辣椒细菌性果实条斑病菌生物学特性及抑菌药剂筛选

研究辣椒细菌性果实条斑病菌的生物学特性,对病原菌进行室内药剂筛选,旨在为田间病害规律研究和病害防治提供依据。采用紫外分光光度计测定病原菌在不同温度、pH、NaCl浓度下的生长速率。采用滤纸片抑菌圈法对病原菌进行室内药剂筛选试验。结果表明,辣椒细菌性果实条斑病菌最适生长温度为28℃,最适pH 7.0,病原菌的耐盐性为12%。18种供试药剂在药剂推荐使用剂量范围内,四霉素、乙蒜素、77%志信、氯溴异氰尿酸4种药剂对病原菌有抑菌效果,对这4种药剂进行毒力测定,结果表明四霉素的毒力较强,EC₅₀值为11.75 mg/L,毒力强于其他3种试剂,四霉素可作为田间防治药剂的首选。     

松阳县古松树的松材线虫病现状调查与分析

古松树是历史的见证,是文化的承载,其保护工作具有重要意义,但因松材线虫的传播侵染,其生存状况不容乐观。本研究以松阳古松树为对象,基于分子检测技术,对其松材线虫病现状进行了调查与分析。结果表明,有12个乡镇近60%的古松树均已感病,疫情较为严重,需及时开展抗病保护工作。     

嗜线虫致病杆菌抗菌代谢产物Xcn1的研究进展

Xenocoumacin1（Xcn1）是嗜线虫致病杆菌产生的一种水溶性小肽类抗菌化合物，对真菌和细菌具有广谱的抑菌活性，在农业病害防治中具有良好的开发和应用前景。本文综述了Xcn1化合物的合成和降解机制、合成调控机制、抗菌作用机制以及防控植物疫病的研究进展，并展望了高产Xcn1生产菌株的改良和产业化应用前景。     

3株鸽源新城疫病毒的分子特征及对鸽的致病性

2011―2013年在新城疫流行病学调查中分离到3株鸽源新城疫病毒(SDS,SD01和SD02),为了进一步了解其生物学特性和遗传进化规律,对3株病毒进行了测序和生物活性分析,并对分离株SDS对鸽的致病性进行了评价。结果表明,毒株SDS基因组全长为15192 bp,基因排列方式为3′-NP-P-M-F-HN-L-5′,3个分离株F蛋白裂解位点氨基酸序列均为¹¹²RRQKRF¹¹⁷,具有典型的新城疫强毒的分子特征。系统进化分析表明这3个毒株与基因Ⅵ型NDV毒株聚于一簇,属于ClassⅡ系Ⅵb基因型。3个分离株F蛋白的融合肽和七肽重复区,HN蛋白的抗原中和表位存在多处突变,与单克隆抗体1E5、2F10的反应性也发生改变,表明3个分离株与疫苗株LaSota有明显的抗原差异。SDS攻毒试验结果显示,试验鸽自攻毒后5 d出现明显的临床症状,滴鼻组和肌肉注射组的死亡率分别为60%和70%;病死鸽剖检可见脑膜充血出血,腺胃乳头、腺胃肌胃交界及肠道出血,肝脏肿大,脾有淤血斑。滴鼻组在攻毒后5~14 d于喉头和泄殖腔检出排毒,而肌肉注射组在攻毒后3~14 d于喉头和泄殖腔有排毒。     

Screening and identification of soybean varieties with high phosphorus efficiency at seedling stage

In order to simplify the workload and find a reliable method for screening soybeans with different phosphorus(P)efficiencies,47 soybean varieties were screened from 90 varieties according to yield under normal phosphorus,and 10 indicators including root phenotype,phosphorus utilization efficiency at the seedling stage and yield were measured.Through single-index cluster analysis,the performance value and relative value of the abovementioned indexes under low-phosphorus conditions were analyzed,and then,the combined indexes were analyzed by principal components method.The membership function method and the cluster analysis method was used to calculate and analyze the comprehensive score value.Three indexes of root P utilization efficiency(X₁),relative value of root P utilization efficiency(X₂),and root volume(X₃)under low P at the seedling stage were selected as the most consistent with the yield screening results under low P condition,and the mathematical model of the comprehensive score value was obtained:D=1.218X₁t 0.320X₂t 0.007X₃-0.664(P=0.000,R²=1.000).The comprehensive score can screen and identify the P efficiency of soybeans with fewer indexes in the early growth stages,which provides a more rapid and reliable mathematical model for screening and identifying a larger number of germplasm resources for P efficiency.     

基于卷积神经网络马铃薯叶片病害识别和病斑检测*

针对自然环境下马铃薯叶片病害识别率低和晚疫病斑定位难的问题,基于大田环境中采集的马铃薯叶片图像,首先对马铃薯叶片病害进行识别,对比AlexNet、VGG16、InceptionV3、ResNet50、MobileNet五种神经网络模型,结果表明InceptionV3模型的识别效果准确率最高,可达98.00%。其次对马铃薯叶片的晚疫病斑进行检测,提出一种改进型的CenterNet-SPP模型,该模型通过特征提取网络获取对象的中心点,再通过中心点回归获得中心点偏移量、目标大小等图像信息,训练后的模型在验证集下的mAP可达90.03%,以F₁为评价值分析对比其它目标检测模型,CenterNet-SPP模型的效果最好,准确率为94.93%,召回率为90.34%,F₁值为92.58%,平均检测一张图像耗时0.10 s。为自然环境下马铃薯叶片病害识别和检测提供较为全面的深度学习算法和模型研究基础。     

基于卷积神经网络的玉米病害识别方法研究

为解决传统的玉米病害识别方法中特征提取主观性强及误识率高的问题,提出利用卷积神经网络对玉米病害进行识别。以玉米病害图像和健康图像共5种类别的玉米图像为研究对象,并采用LeNet模型进行试验。首先,按照8∶2的比例为每种玉米病害图像选择训练集和测试集。然后,通过试验组合和对比分析的方法比较不同卷积神经网络结构设置对准确率的影响,选出最佳参数。另外,选用Adam算法代替SGD算法来优化模型,通过指数衰减法调整学习率,将L2正则项添加到交叉熵函数中,并选择Dropout策略和ReLU激励函数。最后,确定了一个10层CNN网络结构。试验结果显示,玉米花叶病、灰斑病、锈病、叶斑病和玉米健康识别率分别为95.83%、90.57%、100%、93.75%、100%,平均识别率达96%,平均计算时间为0.15 s。经试验结果比较,该模型识别效果明显高于传统方法,为玉米病害的防治提供技术支持。     

Dynamics of Cotton Rhizosphere Bacterial Community Structure in Cotton Continuous Cropping Field Soil

[Objective] The aim of this study is to understand the dynamics of cotton rhizosphere bacterial community structure in cotton continuous cropping field soil. [Method] 16S rDNA genes were sequenced by high-throughput sequencing technology to determine the community structure of cotton rhizosphere bacteria in different developmental stages using an upland cotton cultivar (Gossypium hirsutum cv. TM-1). [Result] Four dominant phyla were found in the cotton rhizosphere bacterial community including Proteobacteria, Acidobacteria, Planctomycetes, and Bacteroidetes. The four dominant phyla and Firmicutes were largely influenced by cotton root. The relative abundances of Acidobacteria, Planctomycetes, Bacteroidetes were promoted, and Proteobacteria and Firmicutes were inhibited, by cotton root. There were significant differences in community structure, but not species richness or α-diversity among different developmental stages of the cotton rhizosphere bacterial community; the differences between the flowering stage and the budding stage were greater than the differences between the budding stage and the seedling stage. The diversity of the rhizosphere bacterial community in cotton continuous cropping field soil was significantly higher than that of the bulk soil; the β-diversity values of both the rhizosphere and bulk soil bacterial communities were highest in the flowering stage. [Conclusion] The structure and dynamics of the cotton rhizosphere bacterial community in cotton continuous cropping field soil was defined by high-throughput sequencing. The effect of cotton on the rhizosphere bacterial community structure was most significant in the flowering stage.