2024年2月26日 · 预测对齐误差( Predicted Aligned Error，PAE )是AlphaFold系统的另一个输出结果。 AlphaFold DB提供给结构PAE的图片和数据.json文件。 它表示如果预测结构和实际结构在残基y (使用Cα、N和C原子)上对齐，显示在残基x处的期望位置误差。

The Predicted Aligned Error (PAE) is a quantitative output produced by AlphaFold, a protein structure prediction system developed by DeepMind. [1] PAE estimates the expected positional error for each residue in a predicted protein structure if it were aligned to a corresponding residue in the true protein structure.

pae（Predicted Aligned Error，预测对齐误差）:PAE表示预测的残基对之间的平均距离误差，用于评估预测模型的精度。较低的PAE 值通常意味着 预测的结构与实际结构更为接近。

2024年2月15日 · pTM (predicted template modeling) 是另一种置信度指标，取值0-1，也是越高越可信。tol是与上一轮循环model的RMSD (root-mean-square deviation)，表示两轮循环结果的差别，值越小差别越小。 得到的5个model默认按pLDDT排序，保存为unrelaxed的 pdb 文件。

Predicted aligned error (PAE) is a measure of how confident AlphaFold2 is in the relative position of two residues within the predicted structure. PAE is defined as the expected positional error at residue X, measured in Ångströms (Å), if the predicted and …

第二个度量是 PAE （预测对齐误差），当预测和真实结构在残基y上对齐时，它报告AlphaFold在残基x处的预期位置误差。这对于评估对全局特征（尤其是域包装）的信心很有用。

2022年6月23日 · We show that PAE maps from AF2 are correlated with the distance variation (DV) matrices from molecular dynamics (MD) simulations, which reveals that the PAE maps can predict the dynamical...

The “predicted aligned error” or PAE values can be shown as a 2D plot using the command alphafold pae (details below), the AlphaFold Error Plot tool, alphafold fetch or alphafold match with the option pae true, or the open command. PAE and other pairwise metrics associated with ModelArchive entries can also be plotted (see modelcif pae).

2024年6月18日 · Our PAE model is specifically designed to maintain crucial protein symmetries, particularly rotation and translation, through carefully chosen architectural features. To address rotational symmetry, the model incorporates random rotations during training, enhancing its ability to recognize and adapt to various orientations of protein structures.

The PAE is a two-stage generative model, composed of an Auto-Encoder (AE) that is interpreted probabilistically after training with a Normalizing Flow (NF). The AE compresses the data and maps it to a lower dimensional latent space.