Skip to main content

Table 2 PPB prediction results of small molecules for sparse modeling by ELS and FBS compared to the baseline results. These situations are shown in a part of Fig. 1(b)

From: Computational prediction of plasma protein binding of cyclic peptides from small molecule experimental data using sparse modeling techniques

Method

Training set

Test set

RMSE (fb)

MAE (fb)

R (ln Ka)

ridge-SM

(baseline #1)

Small molecules (training data)

Small molecules (test data)

0.212

0.155

0.781

lasso-SM

(baseline #2)

Small molecules (training data)

Small molecules (test data)

0.233

0.176

0.707

ELS

Small molecules (training data)

Small molecules (test data)

0.228

0.172

0.714

FBS

Small molecules (training data)

Small molecules (test data)

0.230

0.167

0.725

Non-linear model [36]

Small molecules (training data)

Small molecules (test data)

0.225–0.251

0.155–0.177

0.707–0.787