In the year 1971,the world’s biggest structural biology collaboration name—The Research Collaboratory for Structural Bioinformatics(RCSB),was formed to gather all the structural biologists at a single platform and t...In the year 1971,the world’s biggest structural biology collaboration name—The Research Collaboratory for Structural Bioinformatics(RCSB),was formed to gather all the structural biologists at a single platform and then extended out to be the world’s most extensive structural data repository named RCSB-Protein Data Bank(PDB)(https://www.rcsb.org/)that has provided the service for more than 50 years and continues its legacy for the discoveries and repositories for structural data.The RCSB has evolved from being a collaboratory network to a full-fledged database and tool with a huge list of protein structures,nucleic acid-containing structures,ModelArchive,and AlphaFold structures,and the best is that it is expanding day by day with computational advancement with tools and visual experiences.In this review article,we have discussed how RCSB has been a successful collaboratory network,its expansion in each decade,and how it has helped the ground-breaking research.The PDB tools that are helping the researchers,yearly data deposition,validation,processing,and suggestions that can help the developer improve for upcoming years are also discussed.This review will help future researchers understand the complete history of RCSB and its developments in each decade and how various future collaborative networks can be developed in various scientific areas and can be successful by keeping RCSB as a case study.展开更多
文摘In the year 1971,the world’s biggest structural biology collaboration name—The Research Collaboratory for Structural Bioinformatics(RCSB),was formed to gather all the structural biologists at a single platform and then extended out to be the world’s most extensive structural data repository named RCSB-Protein Data Bank(PDB)(https://www.rcsb.org/)that has provided the service for more than 50 years and continues its legacy for the discoveries and repositories for structural data.The RCSB has evolved from being a collaboratory network to a full-fledged database and tool with a huge list of protein structures,nucleic acid-containing structures,ModelArchive,and AlphaFold structures,and the best is that it is expanding day by day with computational advancement with tools and visual experiences.In this review article,we have discussed how RCSB has been a successful collaboratory network,its expansion in each decade,and how it has helped the ground-breaking research.The PDB tools that are helping the researchers,yearly data deposition,validation,processing,and suggestions that can help the developer improve for upcoming years are also discussed.This review will help future researchers understand the complete history of RCSB and its developments in each decade and how various future collaborative networks can be developed in various scientific areas and can be successful by keeping RCSB as a case study.
文摘启动子和增强子之间的相互作用(Promoter-Enhancer interaction,PEI)与基因的转录与调控密切相关。本文以人B淋巴细胞系(GM12878)为研究对象,基于染色质环(Loop)数据库构建了启动子-增强子(Promoter-Enhancer,P-E)相互作用数据集,分析了P-E结构中149种转录因子(Transcription factor,TF)以及11种组蛋白修饰(Histone madification,HM)的相关性,筛选出与P-E结构具有较强关联的表观遗传修饰特征,并利用卷积神经网络(Convolutional neural network,CNN)和随机森林(Random forest,RF)算法预测了P-E相互作用对。结果显示RF预测的AUC值(Area under the curve)介于0.84至0.88之间,而CNN的AUC值在0.69至0.77之间,表明RF的预测性能略优于CNN。此外,仅使用TF信号作为特征的AUC值优于仅使用HM信号的情况,表明TF信号对P-E结构的识别具有更佳的效果。最后将TF和HM特征组合后,预测效果能够进一步提升,我们发现EGR1、H3K4me2、EP300等12种特征是预测PEI的重要特征。
