell cycle regulation and drug metabolism. CONCLUSION: Our study found that hsa-mir-30a was differentially expressed in NSCLC, and it mainly affected NSCLC by regulating the cell cycle and drug metabolism.