Table 2 Pearson correlation coefficients between experimental and calculated ΔΔG values for interface and non-interface mutations.

From: PremPLI: a machine learning model for predicting the effects of missense mutations on protein-ligand interactions

Dataset

Method

Interface

Non-interface

PCC

RMSE

PCC

RMSE

S144

PremPLI

0.45

0.92

0.28

0.56

mCSM-lig17

0.41

1.11

—

0.61

ML119

—

1.05

—

0.61

R1519

0.74**

0.82

—

0.59

Prime25

0.25

2.40

—

0.76

FEP+25

0.64*

1.15

−0.26**

0.97

S99

PremPLIC

0.70

1.22

0.67

0.75

A1427

0.57

1.31

—

1.44

R1427

0.36**

1.48

—

1.02

RMD27

0.59

1.25

−0.36**

1.19

  1. Only correlation coefficients statistically significantly different from zero are shown (p-value < 0.05, t-test). * and ** indicate statistically significant difference between PremPLI and other methods in terms of PCC (Hitter et al.51 test) with p-value < 0.05 and p-value < 0.01, respectively.
Back to article page