[目的/意义]识别新兴研究主题并预测其发展趋势,对科技创新和科研决策具有重要意义。[方法/过程]提出基于主题模型和曲线拟合的新兴主题趋势预测方法。该方法应用LDA主题模型进行科技文献主题划分,然后将主题新颖性、增长性、影响力等...[目的/意义]识别新兴研究主题并预测其发展趋势,对科技创新和科研决策具有重要意义。[方法/过程]提出基于主题模型和曲线拟合的新兴主题趋势预测方法。该方法应用LDA主题模型进行科技文献主题划分,然后将主题新颖性、增长性、影响力等特征指标依次赋权叠加构建主题新兴指标,利用多维尺度绘制主题分布矩阵以识别和探测新兴主题。最后基于主题新兴指标时序特征进行曲线拟合,预测新兴主题未来发展趋势。[结果/结论]利用Web of Science数据库中1997—2017年燃料电池领域的94661篇文献,进行实证研究。结果表明该方法能够有效识别燃料电池领域新兴研究主题,相对于时间序列自回归预测方法,曲线拟合预测方法具有较高准确率。展开更多
Comprehensive characterization of metabolites and metabolic profiles in plasma has considerable significance in determining the efficacy and safety of traditional Chinese medicine(TCM)in vivo.However,this process is u...Comprehensive characterization of metabolites and metabolic profiles in plasma has considerable significance in determining the efficacy and safety of traditional Chinese medicine(TCM)in vivo.However,this process is usually hindered by the insufficient characteristic fragments of metabolites,ubiquitous matrix interference,and complicated screening and identification procedures for metabolites.In this study,an effective strategy was established to systematically characterize the metabolites,deduce the metabolic pathways,and describe the metabolic profiles of bufadienolides isolated from Venenum Bufonis in vivo.The strategy was divided into five steps.First,the blank and test plasma samples were injected into an ultra-high performance liquid chromatography/linear trap quadrupole-orbitrap-mass spectrometry(MS)system in the full scan mode continuously five times to screen for valid matrix compounds and metabolites.Second,an extension-mass defect filter model was established to obtain the targeted precursor ions of the list of bufadienolide metabolites,which reduced approximately 39%of the interfering ions.Third,an acquisition model was developed and used to trigger more tandem MS(MS/MS)fragments of precursor ions based on the targeted ion list.The acquisition mode enhanced the acquisition capability by approximately four times than that of the regular data-dependent acquisition mode.Fourth,the acquired data were imported into Compound Discoverer software for identification of metabolites with metabolic network prediction.The main in vivo metabolic pathways of bufadienolides were elucidated.A total of 147 metabolites were characterized,and the main biotransformation reactions of bufadienolides were hydroxylation,dihydroxylation,and isomerization.Finally,the main prototype bufadienolides in plasma at different time points were determined using LC-MS/MS,and the metabolic profiles were clearly identified.This strategy could be widely used to elucidate the metabolic profiles of TCM preparations or Chinese patent medicines展开更多
文摘[目的/意义]识别新兴研究主题并预测其发展趋势,对科技创新和科研决策具有重要意义。[方法/过程]提出基于主题模型和曲线拟合的新兴主题趋势预测方法。该方法应用LDA主题模型进行科技文献主题划分,然后将主题新颖性、增长性、影响力等特征指标依次赋权叠加构建主题新兴指标,利用多维尺度绘制主题分布矩阵以识别和探测新兴主题。最后基于主题新兴指标时序特征进行曲线拟合,预测新兴主题未来发展趋势。[结果/结论]利用Web of Science数据库中1997—2017年燃料电池领域的94661篇文献,进行实证研究。结果表明该方法能够有效识别燃料电池领域新兴研究主题,相对于时间序列自回归预测方法,曲线拟合预测方法具有较高准确率。
基金supported by the National Natural Science Foundation of China (Grant Nos.: 81530095 and 81673591)Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No.: XDA12020348)+1 种基金National Standardization of Traditional Chinese Medicine Project (Grant No.: ZYBZH-K-LN-01)Science and Technology Commission Foundation of Shanghai (Grant No.: 15DZ0502800)
文摘Comprehensive characterization of metabolites and metabolic profiles in plasma has considerable significance in determining the efficacy and safety of traditional Chinese medicine(TCM)in vivo.However,this process is usually hindered by the insufficient characteristic fragments of metabolites,ubiquitous matrix interference,and complicated screening and identification procedures for metabolites.In this study,an effective strategy was established to systematically characterize the metabolites,deduce the metabolic pathways,and describe the metabolic profiles of bufadienolides isolated from Venenum Bufonis in vivo.The strategy was divided into five steps.First,the blank and test plasma samples were injected into an ultra-high performance liquid chromatography/linear trap quadrupole-orbitrap-mass spectrometry(MS)system in the full scan mode continuously five times to screen for valid matrix compounds and metabolites.Second,an extension-mass defect filter model was established to obtain the targeted precursor ions of the list of bufadienolide metabolites,which reduced approximately 39%of the interfering ions.Third,an acquisition model was developed and used to trigger more tandem MS(MS/MS)fragments of precursor ions based on the targeted ion list.The acquisition mode enhanced the acquisition capability by approximately four times than that of the regular data-dependent acquisition mode.Fourth,the acquired data were imported into Compound Discoverer software for identification of metabolites with metabolic network prediction.The main in vivo metabolic pathways of bufadienolides were elucidated.A total of 147 metabolites were characterized,and the main biotransformation reactions of bufadienolides were hydroxylation,dihydroxylation,and isomerization.Finally,the main prototype bufadienolides in plasma at different time points were determined using LC-MS/MS,and the metabolic profiles were clearly identified.This strategy could be widely used to elucidate the metabolic profiles of TCM preparations or Chinese patent medicines
