基于记忆模拟退火和A*算法的农业机器人 遍历路径规划

【目的】解决农业机器人大田作业时遍历路径规划的问题。【方法】提出一种记忆模拟退火与A*算法相结合的遍历算法。首先通过记忆模拟退火算法搜索出任务最优目标点行走顺序,然后使用A*算法进行跨区域衔接路径规划。【结果】仿真试验结果表明,该算法规划的遍历路径曼哈顿距离比传统模拟退火算法减少了9.4%,遍历路径覆盖率能达到100%,重复率控制为4.2%。【结论】记忆模拟退火通过为传统模拟退火算法增加记忆器,增强了跳出局部最优陷阱的能力,提高了算法所得解的质量。该研究结果可为农业机器人遍历路径规划提供理论基础。     

H9N2禽流感病毒全基因组密码子使用偏好性及影响因素分析

【目的】探究H9N2禽流感病毒(AIV)全基因组的密码子使用偏好性及影响因素。【方法】选取2010—2018年国内H9N2 AIV流行毒株的全基因组为研究对象,分析其碱基组成特性、最优密码子、密码子使用偏好性的影响因素以及病毒对宿主密码子使用模式的适应性。【结果】H9N2 AIV的全基因组中AU含量高于GC。大部分最优密码子以A或U结尾,有效密码子数(ENC)平均值为52.86,提示存在密码子使用偏好性但偏好性较低。密码子使用偏好性主要受到突变压力和自然选择的共同作用,其中自然选择(所占比例为61.79%～76.15%)作用大于突变压力(所占比例为23.85%～38.21%)。H9N2 AIV对人Homo sapiens的密码子适应指数平均值为0.739～0.741,提示H9N2AIV禽流感病毒可能已适应人类的密码子使用模式。【结论】本研究为H9N2 AIV的基因进化分析、已有疫苗的密码子优化和新型疫苗(密码子去优化疫苗)研制提供了理论依据。     

基于WT-CNN-LSTM的溶解氧含量预测模型

溶解氧（Dissolved oxygen, DO）含量是影响水产养殖产量的重要因素之一，具有时序性、不稳定性和非线性等特点，且其影响因子过多、存在复杂的耦合关系，难以实现精准预测。针对传统长短时记忆神经网络（Long short term memory, LSTM）预测模型易引入冗余数据，且在训练过长序列时会出现梯度消失现象，从而不能捕捉因子间长期的依赖性问题，提出了基于小波变换（Wavelet transform, WT）、卷积神经网络（Convolutional neural network, CNN）和LSTM的溶解氧含量预测模型。首先，使用WT降低数据噪声；然后，使用CNN深度挖掘各变量之间的潜在关系；最后，利用LSTM的时序性预测2h后的水产养殖溶解氧含量。结果表明，本文提出的WT-CNN-LSTM模型预测效果良好，其平均绝对误差、均方根误差和决定系数分别为0.138、0.229和0.954，比传统LSTM模型分别优化了28.87%、21.03%和4.61%。     

基于部首嵌入和注意力机制的病虫害命名实体识别

为了解决农业病虫害命名实体识别过程中存在的内在语义信息缺失、局部上下文特征易被忽略和捕获长距离依赖能力不足等问题，以农业病虫害文本为研究对象，提出一种基于部首嵌入和注意力机制的农业病虫害命名实体识别模型(Chinese agricultural diseases and pests named entity recognition with joint radical embedding and self attention, RS-ADP)。首先，该模型将部首嵌入集成到字符嵌入中作为输入，用以丰富语义信息。其中，针对部首嵌入设计了3种特征提取策略，即卷积神经网络(Convolutional neural network, CNN)、双向长短时记忆网络(Bidirectional long short term memory network, BiLSTM) 和CNN-BiLSTM；其次，采用多层不同窗口尺寸的CNNs层提取不同尺度的局部上下文信息；然后，在BiLSTM提取全局序列特征的基础上，采用自注意力机制进一步增强模型提取更长距离依赖的能力；最后，采用条件随机场(Conditional random field, CRF)联合识别实体边界和划分实体类别。在包含11个类别和24715条标注样本的农业病虫害自制语料上进行了实验。结果表明，本文模型RS-ADP在该数据集上精确率、召回率和F1值分别为94.16%、94.47%和94.32%；在具体实体类别上，RS-ADP在作物、病害、虫害等易识别实体上F1值高达95.81%、97.76%和97.23%。同时，RS-ADP在草害、病原等难以识别实体上F1值仍保持86%以上。实验结果表明，本文所提模型能够有效识别农业病虫害命名实体，其识别精度优于其他模型，且具有一定的泛化性。     

低频声波频率下瘤背石磺CaM-like、CaMKII基因的表达及功能

长期栖息于潮间带的动物能够感知当地的潮汐规律,形成潮汐记忆的能力。本研究以栖息于潮间带的瘤背石磺(Onchidium reevesii)为对象,研究其通过感知潮汐来临时产生的低频声音来调节其行为与潮汐节律相协调的分子机制。通过RACE-PCR技术克隆到CaM-like基因的cDNA序列,并进行生物信息学分析和qRT-PCR实验。结果显示,瘤背石磺CaM-like基因的cDNA全长2321 bp,包括5′非编码区(UTR)366 bp,3′非编码区(UTR)1337 bp,618 bp的开放阅读框编码206个氨基酸;预测该基因编码的多肽链的原子数量是3165,分子量约为23029.64 kD,理论等电点4.64,分子式是C_(1018)H_(1544)N_(274)O_(320)S_9。N端信号肽由29个氨基酸组成。氨基酸序列比对构建系统进化树,结果显示瘤背石磺CaM-like基因与静水椎实螺(Lymnaea stagnalis)、光滑双脐螺(Biomphalaria glabrata)的CaM-like基因的亲缘关系最接近,这与传统形态学分类相吻合。在实验室内模拟潮汐中低频声音刺激瘤背石磺,应用qRT-PCR检测CaM-like基因在不同组织的分布情况,结果显示CaM-like mRNA在瘤背石磺不同组织均有表达,但神经节部位的表达量显著高于背部皮肤、腹足、肠、肝胰腺、口器和蛋白腺等组织(P0.05),推测其可能参与神经系统的可塑性调节。荧光定量结果显示CaM-like、CaMKII基因分别在25 Hz和50 Hz低频声波下刺激12.4 h表达量较高,初步推测其能感知25～50 Hz声波频率。这可能与栖息于潮间带中的瘤背石磺长期感知12.4 h半日潮潮汐周期节律,形成潮汐记忆有关。本研究将为进一步深入了解瘤背石磺感知潮汐节律的分子机制奠定基础,也为探究海洋动物由海洋向陆地进化过程中对环境的适应机制提供依据。     

Effect of pancreas disease caused by SAV 2 on protein and fat digestion in Atlantic salmon

Salmonid alphavirus (SAV) causes pancreas disease (PD) in farmed Atlantic salmon (Salmo salar L.), and exocrine pancreas tissue is a primary target of the virus. Digestive enzymes secreted by the exocrine pancreas break down macromolecules in feed into smaller molecules that can be absorbed. The effect of SAV infection on digestion has been poorly studied. In this study, longitudinal observations of PD outbreaks caused by SAV subtype 2 (SAV2) in Atlantic salmon at two commercial sea sites were performed. The development of PD was assessed by measurement of SAV2 RNA load and evaluation of histopathological lesions typical of PD. Reduced digestion of both protein and fat co‐varied with the severity of PD lesions and viral load. Also, the study found that during a PD outbreak, the pen population comprise several subpopulations, with different likelihoods of being sampled. The body length of sampled fish deviated from the expected increase or steady state over time, and the infection status in sampled fish deviated from the expected course of infection in the population. Both conditions indicate that disease status of the individual fish influenced the likelihood of being sampled, which may cause sampling bias in population studies.     

基于MFO-LSTM的母猪发情行为识别

及时准确识别母猪的发情行为可以有效增加受胎率和产仔量,对提高养殖企业的繁育水平和经济效益具有重要意义。该研究针对生猪养殖过程中母猪发情行为识别存在主观性强、智能化水平低、假警报和错误率高、识别不及时等问题,提出了一种基于飞蛾扑火算法(Moth-Flame Optimization,MFO)优化长短时记忆网络(Long Short Term Memory,LSTM)的母猪发情行为识别方法。利用安装在母猪颈部的姿态传感器获得母猪姿态数据,然后使用姿态数据训练MFO-LSTM姿态分类模型,将母猪姿态分为立姿、卧姿和爬跨3类。通过对姿态分类结果进行分析,确定以爬跨行为和活动量2个特征作为发情行为识别依据,使用MFO-LSTM分类算法判断母猪是否发情。以山西省太原市杏花岭区五丰养殖场的试验数据对该方法进行验证,结果表明,该方法在以30 min为发情行为识别时间时的识别效果最好,发情行为识别的错误率为13.43%,召回率为90.63%,特效性为81.63%,与已有的母猪发情行为识别方法相比错误率降低了80%以上。该方法在保证识别准确率的情况下有效降低了错误率,可满足母猪养殖生产过程中发情行为自动识别要求。     

Clinical feasibility of cognitive testing in dogs (Canis lupus familiaris)

Several cognitive tests have been developed to evaluate specific aspects of human and animal learning and memory. These tests have been used for early detection of cognitive deficits and to monitor the treatment of dogs with cognitive impairment. Thus, this article evaluated the feasibility of cognitive tests for use in canine neurology clinical routines and the suitability of the different tests to accomplish this aim. Fifteen healthy adult dogs were used for the cognitive tests of reward approach learning, object approach learning, object discrimination learning, reversal learning, delayed non-matched to position, and delayed non-matched to sample to assess different aspects of memory. No difference was observed between tests of delayed non-matched to position (3.13 ± 2.23 days) and delayed non-matched to sample (3.20 ± 2.40 days) (P = 0.944). However, dogs had greater difficulty in reversal learning (8.47 ± 2.61 days) than in object discrimination learning (4.60 ± 1.64 days) (P ≤ 0.001). Based on the tests performed, the delayed non-matched to position test may be performed in clinical routine if the owner and the veterinarian have time available, because this test is sensitive to evaluate dogs with cognitive impairments, but requires approximately 10 days of training. Thus, elderly dogs are excellent experimental models to study pathological aging based on their similarities with some human brain diseases, such as Alzheimer disease.     

Deficits of learning and memory in Hemojuvelin knockout mice

Iron is involved in various physiological processes of the human body to maintain normal functions. Abnormal iron accumulation in brain has been reported as a pathogenesis of several neurodegenerative disorders and cognitive impairments. Hemojuvelin (HVJ) is a membrane-bound and soluble protein in mammals that is responsible for the iron overload condition known as juvenile hemochromatosis. Although iron accumulation in brain has been related to neurodegenerative diseases, it remains unknown the effect of mutation of HVJ gene on cognitive performance. In our studies, HJV(−/−) mice showed deficits in novel object recognition and Morris water maze tests. Furthermore, the expression ration of apoptotic marker Bax and anti-apoptotic marker Bcl-2 in the hippocampus and prefrontal cortex showed higher levels in HJV(−/−) mice. Our results suggested that deletion of HJV gene could increase apoptosis in brain which might contribute to learning and memory deficits in mutant mice. These results indicated that HJV(−/−) mice would be a useful model to study cognitive impairment induced by iron overload in brain.