Skip to main content

Table 3 The results of the RP and RMSE with standard deviation of the proposed models (GES_PPI and gnn_PPI)

From: An interpretable deep geometric learning model to predict the effects of mutations on protein–protein interactions using large-scale protein language model

Method

Dataset

Rp

RMSE

GES_PPI

S2648

0.6491 \(\pm\) 0.0108

1.1324 \(\pm\) 0.0114

S3421

0.7166 \(\pm\) 0.1022

1.6413 \(\pm\) 0.0840

S4169

0.6892 \(\pm\) 0.0126

1.5633 \(\pm\) 0.0143

M1101

0.5679 \(\pm\) 0.0221

1.7622 \(\pm\) 0.1421

M1707

0.7538 \(\pm\) 0.1097

2.1421 \(\pm\) 0.0988

gnn_PPI*

S2648

0.6272 \(\pm\) 0.0441

1.2197 \(\pm\) 0.0308

S3421

0.6725 \(\pm\) 0.1320

1.6944 \(\pm\) 0.0806

S4169

0.6254 \(\pm\) 0.1052

1.5697 \(\pm\) 0.1231

M1101

0.5433 \(\pm\) 0.2597

1.7607 \(\pm\) 0.2080

M1707

0.7397 \(\pm\) 0.1956

2.1756 \(\pm\) 0.1427

  1. *: gnn_PPI represents the method without the ESM pre-trained model