Skip to main content

Table 7 PCA dimensionality reduction results via 10-fold cross-validation for D. melanogaster

From: Comparative analysis and prediction of nucleosome positioning using integrative feature representation and machine learning algorithms

Method

Feature

Parameters

PCA%

ACC

Sn

Sp

MCC

AUC

SVM

FCGR

K = 1 + 2 + 4

0.88

0.8108

0.7786

0.8435

0.6235

0.8785

 

FCGR + DAC

K = 4, lag = 2

0.95

0.8070

0.7855

0.8288

0.6152

0.8768

 

FCGR + TAC

K = 4, lag = 2

0.93

0.8115

0.7831

0.8404

0.6245

0.8766

 

FCGR + DACC

K = 4, lag = 2

0.95

0.7809

0.7931

0.7684

0.5621

0.8343

 

FCGR + TACC

K = 4, lag = 2

0.95

0.7678

0.6879

0.8491

0.5440

0.8363

 

FCGR + PCPseDNC

K = 4, λ = 8, w = 0.5

0.9

0.8085

0.7752

0.8425

0.6190

0.8773

 

FCGR + PCPseTNC

K = 4, λ = 8, w = 0.5

0.93

0.8097

0.7772

0.8428

0.6215

0.8761

 

All features

 

0.95

0.7593

0.70414

0.81544

0.52275

0.80283

ELM

FCGR + DAC

K = 2, lag = 2

0.95

0.7817

0.7690

0.7947

0.5642

0.8544

 

FCGR + TAC

K = 2, lag = 2

0.95

0.7859

0.7735

0.7986

0.5723

0.8552

 

FCGR + DACC

K = 2, lag = 2

0.9

0.7530

0.7524

0.7537

0.5064

0.8262

 

FCGR + TACC

K = 2, lag = 2

0.95

0.7365

0.7472

0.7256

0.4733

0.8018

 

FCGR + PCPseDNC

K = 2, λ = 8, w = 0.5

0.93

0.7837

0.7597

0.8081

0.5685

0.8587

 

FCGR + PCPseTNC

K = 2, λ = 8, w = 0.5

0.88

0.7678

0.7283

0.8081

0.5379

0.8448

 

All features

K = 2

0.95

0.7727

0.7714

0.7740

0.5455

0.8437

XGBoost

FCGR

K = 1 + 2 + 4

0.9

0.8037

0.7824

0.8253

0.6085

0.8772

 

FCGR + DAC

K = 4, lag = 2

0.9

0.7877

0.7683

0.8074

0.5763

0.8630

 

FCGR + TAC

K = 4, lag = 2

0.88

0.7930

0.7635

0.8232

0.5879

0.8671

 

FCGR + DACC

K = 4, lag = 2

0.93

0.7741

0.7690

0.7793

0.5486

0.8506

 

FCGR + TACC

K = 4, lag = 2

0.88

0.7654

0.7576

0.7733

0.5313

0.8461

 

FCGR + PCPseDNC

K = 4, λ = 8, w = 0.5

0.88

0.7988

0.7769

0.8211

0.5987

0.8753

 

FCGR + PCPseTNC

K = 4, λ = 8, w = 0.5

0.85

0.7974

0.7772

0.8179

0.5960

0.8727

 

All features

 

0.93

0.7647

0.7590

0.7705

0.5295

0.8406

  1. “PCA%” means contributing rate of principal component
  2. Best values are in bold