Due to inter-patient tumor heterogeneity, precision medicine is increasingly replacing the traditional one-size-fits-all approach in cancer treatment. Anticancer drug response prediction (DRP) has emerged as a critical area bridging basic and clinical research in personalized therapy, while also accelerating drug discovery for cancers. The availability of large-scale pharmacological datasets, combined with the success of deep learning in fields such as computer vision and natural language processing, has driven rapid advancements in deep learning-based DRP models. In this review, we discuss the deep learning based DRP models on a problem-driven basis and mainly focus on the multi-drug DRP models for monotherapy. This review systematically examines recent advances in deep learning-based multi-drug DRP models for monotherapy, detailing the methods employed at different modeling stages. We further discuss key challenges and future directions for developing high-performance DRP models, including generalized feature learning for both cancers and drugs, effective information fusion, interpretability, and adaptation from cell lines to patient data. Overall, this review offers valuable insights for biologists and bioinformaticians into current trends and future opportunities in DRP research.

Essential genes are necessary for the survival or reproduction of a living organism. The prediction and analysis of gene essentiality can advance our understanding of basic life and human diseases, and further boost the development of new drugs. We propose a snapshot ensemble deep neural network method, DeEPsnap, to predict human essential genes. DeEPsnap integrates the features derived from DNA and protein sequence data with the features extracted or learned from four types of functional data: gene ontology, protein complex, protein domain, and protein-protein interaction networks. More than 200 features from these biological data are extracted/learned which are integrated together to train a series of cost-sensitive deep neural networks. The proposed snapshot mechanism enables us to train multiple models without increasing extra training effort and cost. The experimental results of 10-fold cross-validation show that DeEPsnap can accurately predict human gene essentiality with an average AUROC of 96.16%, AUPRC of 93.83%, and accuracy of 92.36%. The comparative experiments show that DeEPsnap outperforms several popular traditional machine learning models and deep learning models, while all those models show promising performance using the features we created for DeEPsnap. We demonstrated that the proposed method, DeEPsnap, is effective for predicting human essential genes.

Essential genes are necessary for the survival or reproduction of a living organism. The prediction and analysis of gene essentiality can advance our understanding of basic life and human diseases, and further boost the development of new drugs. Wet lab methods for identifying cell essential genes are often costly, time-consuming, and laborious. As a complement, computational methods have been proposed to predict essential genes by integrating multiple biological data sources. Most of these methods are evaluated on model organisms. However, prediction methods for human essential genes are still limited and the relationship between human gene essentiality and different biological information still needs to be explored. In addition, exploring suitable deep learning techniques to overcome the limitations of traditional machine learning methods and improve prediction accuracy is also important and interesting. We propose a snapshot ensemble deep neural network method, DeEPsnap, to predict human essential genes. DeEPsnap integrates sequence features derived from DNA and protein sequence data with features extracted or learned from multiple types of functional data, such as gene ontology, protein complex, protein domain, and protein-protein interaction network. More than 200 features from these biological data are extracted/learned which are integrated together to train a series of cost-sensitive deep neural networks by utilizing multiple deep learning techniques. The proposed snapshot mechanism enables us to train multiple models without increasing extra training effort and cost. The experimental results of 10-fold cross-validation show that DeEPsnap can accurately predict human gene essentiality with an average AUROC (Area Under the Receiver Operating Characteristic curve) of 96.1%, the average AUPRC (Area under the Precision-Recall curve) of 93.82%, the average accuracy of 92.21%, and the average F1 measure about 80.62%. In addition, the comparison of experimental results shows that DeEPsnap outperforms several popular traditional machine learning models and deep learning models. We have demonstrated that the proposed method, DeEPsnap, is effective for predicting human essential genes.

ABSTRACTEssential genes are necessary to the survival or reproduction of a living organism. The prediction and analysis of gene essentiality can advance our understanding to basic life and human diseases, and further boost the development of new drugs. Wet lab methods for identifying essential genes are often costly, time consuming, and laborious. As a complement, computational methods have been proposed to predict essential genes by integrating multiple biological data sources. Most of these methods are evaluated on model organisms. However, prediction methods for human essential genes are still limited and the relationship between human gene essentiality and different biological information still needs to be explored. In addition, exploring suitable deep learning techniques to overcome the limitations of traditional machine learning methods and improve the prediction accuracy is also important and interesting. We propose a deep learning based method, DeepSF, to predict human essential genes. DeepSF integrates sequence features derived from DNA and protein sequence data with features extracted or learned from different types of functional data, such as gene ontology, protein complex, protein domain, and protein-protein interaction network. More than 200 features from these biological data are extracted/learned which are integrated together to train a cost-sensitive deep neural network by utilizing multiple deep leaning techniques. The experimental results of 10-fold cross validation show that DeepSF can accurately predict human gene essentiality with an average AUC of 95.17%, the area under precision-recall curve (auPRC) of 92.21%, the accuracy of 91.59%, and the F1 measure about 78.71%. In addition, the comparison experimental results show that DeepSF significantly outperforms several popular traditional machine learning models (SVM, Random Forest, and Adaboost), and performs slightly better than a recent deep learning model (DeepHE). We have demonstrated that the proposed method, DeepSF, is effective for predicting human essential genes. Deep learning techniques are promising at both feature learning and classification levels for the task of essential gene prediction.

Accurately predicting essential genes using computational methods can greatly reduce the effort in finding them via wet experiments at both time and resource scales, and further accelerate the process of drug discovery. Several computational methods have been proposed for predicting essential genes in model organisms by integrating multiple biological data sources either via centrality measures or machine learning based methods. However, the methods aiming to predict human essential genes are still limited and the performance still need improve. In addition, most of the machine learning based essential gene prediction methods are lack of skills to handle the imbalanced learning issue inherent in the essential gene prediction problem, which might be one factor affecting their performance. We propose a deep learning based method, DeepHE, to predict human essential genes by integrating features derived from sequence data and protein-protein interaction (PPI) network. A deep learning based network embedding method is utilized to automatically learn features from PPI network. In addition, 89 sequence features were derived from DNA sequence and protein sequence for each gene. These two types of features are integrated to train a multilayer neural network. A cost-sensitive technique is used to address the imbalanced learning problem when training the deep neural network. The experimental results for predicting human essential genes show that our proposed method, DeepHE, can accurately predict human gene essentiality with an average performance of AUC higher than 94%, the area under precision-recall curve (AP) higher than 90%, and the accuracy higher than 90%. We also compare DeepHE with several widely used traditional machine learning models (SVM, Naïve Bayes, Random Forest, and Adaboost) using the same features and utilizing the same cost-sensitive technique to against the imbalanced learning issue. The experimental results show that DeepHE significantly outperforms the compared machine learning models. We have demonstrated that human essential genes can be accurately predicted by designing effective machine learning algorithm and integrating representative features captured from available biological data. The proposed deep learning framework is effective for such task.

为了全面分析沥青路面表面构造的分布特性,探索快速、准确的沥青路面表面构造分布的多特征参数表征方法,利用MATLAB软件数字图像处理技术获取AC、SMA及OGFC等6种不同的沥青道路表面构造的二值图像,验证了表面构造分布的多重分形特性,描述了表面构造分布多重分形谱参数的物理意义,分析了不同类型沥青路面的谱参数的变化规律.研究结果表明:多重分形谱参数能从不同角度描述表面构造分布的多重特征;多重分形谱峰值fmax (α)为表面下凹构造的分形维数D0,峰值fmax(α)越大,表面也就越粗糙;谱宽△α整体反映了表面构造分布的不均匀程度,沥青道路表面构造粒径越大、级配越粗,其表面构造分布往往越不均匀,△α也越大;同时,最大、最小奇异性标度指数的分形维数之差△f也从下凹区域集中角度局部反映了表面构造分布不均匀程度,受集料粒径尺寸是否连续影响较大.沥青道路表面构造分布的多重分形特性能为分析表面离析状况及抗滑性能等提供一种新的方法与思路.

Identifying essential proteins is very important for understanding the minimal requirements of cellular life and finding human disease genes as well as potential drug targets. Experimental methods for identifying essential proteins are often costly, time-consuming, and laborious. Many computational methods for such task have been proposed based on the topological properties of protein-protein interaction networks (PINs). However, most of these methods have limited prediction accuracy due to the noisy and incomplete natures of PINs and the fact that protein essentiality may relate to multiple biological factors. In this work, we proposed a new centrality measure, OGN, by integrating orthologous information, gene expressions, and PINs together. OGN determines a protein's essentiality by capturing its co-clustering and co-expression properties, as well as its conservation in the evolution process. The performance of OGN was tested on the species of Saccharomyces cerevisiae. Compared with several published centrality measures, OGN achieves higher prediction accuracy in both working alone and ensemble.

To reveal the characteristics and polishing behaviors of the surface texture on asphalt pave-ment,the multifractal characteristics of the three-dimensional(3D)morphology distribution of the sur-face texture surface were verified based on the multifractal theory.Then,the polishing process of the surface texture of AC-13 limestone mixture was simulated at 0 to 1×105polishing cycles at the low speed.The changes of the multifractal spectrum parameters of the surface texture in the polished area were analyzed.The global and local feature changes of the surface texture before and after polishing were qualified.The results show that the 3D morphology of the asphalt pavement surface has multifrac-tal characteristics,and the multifractal spectrum width Δαqualifies the global characteristics of the rel-ative steepness difference.The height difference between the left and the right spectrum arms Δf cap-tures the dominance degree of the steepest fragment.With the increase of the polishing cycles, the fractal dimension D changes little, while Δαdecreases obviously with the polishing action and the dominance degree of the surface flatness fragments gradually tends to be stable.Thus,the multifractal characteristics provide some new ideas to reflect the polishing behaviors of the surface texture.

Surface segregation of asphalt mixtures is a common problem encountered when determining segregation level, which is a criterion of pavement quality evaluation. A method to evaluate segregation levels of asphalt pavement surface was presented based on the concave multifractal characteristic in a binary image of a pavement surface, which was obtained by digital image processing technology and mathematical morphology. The practical value of the proposed method was verified in a newly built asphalt pavement, and the segregation level was divided into five sections based on the recommended texture ratio in practical engineering. Results show that the multifractal spectrum width (Delta alpha) quantifies the uniformity of the concave distribution. The pixel percentage of the concave (P) characterizes the ratio of the occupied area. The product of Delta alpha and P (PWP) was quantified as the surface segregation level, which has good linear relevance with texture depth evaluation results. The proposed evaluation technique (PWP) can be used as an alternative to the sand patch method.

To study the relevance between asphalt pavement skid resistance performance and asphalt mixture volume indexes,and reveal the influence degrees of different volume indexes on the skid resistance performance,the density experiments of different AC-16 asphalt mixtures were conducted by vacuum method and plastic encapsulation method,and the volume indexes such as volumetric voidage (VV),void in mineral aggregate (VMA),void filled with asphalt (VFA) and void in coarse aggregate (VCA) were compared.Indoor polishing test and pendulum test on asphalt mixtures were conducted,asymptotic model was used to fit the skid resistance's decay trend,and the parameters such as the initial value,steady value and damping value of skid resistance were obtained.The function relationships between the volume indexes and the skid resistance performance were built,and the grey correlation degree ranks between the different volume indexes and the skid resistance performance were analyzed by grey correlation theory.Research result indicates that there are some differences between the influences of different volume indexes on the skid resistance performance.The skid resistance performance of asphalt mixture increases with the increase of VV and VMA,and the decrease of VFA and VCA.Grey correlation degree descending order is VV,VMA,VFA and VCA,which means that VV is the main influence factor of skid resistance performance,the influence of VMA on skid resistance performance is significant,however,the influence degrees of VFA and VCA are not obvious.In the design and construction process,the asphalt pavement skid resistance performance can be improved by controlling mixture voidage and adjusting the dense state and compact condition of mixture.10 tabs,9 figs,25 refs.

Texture roughness of 6 aggregates (diorite,diabase,granite,etc.sizes of 9 .5~1 3 .2 mm, 1 3 .2~1 6 .0 mm)were studied based on multi-fractal theory.Subj ective evaluation on the surfaces was obtained with a microscope (60X),and then aggregates′surface profile were measured by laser profil-ometer.To verify the multi-fractal characteristic of surface texture,high-pass butterworth was used to filter the surface texture roughness from the profiles.The relationship was finally analyzed between the spectrum widthΔαand polishing stone value pPSV .The results show that the multi-fractal spec-trum parameters are more detailed characterization for surface roughness,Δαquantifies the uneven distribution of aggregate surface globally andΔf characterizes the ratio of peak and valley of aggregate surface locally;the spectral widthΔαhave a good linear correlation with pPSV (the fitting goodness is 0.84),the higher pPSV is,the widerΔαis;it is feasible that multi-fractal spectrum evaluates rough-ness of aggregates surface texture,so it will provide fast and comprehensive evaluation method for practical engineering application.

Collecting 3D micro-texture of asphalt pavement is the key work of studying the constitutive relationship between asphalt pavement micro-texture and pavement performance. Due to affine characteristics of asphalt pavement micro-texture, it is very difficult to obtain 3D micro-texture quickly and accurately. The paper provided a 3D reconstruction method for asphalt pavement by establishing 3D mathematical measuring model of binocular vision. The camera interior and exterior parameters was obtained by camera calibration, Homograph H was calculated by feature points extraction and matching with the help of SIFT algorithm, then realized area-points matching, finally the asphalt pavement micro-texture 3D image reconstruction was achieved based on the parameters and area-point coordinates. The reconstruction results show that using binocular vision method to extract the 3D image micro-texture is feasible and rapid, it is potential to acquire a large area 3D reconstruction of asphalt pavement micro-texture with the help of homograph.

BACKGROUND:Many centrality measures have been proposed to mine and characterize the correlations between network topological properties and protein essentiality. However, most of them show limited prediction accuracy, and the number of common predicted essential proteins by different methods is very small.RESULTS:In this paper, an ensemble framework is proposed which integrates gene expression data and protein-protein interaction networks (PINs). It aims to improve the prediction accuracy of basic centrality measures. The idea behind this ensemble framework is that different protein-protein interactions (PPIs) may show different contributions to protein essentiality. Five standard centrality measures (degree centrality, betweenness centrality, closeness centrality, eigenvector centrality, and subgraph centrality) are integrated into the ensemble framework respectively. We evaluated the performance of the proposed ensemble framework using yeast PINs and gene expression data. The results show that it can considerably improve the prediction accuracy of the five centrality measures individually. It can also remarkably increase the number of common predicted essential proteins among those predicted by each centrality measure individually and enable each centrality measure to find more low-degree essential proteins.CONCLUSIONS:This paper demonstrates that it is valuable to differentiate the contributions of different PPIs for identifying essential proteins based on network topological characteristics. The proposed ensemble framework is a successful paradigm to this end.

运用改进的立方体覆盖法计算路表微观构造三维分形维数.利用该方法计算了4种路面常用石料的沥青混合料试件的三维分形雏数,并与摆式摩擦仪测量结果进行对比分析,结果表明:分形维数D越大,微观构造越丰富;分形维敦D和摩擦系数值F是相互对应的,分形雏数D趋大,对应的摩擦系数F也越大,其抗滑性能超好.

为减少长寿命复合式路面的沥青层厚度与反射性裂缝,将冷拌水泥-乳化沥青混凝土应用于高压应力区(联结层),并设置大粒径沥青碎石与水泥稳定碎石复合式基层.用三维有限元方法分析其路面设计控制参数以及路面材料宏观力学参数(弹性模量)与各层结构厚度对长寿面复合式路面的影响,结果表明各层材料最大拉应力、最大剪应力为强度控制指标更为合适,同时结合工程实际提出了水泥乳化沥青混凝土长寿命复合式路面结构,并应用于宜昌城区的试验路段.

石料的磨光性能与沥青路面的抗滑性能密切相关.通过光学显微镜对典型石料进行电镜试验,鉴定出石料的构造以及主要成份,分析得出造成石料磨光性能差异的原因,可为沥青路面石料的选材提供合理的依据.

In order to overcome the disadvantages of mean texture depth(MTD)measurement methods,with section method and digital image technology method combined and with the help of laser vision advanced technology,a new MTD measurement method based on laser vision was proposed.Based on triangulation measurement principle,laser vision 3Dmathematical model was established.According to the measurement requirements,the general image processing method was put forward.By using the mean profile depth(MPD)estimation,MTD measurement was realized.Comparative experiments with sand patch method were conducted.The results show that the new method is of simple operation with moderate price equipment,its measuring results are of higher resolution and accuracy and its measuring data are 3Dvisual,which has high value of engineering practicability.

沥青路面抗滑性能与行车安全密切相关,研究抗滑性能衰变的规律与变异,对指导沥青面层抗滑性能恢复,保证行车的安全与舒适有重要意义.鉴于此,通过室外跟踪检测对国内常用的AC-13C,SMA-13,OGFC-13沥青路面抗滑级配进行对比研究,揭示了沥青路面抗滑性能的衰变规律.

BACKGROUND:Experimental methods for the identification of essential proteins are always costly, time-consuming, and laborious. It is a challenging task to find protein essentiality only through experiments. With the development of high throughput technologies, a vast amount of protein-protein interactions are available, which enable the identification of essential proteins from the network level. Many computational methods for such task have been proposed based on the topological properties of protein-protein interaction (PPI) networks. However, the currently available PPI networks for each species are not complete, i.e. false negatives, and very noisy, i.e. high false positives, network topology-based centrality measures are often very sensitive to such noise. Therefore, exploring robust methods for identifying essential proteins would be of great value.METHOD:In this paper, a new essential protein discovery method, named CoEWC (Co-Expression Weighted by Clustering coefficient), has been proposed. CoEWC is based on the integration of the topological properties of PPI network and the co-expression of interacting proteins. The aim of CoEWC is to capture the common features of essential proteins in both date hubs and party hubs. The performance of CoEWC is validated based on the PPI network of Saccharomyces cerevisiae. Experimental results show that CoEWC significantly outperforms the classical centrality measures, and that it also outperforms PeC, a newly proposed essential protein discovery method which outperforms 15 other centrality measures on the PPI network of Saccharomyces cerevisiae. Especially, when predicting no more than 500 proteins, even more than 50% improvements are obtained by CoEWC over degree centrality (DC), a better centrality measure for identifying protein essentiality.CONCLUSIONS:We demonstrate that more robust essential protein discovery method can be developed by integrating the topological properties of PPI network and the co-expression of interacting proteins. The proposed centrality measure, CoEWC, is effective for the discovery of essential proteins.