Prediction of protein motions from amino acid sequence and its application to protein-protein interaction

BMC Structural Biology

Table 3 Summary of evaluation

Protein name	Partner protein name	PROFbval¹	POODLE-S²	FlexPred¹	Proposed method
Staphylococcus A	Human Fc fragment	×	○	×	○
Ran GTPase	RCC1	×	○	×	○
14-3-3	Serotonin N-acteylase	○	○	×	○
Actin	Profilin	×	×	×	×
Erythropoietin	EPO receptor	×	○	○	○
Fab fragment	Flu virus hemagglutinin	○	×	×	○
TGF-beta	TGF-beta receptor	×	○	○	○
Actin	Dnase I	×	×	×	×
Coagulation factor Vlla	Soluble tissue factor	○	×	×	○
Ran GTPase	Importin-beta	×	×	×	○
HPr kinase C-ter domain	HPr	○	○	○	○
HIV1 reverse transcriptase	Fab28	○	○	×	○
Ecotin	D102N trypsin	○	○	○	○
EPO receptor	Erythropoietin	×	×	×	○
Vitamin D bindings	Actin	○	×	○	○
Nitrogenase Fe	Nitrogenase Mo-Fe protein	○	×	×	○
CDK2 kinase	CDK inhibitor 3	×	×	×	○
Gelsolin	Actin	○	○	○	○
Importin-beta	Ran GTPase	○	×	×	○
Hirustatin	Kallikrein	×	×	×	×

'○' means that the predictor detects at least one overlap between the observed conformational change region and predicted high motion region. Conversely, '×' means that there is no overlap. PROFbval, POODLE-S, and FlexPred were performed with the default parameters; then candidate conformational change regions were determined as follows.
¹ Three or more consecutive residues predicted to be flexible.
² Disordered region longer than three residues.

ISSN: 1472-6807