Identification of novel DNA repair proteins via primary sequence, secondary structure, and homology

BMC Bioinformatics

Table 1 Data transformations

Method	Feature Vector Representation	Data Used
P	<φ_s(α^k, Q) >	Primary Sequence
PS	$< ϕ_{s} (α^{k_{1}}, Q_{p}), ϕ_{s} ({C, E, H}^{k_{2}}, Q_{s}) >$	Primary Sequence and Secondary Structure
PF	$< ϕ_{s} (α^{k}, Q), ϕ_{f}^{C, E} (Q) >$	Primary Sequence and Frequency Priors
PH	<φ_s(α^k, Q), φ_H(Q) >	Primary Sequence and Homology
PSH	$< ϕ_{s} (α^{k_{1}}, Q_{p}), ϕ_{s} ({C, E, H}^{k_{2}}, Q_{s}), ϕ_{H} (Q) >$	Primary Structure, Secondary Structure, Homology
BLAST	BLAST	Homology

A summary of the six data processing methods used to identify and classify DNA repair proteins.

ISSN: 1471-2105