Prediction of selective estrogen receptor beta agonist using open data and machine learning approach.

BACKGROUND: Estrogen receptors (ERs) are nuclear transcription factors that are involved in the regulation of many complex physiological processes in humans. ERs have been validated as important drug targets for the treatment of various diseases, including breast cancer, ovarian cancer, osteoporosis, and cardiovascular disease. ERs have two subtypes, ER-α and ER-ß. Emerging data suggest that the development of subtype-selective ligands that specifically target ER-ß could be a more optimal approach to elicit beneficial estrogen-like activities and reduce side effects. METHODS: Herein, we focused on ER-ß and developed its in silico quantitative structure-activity relationship models using machine learning (ML) methods. RESULTS: The chemical structures and ER-ß bioactivity data were extracted from public chemogenomics databases. Four types of popular fingerprint generation methods including MACCS fingerprint, PubChem fingerprint, 2D atom pairs, and Chemistry Development Kit extended fingerprint were used as descriptors. Four ML methods including Naïve Bayesian classifier, k-nearest neighbor, random forest, and support vector machine were used to train the models. The range of classification accuracies was 77.10% to 88.34%, and the range of area under the ROC (receiver operating characteristic) curve values was 0.8151 to 0.9475, evaluated by the 5-fold cross-validation. Comparison analysis suggests that both the random forest and the support vector machine are superior for the classification of selective ER-ß agonists. Chemistry Development Kit extended fingerprints and MACCS fingerprint performed better in structural representation between active and inactive agonists. CONCLUSION: These results demonstrate that combining the fingerprint and ML approaches leads to robust ER-ß agonist prediction models, which are potentially applicable to the identification of selective ER-ß agonists.

[Investigation of chigger mites on the rat Eothenomys miletus in Yunnan].

OBJECTIVE: To understand the species, species distribution, the dominant species and their interspecies interaction of chigger mites on Eothenomys miletus(a dominant species of rats) in Yunnan. METHOD: The rats were captured with mouse traps in 16 counties (or cities) during 2000-2004. All mites on the surface of two auricles of the hosts were collected and identified. The patch index (m*/m) and the coefficient of association (V) were adopted to judge the spatial distribution patterns and interspecies interaction of the dominant chigger mite species among different individuals of the rats (E. miletus). RESULTS: 1157 individuals of E. miletus were captured from 16 counties (citys). 37613 chigger mites (belonging to 3 subfamily, 9 genus and 80 species) were collected from the auricles (body surface) of 1157 rat hosts with a high "overall mite infestation rate" (68.2%) and "overall mite index" (32.5). Six species of mites were found dominant on E. miletus: Leptotrombidium scutellare, Leptotrombidium sinicum, Helenicula simena, Leptotrombidium eothenomydis, Herpetacarus hastoclavus and Leptotrombidium hiemalis. The distribution of the chigger mites among different individuals of E.miletus showed an aggregation pattern. Both positive and negative association existed between each two dominant species of chigger mites. CONCLUSION: The species composition of chigger mites on Eothenomys miletus is complex with abundant individuals, which reflects a high species diversity of the mites. The main species of chigger mites tend to an aggregation on the body surface of E. miletus.