Skip to main content

Table 1 Related works of predicting DNA-binding domains/proteins from sequences

From: Predicting and analyzing DNA-binding domains using a systematic approach to identifying a set of informative physicochemical and biochemical properties

Reference

Sequence type

Identity

Feature number

Representation

Feature type

Classifier

Shao et al. 2009 [8]

protein

25%

343

Seven class Conjoint triad

PCP

SVM

Fang et al. 2008 [6]

protein

35%

40

Pseudo-AA composition

PCP

SVM

Yu et al. 2006 [9]

protein

25%

132

Combined descriptors

PCP

SVM

Cai et al 2003 [5]

protein

40%

40

Pseudo-AA composition

PCP

SVM

Kumar et al. 2007 [7]

domain and protein

25%

400

PSSM

PSSM

SVM

Auto-IDPCPs

domain and protein

25%

m*

Mean value of sequence#

PCP

SVM

  1. PCP: physicochemical property and biochemical property
  2. *: a small number (<30) of feature vectors selected from 531 vectors
  3. #: The averaged value of amino acids in a sequence for one property