Background: Infant health inequalities responsible for high infant sicknesses and deaths in our setting could depend to a large extend on maternal inequalities like socioeconomic class (SEC), age and human immunodefic...Background: Infant health inequalities responsible for high infant sicknesses and deaths in our setting could depend to a large extend on maternal inequalities like socioeconomic class (SEC), age and human immunodeficiency virus (HIV). Objective: To look at maternal inequalities (SEC, Age and HIV), to predict well-being of neonates during infancy. Methods: Subjects were selected using systematic random sampling. Maternal education, occupation, age and HIV status were obtained using a questionnaire;their SEC was derived using the Oyedeji’s model. Gestational age (GA) of the neonates was estimated from their mother’s last menstrual period, obstetric ultrasound scan reports or the Dubowitz criteria;and birthweight (BW) was determined using the basinet weighing scale, which has a sensitivity of 50 grams. Results: Ninety mother-neonatal pairs were enrolled, 47 (52.2%) neonates were males and 43 (47.8%) females. Most of the neonates were term 66 (73.3%) and of normal BW 75 (83.4%). A significant association existed between maternal variables and the likely hood of the subjects being less healthy during infancy (χ2 = 126.528, p < 0.005). Maternal age had a negative correlation coefficient with GA (r = -0.200) and BW (r = -0.115) and comparison of MA, GA and BW was significant (F = 2662.92, p < 0.0001). Conclusion: The combine effects of maternal SEC, Age and HIV have predicted less healthy neonates during infancy. Neonates in the present work are more prone to sicknesses and ill-health during infancy.展开更多
针对正常基因和异常基因在样本中的占比差异较大、变异断点位置难以准确确定的问题,提出了一种基于OCSVM(one-class support vector machine)的多策略融合拷贝数变异检测算法。算法融合读对深度、分裂读段和双端映射三种策略,建立多信...针对正常基因和异常基因在样本中的占比差异较大、变异断点位置难以准确确定的问题,提出了一种基于OCSVM(one-class support vector machine)的多策略融合拷贝数变异检测算法。算法融合读对深度、分裂读段和双端映射三种策略,建立多信号通道,并使用OCSVM模型解决正常基因和异常基因占比差异较大的影响以提高算法的拷贝数变异检测性能;对串联重复区域、穿插重复区域和缺失区域进行了分析探索,利用分裂读段信号实现变异点位置的精确定位,并确定变异类型。在240个模拟数据集和4个真实数据集上进行测试,并与其它几种算法进行比较。实验结果表明,该算法可以显著提高拷贝数变异检测的灵敏度、精度、F1评分以及重叠密度评分,同时减小了检测结果的边界偏差。展开更多
文摘Background: Infant health inequalities responsible for high infant sicknesses and deaths in our setting could depend to a large extend on maternal inequalities like socioeconomic class (SEC), age and human immunodeficiency virus (HIV). Objective: To look at maternal inequalities (SEC, Age and HIV), to predict well-being of neonates during infancy. Methods: Subjects were selected using systematic random sampling. Maternal education, occupation, age and HIV status were obtained using a questionnaire;their SEC was derived using the Oyedeji’s model. Gestational age (GA) of the neonates was estimated from their mother’s last menstrual period, obstetric ultrasound scan reports or the Dubowitz criteria;and birthweight (BW) was determined using the basinet weighing scale, which has a sensitivity of 50 grams. Results: Ninety mother-neonatal pairs were enrolled, 47 (52.2%) neonates were males and 43 (47.8%) females. Most of the neonates were term 66 (73.3%) and of normal BW 75 (83.4%). A significant association existed between maternal variables and the likely hood of the subjects being less healthy during infancy (χ2 = 126.528, p < 0.005). Maternal age had a negative correlation coefficient with GA (r = -0.200) and BW (r = -0.115) and comparison of MA, GA and BW was significant (F = 2662.92, p < 0.0001). Conclusion: The combine effects of maternal SEC, Age and HIV have predicted less healthy neonates during infancy. Neonates in the present work are more prone to sicknesses and ill-health during infancy.
文摘针对正常基因和异常基因在样本中的占比差异较大、变异断点位置难以准确确定的问题,提出了一种基于OCSVM(one-class support vector machine)的多策略融合拷贝数变异检测算法。算法融合读对深度、分裂读段和双端映射三种策略,建立多信号通道,并使用OCSVM模型解决正常基因和异常基因占比差异较大的影响以提高算法的拷贝数变异检测性能;对串联重复区域、穿插重复区域和缺失区域进行了分析探索,利用分裂读段信号实现变异点位置的精确定位,并确定变异类型。在240个模拟数据集和4个真实数据集上进行测试,并与其它几种算法进行比较。实验结果表明,该算法可以显著提高拷贝数变异检测的灵敏度、精度、F1评分以及重叠密度评分,同时减小了检测结果的边界偏差。
