Changes in the acoustic behavior of resident bottlenose dolphins near operating vessels

Maritime traffic is an issue of major ecological concern, and vessel noise may be an important source of disturbance for coastal cetaceans. In the Sado estuary, Portugal, core habitat areas of a small resident population of bottlenose dolphins (Tursiops truncatus) overlap with routes of intense maritime traffic, which presents an opportunity to assess vocal responses of these dolphins to specific vessel noise sources. Field recordings of dolphin vocalizations were made from April to November 2011, using a calibrated system. Dolphin behavior and group size were recorded, as well as the operating boat condition (no boats or specific boat type) in a 1,000 m radius. Spectral analyses of vocalizations allowed the categorization and quantitative analysis of echolocation click trains and social calls, including whistles. Mean overall call rates decreased significantly in the presence of operating vessels. Creaks (fast click trains) were significantly reduced in the presence of ferry boats. Significant differences were also observed in the whistles' minimum, maximum, and start frequencies. These changes in call emission rates and temporary shifts in whistles characteristics may be a vocal response to the proximity of operating vessels, facilitating communication in this busy, noisy estuary.     

免疫细胞浸润在非小细胞肺癌诊断与预后中的应用

免疫细胞浸润对癌症的诊断与预后有着重要意义。文中收集TCGA数据库已收录的非小细胞肺癌肿瘤与正常组织基因表达数据,利用CIBERSORT工具得到22种免疫细胞占比来评估免疫细胞浸润情况。以22种免疫细胞占比为特征,用机器学习方法构建了非小细胞肺癌肿瘤与正常组织的分类模型,其中随机森林方法构建的模型分类效果AUC=0.987、敏感性0.98及特异性0.84。并且用随机森林方法构建的肺腺癌和肺鳞癌肿瘤组织分类模型效果AUC=0.827、敏感性0.75及特异性0.77。用LASSO回归筛选22种免疫细胞特征,保留8种强相关特征组成的免疫细胞评分结合临床特征构建了非小细胞肺癌预后模型。经评估及验证,预后模型C-index=0.71并且3年和5年的校准曲线拟合良好,可以对预后风险度进行准确预测。本研究基于免疫细胞浸润所构建的分类模型与预后模型,旨在对非小细胞肺癌的诊断与预后研究提供新的策略。