论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins

论文:Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
**TS-PIN:**本文提出了一种利用基因表达谱和亚细胞定位信息构建精蛋白的新方法。提炼PIN的基本思想是,如果两种蛋白质在同一亚细胞位置同时出现,并且至少在细胞周期的某个时间点同时活跃,那么它们在物理上相互作用的可能性应该更高。
因此,在本研究中,提出了一种新的过滤假阳性的方法,假设两种蛋白质在同一亚细胞位置上,并且至少在细胞周期的某个时间点一起活动时,它们相互作用的可能性更高。相应地,我们构建了TS-PIN,并试图提高基于新网络识别关键蛋白的准确性。

基因表达数据反映的是直接或间接测量得到的基因转录产物 mRNA在细胞中的丰度,这些数据可以用于分析哪些基因的表达发生了改变,基因之间有何相关性,在不同条件下基因的活动是如何受影响的。

亚细胞定位是指某种蛋白或表达产物在细胞内的具体存在部位。例如在核内、胞质内或者细胞膜上存在。GFP是绿色荧光蛋白,在扫描共聚焦显微镜的激光照射下会发出绿色荧光,从而可以精确地定位蛋白质的位置。
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
论文阅读Construction of Refined Protein Interaction Network for Predicting Essential Proteins
sensitivity (Sn):TP/(TP+FN),代表分类器预测的正类中实际正实例占所有正实例的比例。
specificity (Sp):TN/(FP+TN),代表分类器预测的负类中实际负实例占所有负实例的比例,
F-measure (F) F1值的一般形式为sp和sn的调和均值。 是分类与信息检索中最常用的指标之一。只有当精确率sp和召回率sn都很高时,F1值才会高
positive predictive value (PPV):指的是模型判为正的所有样本中有多少是真正的正样本
negative predictive value (NPV) 指的是模型判为负的所有样本中有多少是真正的负样本
accuracy (ACC) 分类正确的样本数/所有样本数量

TS-PIN的10种方法的灵敏度(Sn)、特异性(Sp)、F-测量(F)、阳性预测值(PPV)、阴性预测值(NPV)和准确度(ACC)均高于S-PIN和NF-APIN。