PAIR

The Predicted Arabidopsis Interactome Resource (PAIR) contains 149,900 protein-protein interactions involving 10,380 proteins. This dataset includes 145,494 interactions predicted by the PAIR V3 prediction model and 5590 experimentally reported interactions collected from the major interaction databases. The predicted interactions are expected to cover 24.47% of the entire Arabidopsis interactome with a high reliability of 43.52%. In addition, another set of predicted interactions covering 74.55% of the entire Aradopsis interactome but with low reliability is also available for download. The prediction accuracies have been validated by two external benchmark datasets.预测型拟南芥互作组资源（Predicted Arabidopsis Interactome Resource, PAIR）包含149900条蛋白质-蛋白质互作，涵盖10380个蛋白质。该数据集包含145494条由PAIR V3预测模型生成的预测互作，以及从主流互作数据库中采集的5590条实验验证的互作记录。上述预测互作预计覆盖全拟南芥互作组的24.47%，预测可靠性高达43.52%。此外，另有一组覆盖全拟南芥互作组74.55%的预测互作可供下载，但该组互作的预测可靠性偏低。该预测模型的准确率已通过两个外部基准数据集完成验证。

搜集汇总

背景与挑战

背景概述

PAIR数据集包含149,900个蛋白质相互作用，涉及10,380个蛋白质，涵盖由PAIR V3模型预测的145,494个相互作用和从主要数据库收集的5,590个实验报道相互作用。预测相互作用覆盖了拟南芥相互作用组的24.47%，可靠性达43.52%，同时提供另一组覆盖率达74.55%但可靠性较低的预测相互作用，预测准确性已通过外部基准数据集验证。