Click-Through Rate Prediction Network Based on User Behavior Sequences and Feature Interactions

In recent years,deep learning has been widely applied in the fields of recommendation systems and click-through rate(CTR)prediction,and thus recommendation models incorporating deep learning have emerged.In addition,the design and implementation of recommendation models using information related to user behavior sequences is an important direction of current research in recommendation systems,and models calculate the likelihood of users clicking on target items based on their behavior sequence information.In order to explore the relationship between features,this paper improves and optimizes on the basis of deep interest network(DIN)proposed by Ali’s team.Based on the user behavioral sequences information,the attentional factorization machine(AFM)is integrated to obtain richer and more accurate behavioral sequence information.In addition,this paper designs a new way of calculating attention weights,which uses the relationship between the cosine similarity of any two vectors and the absolute value of their modal length difference to measure their relevance degree.Thus,a novel deep learning CTR prediction mode is proposed,that is,the CTR prediction network based on user behavior sequence and feature interactions deep interest and machines network(DIMN).We conduct extensive comparison experiments on three public datasets and one private music dataset,which are more recognized in the industry,and the results show that the DIMN obtains a better performance compared with the classical CTR prediction model.

Fast interactive segmentation algorithm of image sequences based on relative fuzzy connectedness

A fast interactive segmentation algorithm of image-sequences based on relative fuzzy connectedness is presented. In comparison with the original algorithm, the proposed one, with the same accuracy, accelerates the segmentation speed by three times for single image. Meanwhile, this fast segmentation algorithm is extended from single object to multiple objects and from single-image to image-sequences. Thus the segmentation of multiple objects from complex hackground and batch segmentation of image-sequences can be achieved. In addition, a post-processing scheme is incorporated in this algorithm, which extracts smooth edge with one-pixel-width for each segmented object. The experimental results illustrate that the proposed algorithm can obtain the object regions of interest from medical image or image-sequences as well as man-made images quickly and reliably with only a little interaction.

Identification of the interactive region by the homology of the sequence spectrum

The base sequence in genome was governed by some fundamental principles such as reverse-complement symmetry, multiple fractality and so on, and the analytical method of the genome structure, the “Sequence Spectrum Method (SSM)”, based on the structural features of genomic DNA faithfully visualized these principles. This paper reported that the sequence spectrum in SSM closely reflected the biological phenomena of protein and DNA, and SSM could identify the interactive region of protein-protein and DNA-protein uniformly. In order to investigate the effectiveness of SSM we analyzed the several protein-protein and DNA-protein interaction published primarily in the genome of Saccharomyces cerevisiae. The method proposed here was based on the homology of sequence spectrum, and it advantageously and surprisingly used only base sequence of genome and did not require any other information, even information about the amino-acid sequence of protein. Eventually it was concluded that the fundamental principles in genome governed not only the static base sequence but also the dynamic function of protein and DNA.

Single-cell RNA sequencing to understand host-virus interactions

Single-cell RNA sequencing(scRNA-seq)has allowed for the profiling of host and virus transcripts and host-virus interactions at single-cell resolution.This review summarizes the existing scRNA-seq technologies together with their strengths and weaknesses.The applications of scRNA-seq in various virological studies are discussed in depth,which broaden the understanding of the immune atlas,host-virus interactions,and immune repertoire.scRNA-seq can be widely used for virology in the near future to better understand the pathogenic mechanisms and discover more effective therapeutic strategies.

Protist Interactions and Seasonal Dynamics in the Coast of Yantai, Northern Yellow Sea of China as Revealed by Metabarcoding

Facilitated by the high-throughput sequencing(HTS)technique,the importance of protists to aquatic systems has been widely acknowledged in the last decade.However,information of protistan biotic interactions and seasonal dynamics is much less known in the coast ecosystem with intensive anthropic disturbance.In this study,year-round changes of protist community composition and diversity in the coastal water of Yantai,a city along the northern Yellow Sea in China,were investigated using HTS for the V4 region of 18S rDNA.The interactions among protist groups were also analyzed using the co-occurrence network.Data analyses showed that Alveolata,Chlorophyta,and Stramenopiles are the most dominant phytoplanktonic protists in the investigated coastal area.The community composition displayed strong seasonal variation.The abundant families Dino-Group-I-Clade-1 and Ulotrichales_X had higher proportions in spring and summer,while Bathycoccaceae exhibited higher ratios in autumn and winter.Alpha diversities(Shannon and Simpson)were the highest in autumn and the lowest in spring(ANOVA test,P<0.05).Nutrients(SiO42−,PO43−),total organic carbon(TOC),and pH seemed to drive the variation of alpha diversity,while temperature,PO43−and TON were the most significant factors influencing the whole protist community.Co-variance network analyses reveal frequent co-occurrence events among ciliates,chlorophytes and dinoflagellate,suggesting biotic interactions have been induced by predation,parasitism and mixotrophy.

Recent developments in application of single-cell RNA sequencing in the tumour immune microenvironment and cancer therapy

The advent of single-cell RNA sequencing(scRNA-seq)has provided insight into the tumour immune microenvironment(TIME).This review focuses on the application of scRNA-seq in investigation of the TIME.Over time,scRNA-seq methods have evolved,and components of the TIME have been deciphered with high resolution.In this review,we first introduced the principle of scRNA-seq and compared different sequencing approaches.Novel cell types in the TIME,a continuous transitional state,and mutual intercommunication among TIME components present potential targets for prognosis prediction and treatment in cancer.Thus,we concluded novel cell clusters of cancerassociated fibroblasts(CAFs),T cells,tumour-associated macrophages(TAMs)and dendritic cells(DCs)discovered after the application of scRNA-seq in TIME.We also proposed the development of TAMs and exhausted T cells,as well as the possible targets to interrupt the process.In addition,the therapeutic interventions based on cellular interactions in TIME were also summarized.For decades,quantification of the TIME components has been adopted in clinical practice to predict patient survival and response to therapy and is expected to play an important role in the precise treatment of cancer.Summarizing the current findings,we believe that advances in technology and wide application of single-cell analysis can lead to the discovery of novel perspectives on cancer therapy,which can subsequently be implemented in the clinic.Finally,we propose some future directions in the field of TIME studies that can be aided by scRNA-seq technology.

Tracking a maneuvering target in clutter with out-of-sequence measurements for airborne radar

There are many proposed optimal or suboptimal al- gorithms to update out-of-sequence measurement（s） （OoSM（s）） for linear-Gaussian systems, but few algorithms are dedicated to track a maneuvering target in clutter by using OoSMs. In order to address the nonlinear OoSMs obtained by the airborne radar located on a moving platform from a maneuvering target in clut- ter, an interacting multiple model probabilistic data association （IMMPDA） algorithm with the OoSM is developed. To be practical, the algorithm is based on the Earth-centered Earth-fixed （ECEF） coordinate system where it considers the effect of the platform＇s attitude and the curvature of the Earth. The proposed method is validated through the Monte Carlo test compared with the perfor- mance of the standard IMMPDA algorithm ignoring the OoSM, and the conclusions show that using the OoSM can improve the track- ing performance, and the shorter the lag step is, the greater degree the performance is improved, but when the lag step is large, the performance is not improved any more by using the OoSM, which can provide some references for engineering application.

赤眼鳟LGP2序列结构、组织表达及与MDA5互作特征

为探究赤眼鳟遗传学和生理学实验室蛋白2(laboratory of genetics and physiology 2,LGP2)的功能特征及抗草鱼呼肠孤病毒(grass carp reovirus,GCRV)育种参考潜力,实验克隆获得了2940 bp的赤眼鳟lgp2(Sclgp2)全长cDNA和721 bp的5′端上游序列。Sclgp2 cDNA编码680个氨基酸,包含DEXDc(DExD/H-box helicase domain)、HELICc(helicase superfamily C-terminal domain)和CTD(C-terminal regulatory domain)结构域;其5′端上游序列含有MafB(muscle aponeurosis fibromatosis B)和IRF3(interferon regulatory factor 3)等转录因子结合位点。不同物种LGP2的功能结构域、磷酸化修饰位点数具有相似性,同时也存在结构域排布位置及序列的差异。赤眼鳟和草鱼lgp2 cDNA序列比较初步发现2个位于RNA结合功能区的GCRV抗性关联位点。系统进化分析显示,赤眼鳟LGP2先与草鱼、鲫和青鱼聚在一起,再与鲤科鱼类等聚为一大支。荧光定量表达分析显示,赤眼鳟脾脏中sclgp2表达水平显著高于其他组织,肌肉、心脏中表达量次之,而肠中表达量最低。GCRV感染后,肝脏中ifn1表达水平在24~72 h显著下降,其他组织sclgp2和ifn1表达水平未有显著变化。相关性分析结果显示,赤眼鳟肌肉sclgp2与ifn1表达水平呈极显著正相关(0.999)。酵母双杂交互作检测发现,赤眼鳟LGP2与MDA5存在弱相互作用,而其DEXDc(1~201 aa)、HELICc(390~476 aa)以及CTD(553~668 aa)结构域与MDA5无互作。该研究成功获得了sclgp2全长cDNA及5′端上游序列,明确了其序列结构、免疫表达及与MDA5的互作特征,为赤眼鳟LGP2免疫功能属性研究奠定了基础,并为草鱼抗GCRV育种提供了参考。

智慧课堂能够促进课堂互动吗?——来自行为序列分析的经验证据

充分的课堂互动是激发学生学习主动性和自主性,培养学生高阶思维能力和提升教学质效的重要途径。以安徽省蚌埠市某高校2022—2023年度第一、二学期的核心专业课教学过程为对象进行准自然实验,通过互动行为编码和序列分析,探讨智慧课堂环境对师生互动行为的影响。研究显示:与传统多媒体课堂相比,智慧课堂中教师发起的主动谈话频率显著下降,教师的观察和指导行为以及技术操作和测试方面的活动频率显著增加;智慧课堂中的学生主动谈话和行动频率显著增加,被动行为频率显著减少;智慧课堂通过激发更丰富的学生主动行为和教师被动行为,有助于提升学生学习的主体性和积极性。

胶质母细胞瘤恶性进展中不同细胞亚群的动态轨迹和细胞通讯网络

目的:探索胶质母细胞瘤(glioblastoma,GBM)恶性进展过程中细胞亚群的动态轨迹以及免疫细胞亚群之间的通讯网络,结合GBM患者的转录组数据和临床信息,挖掘GBM恶性进展过程中的关键风险标志物,以期为该疾病的治疗和预后提供科学依据。方法:基于单细胞测序数据分析方法,构建GBM恶性进展中的细胞亚群图谱,利用Monocle2技术构建GBM恶性进展中肿瘤细胞亚群的动态进展轨迹,基于基因富集分析,挖掘肿瘤细胞亚群随GBM恶性进展中显著变化的基因所富集的生物学过程,利用CellChat软件识别不同免疫细胞亚群间的复杂通讯网络,通过生存分析识别GBM恶性进展中影响患者预后的关键风险分子标记物。结果:单细胞测序数据分析识别出6种不同的细胞类型,包括淋巴细胞、周细胞、少突神经胶质细胞、巨噬细胞、胶质瘤细胞、小胶质细胞,单细胞数据集中了27151个细胞,其中包含3881个来源于低级别胶质瘤患者的细胞,10166个来源于新诊断GBM患者的细胞,13104个来源于复发性胶质瘤患者的细胞。胶质瘤细胞亚群逆时序分析提示,胶质瘤细胞亚群在恶性进展中存在着明显的细胞异质性;免疫细胞亚群的细胞相互作用分析揭示,GBM恶性进展中不同免疫细胞亚群之间的通讯网络共识别出22条具有生物学意义的配体-受体对,涉及12条通路;生存分析识别出8个与GBM患者预后密切相关的基因,其中SERPINE1、COL6A1、SPP1、LTF、C1S、AEBP1、SAA1L是GBM患者的高风险基因,ABCC8是GBM患者的低风险基因。结论:深入揭示了GBM恶性进展中胶质瘤细胞亚群的动态变化以及免疫细胞亚群之间的通讯模式,对于理解GBM的复杂生物学过程具有重要意义,为GBM的精准医疗和治疗决策提供了科学依据,也为GBM患者更准确的预后评估提供了新的线索。

基于人-物交互关系检测的带电作业人员行为识别方法研究

为解决现有视频行为识别方法难以区分带电作业过程中某些相似行为、可识别行为种类少、未高效利用人员与物品间交互关系等问题,提出1种基于人-物交互关系检测的配网带电作业人员行为识别方法。利用轻量化姿态估计算法识别人员骨架序列,然后通过时空图卷积网络(spatial temporal graph convolutional networks,ST-GCN)提取人体运动的时空间特征并进行初步分类。对于由骨骼姿态无法有效区分的相似行为,采用目标检测算法识别人员所用工器具及使用状态,并通过融合人体动作与作业工器具所含行为信息,实现视频行为的精确识别。研究结果表明:该方法能有效识别带电作业行为,对相似行为的识别准确率约为88.9%,相较于现有基于骨架序列的带电作业人员行为方法提升约53个百分点。研究结果可为提高现场安全管控水平提供参考思路。

深度学习在基于序列的蛋白质互作预测中的应用进展

蛋白质-蛋白质相互作用在细胞信号转导、基因表达和代谢调控等生物过程中发挥重要作用,鉴定蛋白质间的相互作用对于理解复杂生物过程至关重要。预测蛋白质间的相互作用可以为药物发现、蛋白质功能研究和设计等领域提供帮助。近年来,随着人工智能技术的蓬勃发展,深度学习技术在预测蛋白质互作领域做出巨大贡献,其中基于序列的深度学习模型通过学习蛋白质序列信息的深层特征进行互作预测。本文综述了深度学习在基于序列的蛋白质互作预测中的应用,按照算法框架和时间线对该领域进展进行分类归纳,介绍了数据处理、序列编码方法、算法架构以及模型的评估指标等内容,并分析了当前面临的问题以及未来的发展方向。随着深度学习技术的发展,预测蛋白质互作的效率大幅提高,未来需要发展泛化能力更强的预测模型,助力蛋白质互作的预测。

落叶松-杨栅锈菌MlpMCM4蛋白和MlpHOG1蛋白互作关系初探

由落叶松-杨栅锈菌侵染引起的杨树叶锈病严重威胁杨树的健康生长。通过家族鉴定和系统发育分析确定落叶松-杨栅锈菌标准菌株ID 48743为MCM4直系同源基因,命名为MlpMCM4。以夏孢子cDNA为模板,运用RT-PCR技术,同源克隆获得中国菌株MlpMCM4基因CDS片段,称之为MlpMCM4(wh03),长度为2460 bp,编码819个氨基酸。结果表明,比对分析显示目的蛋白具有保守结构域,包括Walker A、Walker B和R-finger基序,以及MCM4类型锌指结构。亚细胞定位预测显示其定位在细胞核区域。利用基于分离泛素系统的酵母双杂交技术和萤火虫荧光素酶互补试验,没有检测到落叶松-杨栅锈菌中国菌株MlpMCM4蛋白和MlpHOG1蛋白能够相互作用,推断二者相互作用可能需要外源渗透压刺激。

基于表情向量聚类的课堂交互学生推荐方法

为了提升课堂提问环节交互学生的选择针对性,提出一种基于表情序列向量聚类的课堂交互学生个体推荐方法。首先,使用表情识别模型从学生端视频中获取表情向量,将每一名学生的所有表情向量汇总至教师端,并组合为表情向量序列;其次,基于学生表情向量序列提出基于短期表情向量聚类的学生分类方法和基于长期表情向量聚类的学生推荐方法;最后,通过可视化和表格形式为教师呈现学生的分类和推荐结果,为教师快速选择不同类型中最突出的学生个体提供支持。通过教师评价发现,所提方法能有效表现学生的状态差异,可显著提升课堂交互中学生的判别效率与提问目的性。

基于新奇度量的社交事件推荐方法

在社交事件网络(EBSN)中,推荐工作都是从用户的历史喜好出发建模用户偏好,阻碍了用户接触新事物的范围和途径。针对上述问题,提出基于新奇度量的社交事件推荐模型UER(Unexpectedness-based Event Recommendation)。UER模型包括Base和Unexpected两个子模型,首先,Base子模型基于用户、事件以及用户历史事件交互序列特征,通过注意力机制衡量事件在用户历史喜好中的权重,最终预测用户参加事件的概率;其次,Unexpected子模型通过自注意力机制提取用户的多个兴趣表示来计算用户自身新奇度和候选事件对用户的新奇值,从而衡量推荐事件的新奇程度。在Meetup-加州数据集上,UER模型相较于DIN(Deep Interest Network)和PURS(Personalized Unexpected Recommender System)的推荐命中率(HR)分别提高22.9%和30.3%,归一化折损累积收益(NDCG)分别提高27.5%和42.3%,推荐事件的新奇程度分别提高54.5%和21.4%;在Meetup-纽约数据集上,UER模型相较于DIN和PURS的HR分别提高18.2%和21.8%,NDCG分别提高26.9%和32.0%,推荐事件的新奇程度分别提高52.6%和20.8%。

Mutual regulation of microglia and astrocytes after Gas6 inhibits spinal cord injury

Invasive inflammation and excessive scar formation are the main reasons for the difficulty in repairing nervous tissue after spinal cord injury.Microglia and astrocytes play key roles in the spinal cord injury micro-environment and share a close interaction.However,the mechanisms involved remain unclear.In this study,we found that after spinal cord injury,resting microglia(M0)were polarized into pro-inflammatory phenotypes(MG1 and MG3),while resting astrocytes were polarized into reactive and scar-forming phenotypes.The expression of growth arrest-specific 6(Gas6)and its receptor Axl were significantly down-regulated in microglia and astrocytes after spinal cord injury.In vitro experiments showed that Gas6 had negative effects on the polarization of reactive astrocytes and pro-inflammatory microglia,and even inhibited the cross-regulation between them.We further demonstrated that Gas6 can inhibit the polarization of reactive astrocytes by suppressing the activation of the Yes-associated protein signaling pathway.This,in turn,inhibited the polarization of pro-inflammatory microglia by suppressing the activation of the nuclear factor-κB/p65 and Janus kinase/signal transducer and activator of transcription signaling pathways.In vivo experiments showed that Gas6 inhibited the polarization of pro-inflammatory microglia and reactive astrocytes in the injured spinal cord,thereby promoting tissue repair and motor function recovery.Overall,Gas6 may play a role in the treatment of spinal cord injury.It can inhibit the inflammatory pathway of microglia and polarization of astrocytes,attenuate the interaction between microglia and astrocytes in the inflammatory microenvironment,and thereby alleviate local inflammation and reduce scar formation in the spinal cord.

核酸条形码技术:扩展蛋白质-蛋白质相互作用检测通量的新方法

蛋白质-蛋白质相互作用(protein-protein interaction,PPI)几乎参与了机体内所有重要的生物学过程,在细胞的基本生命过程中扮演了至关重要的角色,开发高通量的PPI检测新方法具有重要的生物学意义。目前,下一代测序技术(next-generation sequencing,NGS)发展快速,能在几天内测定超过10亿个模板的DNA序列。由于并行DNA测序技术所特有的敏感性、特异性、高通量和多路复用优势,其已被用作广谱分子计数器,应用于基因组测序和转录物组测序等领域。核酸条形码技术通过将寡核苷酸标签与目标蛋白质连接起来,从而标记编码蛋白质。之后,利用高通量的测序方法检测相互作用的蛋白质,实现了PPI的高通量检测。这一技术推动了PPI检测方法的飞速发展,提升了单次实验检测的通量,为构建PPI网络提供了强有力的技术支持。本文详细阐述了核酸条形码在PPI检测方法中的设计、生成和读取;通过分析核酸条形码技术在PPI研究中的应用范例,探讨了各自的优势和不足,并评估了数据的可靠性,讨论了基于核酸条形码技术的PPI检测方法未来的发展趋势。

滨海复垦土地不同间混作模式增产及改良效应

[目的]通过研究高粱与野生大豆不同种植模式在滨海废弃盐田复垦土地上的作物增产及改良效应,探讨不同种植模式提高复垦土地生产力的机理及引起高粱根际土壤细菌菌群变化的主要环境因子,可为优化滨海废弃盐田复垦土地种植模式、种地养地相结合提供依据。[方法]在废弃盐田复垦形成的土地进行小区定位试验,设置高粱单作(S)、野生大豆单播(WS)、1行高粱1行野生大豆间作(S1WS1)、1行高粱2行野生大豆间作(S1WS2)、高粱与野大豆同行混播(SWS)共5个处理,以高粱生物量、产量及根际土壤为研究对象,采用高通量测序技术,解析不同种植模式根际土壤细菌的菌群变化,并对细菌群落结构与土壤环境因子进行冗余分析。[结果]与高粱单作(S)相比,高粱-野生大豆间混作S1WS1、S1WS2、SWS模式作物增产率分别为10.16%、13.15%、40.68%,作物增产率与土壤硝态氮、碱性磷酸酶呈显著正相关,土地当量比均大于1,体现了间混作优势;S1WS1、S1WS2、SWS模式提高了土壤有机质、氮磷养分含量、酶活性,降低了土壤pH和含盐量,SWS模式土壤改良效果最好;S1WS2、SWS模式提高了土壤细菌多样性,SWS模式细菌多样性指数最高,Ace、Chao、Shannon指数分别为S模式的1.12倍、1.12倍、1.04倍,细菌多样性指数与土壤硝态氮、脲酶呈显著正相关;放线菌门、变形菌门与绿弯菌门为丰度大于10%的优势菌门,土壤硝态氮、碱性磷酸酶、氮磷比为影响细菌群落结构的主要环境因子。S1WS1、S1WS2、SWS模式均提高了作物产量、土地当量比、土壤有机质和氮磷含量、酶活性。[结论]本试验条件下,相比于其它模式,SWS模式作物增产率与土地当量比均最大,最有利于发挥间混作优势;土壤有机质和氮磷含量均最高,土壤改良效果最好。

单细胞转录组测序技术在肝纤维化中的研究进展

肝纤维化是一种严重威胁人类健康的疾病,单细胞转录组测序技术为揭示其复杂的病理机制提供了全新途径。传统的研究方法在识别肝纤维化中不同细胞亚群及其基因表达变化方面存在局限,难以深入理解疾病机制。概述了单细胞转录组测序技术在肝纤维化过程中对不同细胞亚群类型的研究进展,单细胞RNA测序技术能够精确地解析不同细胞类型的基因表达及异质性,揭示肝纤维化过程中细胞亚群的动态变化及相关基因的表达,进而有助于理解各类细胞亚型在肝纤维化中的功能、相互作用及其对疾病进展的贡献。进一步探讨了该技术在肝纤维化研究中的重要意义与应用前景,通过这一技术,可以鉴定出与纤维化相关的关键基因和信号通路,为早期诊断、治疗靶点的发现以及新疗法的制定提供理论依据。此外,结合空间转录组测序技术,研究者可以在空间维度上观察细胞在组织中的分布,进一步提升对肝纤维化微环境的理解。该技术有助于深入理解肝纤维化的病理机制,为寻找新的治疗靶点和制定早期诊断及治疗策略提供了创新思路。

基于多场景序列匹配的视音频互动行为数据在线共享方法

视音频在多场景下共享时,由于过多干扰量导致匹配难度较大,因此提出一种基于单播流与多播流的互动行为数据在线共享算法来有效解决此问题。考虑到网络具有自我保护机制,本身就存在共享障碍,需采集历史数据建立安全保护函数,结合网络流量剩余定理计算不同大小和私密程度的视音频文本,记录每一次身份验证时流量的变化,设立验证门限。根据视音频互动行为特征,建立观测序列,按照特征值查找匹配度最高的场景,结合每次共享时视音频产生的资源增益,分别计算单播流和多播流形式下资源增益的对应带宽,生成适应度最高的共享序列。实验证明,所提方法对单播流和多播流形式的视音频数据均能实现精准共享,效率较好,共享代价很低,具有极高的实用价值。