Skip to main content

Table 5 Comparison of KnotAli with RNAalifold, Hxmatch, and Cacofold

From: KnotAli: informed energy minimization through the use of evolutionary information

Family

KnotAli

RNAalifold

Hxmatch

Cacofold

Sen

ppv

F

Sen

ppv

F

Sen

ppv

F

Sen

ppv

F

(a) Input alignment created through MUSCLE

5s

.899

.876

.887

.761

.884

.817

.424

.917

.579

.835

.859

.846

16s

.494

.501

.494

.506

.748

.602

.385

.724

.502

.172

.455

.250

23s

.589

.545

.566

.798

.767

.782

.552

.625

.587

.242

.341

.283

Group I intron

.490

.444

.461

.047

.588

.087

.034

.529

.064

.039

.300

.068

Group II intron

.177

.139

.154

0

0

0

.010

.108

.018

0

0

0

RNaseP

.498

.491

.493

.235

.602

.334

.135

.699

.225

.164

.552

.251

SRP

.580

.556

.564

.165

.764

.241

.166

.897

.25

.186

.496

.255

Telomerase

.289

.233

.256

.508

.636

.563

.292

.711

.413

.380

.480

.423

tmRNA

.491

.468

.477

.255

.803

.386

.176

.852

.291

.234

.868

.367

tRNA

.950

.917

.932

.931

.972

.949

.764

.974

.854

.886

.970

.925

(b) Input alignment created through MAFFT

5s

.902

.871

.885

.644

.843

.729

.385

.927

.541

.739

.852

.790

16s

.548

.495

.519

.548

.794

.647

.440

.802

.567

.198

.467

.277

23s

.329

.468

.361

.795

.771

.783

.563

.627

.593

.281

.336

.281

Group I intron

.424

.378

.396

.036

.719

.068

.036

.719

.069

.051

.384

.090

Group II intron

.382

.244

.295

.106

.773

.186

.103

.580

.175

.105

.920

.187

RNaseP

.592

.583

.585

.207

.759

.322

.067

.700

.122

.301

.639

.400

SRP

.420

.416

.415

.078

.661

.129

.078

.885

.134

.148

.377

.206

Telomerase

.243

.186

.211

.25

.361

.294

.255

.483

.333

.442

.517

.475

tmRNA

.504

.492

.495

.196

.799

.313

.047

.435

.084

.255

.650

.362

tRNA

.898

.878

.886

.8

.931

.855

.642

.940

.758

.853

.925

.882*

  1. Each column corresponds to algorithm used and each sub-column represents a metric: F-measure, Sensitivity or PPV. BOLD represents the significantly highest accuracy compared to others. In the case of two algorithms whose accuracy outperformed the rest while not significantly better than each other, both were represented in bold. An accompanying * is then used to denote a p-value close to but not below .05.