Table 6

Performance benchmark with the Remark 465 dataset under parameters at default values.

Method

threshold

sensitivity

specificity

MCC

PE


DISOPRED2

0.05

0.566

0.874

0.373

0.441

DisEMBL Rem465

1.2

0.273

0.969

0.330

0.242

IUPred short

0.5

0.406

0.920

0.327

0.326

IUPred long

0.5

0.394

0.899

0.276

0.293

SEG25

3.00;3.30

0.269

0.908

0.182

0.177

SEG12

2.20;2.50

0.136

0.967

0.160

0.103

DisEMBL Hotloops

1.4

0.432

0.780

0.159

0.211

CAST

40

0.179

0.940

0.146

0.119

SEG45

3.40;3.75

0.293

0.865

0.141

0.157

DisEMBL Coils

1.2

0.733

0.457

0.124

0.190


Predictors were run and ranked as described in Table 5 but over the Remark 465 dataset (short disorder). As in Table 5, rankings differ if based on MCC or PE.

Sirota et al. BMC Genomics 2010 11(Suppl 1):S15   doi:10.1186/1471-2164-11-S1-S15

Open Data