Fe reducing bacteria(FRB),through extracellular electron transfer(EET)pathway,can reduce Fe(III)nanoparticles,thereby affecting the migration,transformation,and degradation of pollutants.However,the interaction of Fe(...Fe reducing bacteria(FRB),through extracellular electron transfer(EET)pathway,can reduce Fe(III)nanoparticles,thereby affecting the migration,transformation,and degradation of pollutants.However,the interaction of Fe(III)nanoparticles with the most commonly identified FRB,Geobacter sulfurreducens PCA,remains poorly understood.Herein,we demonstrated that the synergistic role of outer membrane proteins and periplasmic proteins in the EET process for-Fe_(2)O_(3),Fe3O4,and𝛽α-FeOOH nanoparticles by construction of multiple gene knockout strain.oxpG(involved in the type II secretion system)and omcST(outer membrane c-type cytochrome)medi-ated pathways accounted for approximately 67%of the total reduction of𝛼α-Fe_(2)O_(3) nanoparticles.The residual reduction of𝛼α-Fe_(2)O_(3) nanoparticles in∆oxpG-omcST strain was likely caused by redox-active substances in cell supernatant.Conversely,the reduction of dissolved Fe(III)was almost unaffected in∆oxpG-omcST strain at the same concentration.However,at high dissolved Fe(III)concentration,the reduction significantly decreased due to the formation of Fe(III)nanoparticles,suggesting that this EET process is specific to Fe(III)nanoparticles.Overall,our study provided a more comprehensive understanding for the EET pathways between G.sulfurreducens PCA and different Fe(III)species,enriching our knowledge on the role of microorganisms in iron biogeochemical cycles and remediation strategies of pollutants.展开更多
针对传统方法在运动鞋用户评论的感性因子提取中存在的效率低下、维度冗余问题,提出一种结合大语言模型(large language model,LLM)与主成分分析(principal component analysis,PCA)的自动化提取方法。以亚马逊电商平台的8680条运动鞋...针对传统方法在运动鞋用户评论的感性因子提取中存在的效率低下、维度冗余问题,提出一种结合大语言模型(large language model,LLM)与主成分分析(principal component analysis,PCA)的自动化提取方法。以亚马逊电商平台的8680条运动鞋用户评论为研究对象,采用GLM-4-9B-Chat模型自动生成感性词汇对,经数据清理后获得7619条有效数据;通过TF-IDF向量化处理后,设计k=10、15、20、25四组K-means聚类实验,对冗余维度进行合并优化,最终收敛得到6个核心感性因子。该方法通过整合LLM自动化提取、多聚类去冗余与PCA分析,为运动鞋感性工学的自动分析提供了一条技术路径,也为纺织服装领域的感性因子自动化提取研究提供了有益参考。展开更多
基金supported by the National Key Research and Development Project(No.2020YFA0907500)the National Natural Science Foundation of China(No.22476206)+1 种基金the supports from the National Young Top-Notch Talents(No.W03070030)Youth Innovation Promotion Association of the Chinese Academy of Sciences(No.Y202011).
文摘Fe reducing bacteria(FRB),through extracellular electron transfer(EET)pathway,can reduce Fe(III)nanoparticles,thereby affecting the migration,transformation,and degradation of pollutants.However,the interaction of Fe(III)nanoparticles with the most commonly identified FRB,Geobacter sulfurreducens PCA,remains poorly understood.Herein,we demonstrated that the synergistic role of outer membrane proteins and periplasmic proteins in the EET process for-Fe_(2)O_(3),Fe3O4,and𝛽α-FeOOH nanoparticles by construction of multiple gene knockout strain.oxpG(involved in the type II secretion system)and omcST(outer membrane c-type cytochrome)medi-ated pathways accounted for approximately 67%of the total reduction of𝛼α-Fe_(2)O_(3) nanoparticles.The residual reduction of𝛼α-Fe_(2)O_(3) nanoparticles in∆oxpG-omcST strain was likely caused by redox-active substances in cell supernatant.Conversely,the reduction of dissolved Fe(III)was almost unaffected in∆oxpG-omcST strain at the same concentration.However,at high dissolved Fe(III)concentration,the reduction significantly decreased due to the formation of Fe(III)nanoparticles,suggesting that this EET process is specific to Fe(III)nanoparticles.Overall,our study provided a more comprehensive understanding for the EET pathways between G.sulfurreducens PCA and different Fe(III)species,enriching our knowledge on the role of microorganisms in iron biogeochemical cycles and remediation strategies of pollutants.
文摘针对传统方法在运动鞋用户评论的感性因子提取中存在的效率低下、维度冗余问题,提出一种结合大语言模型(large language model,LLM)与主成分分析(principal component analysis,PCA)的自动化提取方法。以亚马逊电商平台的8680条运动鞋用户评论为研究对象,采用GLM-4-9B-Chat模型自动生成感性词汇对,经数据清理后获得7619条有效数据;通过TF-IDF向量化处理后,设计k=10、15、20、25四组K-means聚类实验,对冗余维度进行合并优化,最终收敛得到6个核心感性因子。该方法通过整合LLM自动化提取、多聚类去冗余与PCA分析,为运动鞋感性工学的自动分析提供了一条技术路径,也为纺织服装领域的感性因子自动化提取研究提供了有益参考。
