ERACC(Extension Rule Based on Accurate Configuration Checking)算法由杨洋等人基于扩展规则和格局检测提出,具有较高的推理效率.为进一步提高ERACC算法在大规模SAT(Satisfiability)问题求解上的性能,本文在搜索由极大项组成的空间时...ERACC(Extension Rule Based on Accurate Configuration Checking)算法由杨洋等人基于扩展规则和格局检测提出,具有较高的推理效率.为进一步提高ERACC算法在大规模SAT(Satisfiability)问题求解上的性能,本文在搜索由极大项组成的空间时,首先利用IMOM(Improved Maximum Occurrences on Clauses of Maximum Size)思想生成初始极大项,接着设计了适用于扩展规则推理的CCA_ER(Configuration Checking with Aspiration for Extension Rule-Based Reasoning)启发式策略,为极大项中格局信息未发生变化的变量对应文字提供一定的翻转机会.同时,为进一步提高扩展规则推理算法在k-SAT问题求解上的性能,设计了适用于扩展规则推理的PAWS_ER(Pure Additive Weighting Scheme for Extension Rule-Based Reasoning)策略,并且给出变量的Subscore_ER(Subscore for Extension Rule-Based Reasoning),CScore_ER(Comprehensive Score for Extension Rule-Based Reasoning)和HScore_ER(Hybrid Score for Extension Rule-Based Reasoning)属性.在此基础上,提出了ERACC_IAPS(ERACC with IMOM,CCA_ER,PAWS_ER and Subscore_ER)和CERACC_IAPS(ERACC with IMOM,CCA_ER,PAWS_ER,CScore_ER and HScore_ER)算法.实验结果表明:ERACC_IAPS和CERACC_IAPS算法的效率明显优于ERACC算法,最高可将其求解效率提高1000多倍.展开更多
采用磷酸盐法对灵芝(Ganoderma lucidum)β-葡聚糖进行磷酸化修饰,在不同反应温度下(60、70、80、90、100℃)得到灵芝β-葡聚糖磷酸化衍生物(PGLP1~5),随着反应温度的上升,衍生物取代度(degree of substitution,DS)逐渐增大,重均分子量(...采用磷酸盐法对灵芝(Ganoderma lucidum)β-葡聚糖进行磷酸化修饰,在不同反应温度下(60、70、80、90、100℃)得到灵芝β-葡聚糖磷酸化衍生物(PGLP1~5),随着反应温度的上升,衍生物取代度(degree of substitution,DS)逐渐增大,重均分子量(average molecular weight,Mw)逐渐下降,反应温度为100℃时取代度最大为1.02。灵芝β-葡聚糖磷酸化衍生物体外对肿瘤细胞K562和L1210的增殖均有抑制作用,随着取代度增加,抑制率增高,且呈浓度依赖性;取代度最大的磷酸化衍生物PGLP5(DS=1.02,Mw=0.8×10^4)在200μg·mL^-1浓度下,对K562和L1210细胞增殖的抑制率分别为58.74%和50.05%,其IC50值分别为99.61μg·mL^-1和187.52μg·mL^-1。研究结果表明,灵芝β-葡聚糖磷酸化衍生物具有体外抗肿瘤活性,且抗肿瘤活性强弱与其取代度大小有关,取代度越大,抗肿瘤活性越强。展开更多
文摘ERACC(Extension Rule Based on Accurate Configuration Checking)算法由杨洋等人基于扩展规则和格局检测提出,具有较高的推理效率.为进一步提高ERACC算法在大规模SAT(Satisfiability)问题求解上的性能,本文在搜索由极大项组成的空间时,首先利用IMOM(Improved Maximum Occurrences on Clauses of Maximum Size)思想生成初始极大项,接着设计了适用于扩展规则推理的CCA_ER(Configuration Checking with Aspiration for Extension Rule-Based Reasoning)启发式策略,为极大项中格局信息未发生变化的变量对应文字提供一定的翻转机会.同时,为进一步提高扩展规则推理算法在k-SAT问题求解上的性能,设计了适用于扩展规则推理的PAWS_ER(Pure Additive Weighting Scheme for Extension Rule-Based Reasoning)策略,并且给出变量的Subscore_ER(Subscore for Extension Rule-Based Reasoning),CScore_ER(Comprehensive Score for Extension Rule-Based Reasoning)和HScore_ER(Hybrid Score for Extension Rule-Based Reasoning)属性.在此基础上,提出了ERACC_IAPS(ERACC with IMOM,CCA_ER,PAWS_ER and Subscore_ER)和CERACC_IAPS(ERACC with IMOM,CCA_ER,PAWS_ER,CScore_ER and HScore_ER)算法.实验结果表明:ERACC_IAPS和CERACC_IAPS算法的效率明显优于ERACC算法,最高可将其求解效率提高1000多倍.
