Skip to main content

Table 3 Correlation coefficients between predicted σ values by Sigma-RF and the actual errors for CACA distances of 22 CASP9 targets are shown

From: Sigma-RF: prediction of the variability of spatial restraints in template-based modeling by random forest

Target ID

With 20 features

With top 10 features

T0517

0.5622

0.5774

T0523

0.3923

0.3041

T0527

0.3402

0.3355

T0536

0.3138

0.3438

T0538

0.2225

0.2998

T0539

0.5197

0.5093

T0545

0.3312

0.2289

T0552

0.4061

0.4277

T0557

0.4447

0.3720

T0559

0.2589

0.2237

T0560

0.4392

0.4080

T0566

0.4536

0.3619

T0567

0.1997

0.1960

T0580

0.2354

0.2948

T0586

0.3713

0.4038

T0590

0.1218

0.0670

T0594

0.3364

0.3330

T0598

0.3145

0.2489

T0602

0.5608

0.4723

T0610

0.2756

0.2825

T0615

0.4062

0.4177

T0622

0.5028

0.4853

Average

0.3640

0.3452

  1. Results using the full 20 features as well as using top 10 features are shown. On average, by using only half of the features, 95% of the prediction level is achieved.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us