相较于传统药物的研发,药物-靶标的预测方法能够有效降低成本,加快研发进程,但是在实际应用中存在数据集平衡度低、预测精确率不高等问题。基于此,提出一种自适应球形演化的药物-靶标相互作用预测方法ASEKELM(self-Adaptive Spherical E...相较于传统药物的研发,药物-靶标的预测方法能够有效降低成本,加快研发进程,但是在实际应用中存在数据集平衡度低、预测精确率不高等问题。基于此,提出一种自适应球形演化的药物-靶标相互作用预测方法ASEKELM(self-Adaptive Spherical Evolution based on Kernel Extreme Learning Machine)。该方法根据结构相似的药物与靶标更易存在相互作用的原理筛选出高置信度的负样本;并且为了解决球形演化算法易陷入局部最优的问题,利用搜索因子历史记忆的反馈机制及群大小线性递减的策略(LPSR),实现全局搜索和局部搜索的平衡,提高算法的寻优能力;然后利用自适应球形演化算法对核极限学习机(KELM)的参数进行优化。在基于黄金标准的数据集上将ASEKELM与NetLapRLS(Network Laplacian Regularized Least Square)、BLM-NII(Bipartite Local Model with Neighbor-based Interaction profile Inferring)等算法进行对比,验证算法的性能。实验结果表明,在酶(E)、G-蛋白偶联受体(GPCR)、离子通道(IC)和核受体(NR)数据集中,ASE-KELM的ROC曲线下面积(AUC)与PR曲线下面积(AUPR)均优于对比算法;且基于DrugBank等数据库,ASE-KELM在预测新药物-靶标对的验证过程中表现良好。展开更多
The underwater wireless optical communication(UWOC)system has gradually become essential to underwater wireless communication technology.Unlike other existing works on UWOC systems,this paper evaluates the proposed ma...The underwater wireless optical communication(UWOC)system has gradually become essential to underwater wireless communication technology.Unlike other existing works on UWOC systems,this paper evaluates the proposed machine learningbased signal demodulation methods through the selfbuilt experimental platform.Based on such a platform,we first construct a real signal dataset with ten modulation methods.Then,we propose a deep belief network(DBN)-based demodulator for feature extraction and multi-class feature classification.We also design an adaptive boosting(Ada Boost)demodulator as an alternative scheme without feature filtering for multiple modulated signals.Finally,it is demonstrated by extensive experimental results that the Ada Boost demodulator significantly outperforms the other algorithms.It also reveals that the demodulator accuracy decreases as the modulation order increases for a fixed received optical power.A higher-order modulation may achieve a higher effective transmission rate when the signal-to-noise ratio(SNR)is higher.展开更多
P-arsanilic acid(p-ASA),as a kind of organoarsenic feed additive,has been widely used in poultry and swine breeding.However,it has caused the arsenic pollution around the farm.Currently data shows humic acid(HA)is clo...P-arsanilic acid(p-ASA),as a kind of organoarsenic feed additive,has been widely used in poultry and swine breeding.However,it has caused the arsenic pollution around the farm.Currently data shows humic acid(HA)is closely to the migration and transformation of p-ASA.Therefore,the interaction between p-ASA and HA was investigated by using the method of fluorescence quenching titration.The association constant changed from2.74 to 4.88 L·mol-1at a p H varying from 5 to 9 and reached the maximum at p H 7.In addition,log K varied from4.15 to 5.02 L·mol-1when the temperature increased from 15℃to 35℃.The log K increased with an increase in the concentration of HA.The dominant mechanism between p-ASA and HA is static quenching.The primary interaction force was likely the hydrogen bond,and the binding behavior occurred on the As-O stretch of p-ASA and the carboxylic acid C=O stretch of HA.The results showed that dissolved organic matters could affect the fate and biogeochemical cycling of organoarsenic pharmaceuticals in aquatic ecosystems.展开更多
文摘相较于传统药物的研发,药物-靶标的预测方法能够有效降低成本,加快研发进程,但是在实际应用中存在数据集平衡度低、预测精确率不高等问题。基于此,提出一种自适应球形演化的药物-靶标相互作用预测方法ASEKELM(self-Adaptive Spherical Evolution based on Kernel Extreme Learning Machine)。该方法根据结构相似的药物与靶标更易存在相互作用的原理筛选出高置信度的负样本;并且为了解决球形演化算法易陷入局部最优的问题,利用搜索因子历史记忆的反馈机制及群大小线性递减的策略(LPSR),实现全局搜索和局部搜索的平衡,提高算法的寻优能力;然后利用自适应球形演化算法对核极限学习机(KELM)的参数进行优化。在基于黄金标准的数据集上将ASEKELM与NetLapRLS(Network Laplacian Regularized Least Square)、BLM-NII(Bipartite Local Model with Neighbor-based Interaction profile Inferring)等算法进行对比,验证算法的性能。实验结果表明,在酶(E)、G-蛋白偶联受体(GPCR)、离子通道(IC)和核受体(NR)数据集中,ASE-KELM的ROC曲线下面积(AUC)与PR曲线下面积(AUPR)均优于对比算法;且基于DrugBank等数据库,ASE-KELM在预测新药物-靶标对的验证过程中表现良好。
基金supported by the major key project of Peng Cheng Laboratory under grant PCL2023AS31 and PCL2023AS1-2the National Key Research and Development Program of China(No.2019YFA0706604)the Natural Science Foundation(NSF)of China(Nos.61976169,62293483,62371451)。
文摘The underwater wireless optical communication(UWOC)system has gradually become essential to underwater wireless communication technology.Unlike other existing works on UWOC systems,this paper evaluates the proposed machine learningbased signal demodulation methods through the selfbuilt experimental platform.Based on such a platform,we first construct a real signal dataset with ten modulation methods.Then,we propose a deep belief network(DBN)-based demodulator for feature extraction and multi-class feature classification.We also design an adaptive boosting(Ada Boost)demodulator as an alternative scheme without feature filtering for multiple modulated signals.Finally,it is demonstrated by extensive experimental results that the Ada Boost demodulator significantly outperforms the other algorithms.It also reveals that the demodulator accuracy decreases as the modulation order increases for a fixed received optical power.A higher-order modulation may achieve a higher effective transmission rate when the signal-to-noise ratio(SNR)is higher.
基金National Natural Science Foundation of China(41373111)
文摘P-arsanilic acid(p-ASA),as a kind of organoarsenic feed additive,has been widely used in poultry and swine breeding.However,it has caused the arsenic pollution around the farm.Currently data shows humic acid(HA)is closely to the migration and transformation of p-ASA.Therefore,the interaction between p-ASA and HA was investigated by using the method of fluorescence quenching titration.The association constant changed from2.74 to 4.88 L·mol-1at a p H varying from 5 to 9 and reached the maximum at p H 7.In addition,log K varied from4.15 to 5.02 L·mol-1when the temperature increased from 15℃to 35℃.The log K increased with an increase in the concentration of HA.The dominant mechanism between p-ASA and HA is static quenching.The primary interaction force was likely the hydrogen bond,and the binding behavior occurred on the As-O stretch of p-ASA and the carboxylic acid C=O stretch of HA.The results showed that dissolved organic matters could affect the fate and biogeochemical cycling of organoarsenic pharmaceuticals in aquatic ecosystems.