Protein Binding Site Prediction by Combining Hidden Markov Support Vector Machine and Profile-Based Propensities

Table 1

Summary of six data sets.

Data set	Chains	Res.	Surface res.	Interface res.
Heterocomplex I^a	504	109829	92797	26085
Homocomplex I	620	172917	141295	38170
Mix^b I	1124	282746	234092	64255
Heterocomplex II^c	504	109829	92797	32386
Homocomplex II	620	172917	141295	45633
Mix II	1124	282746	234092	78019

aType I data set with minor interface as negative samples.
^bThe mixed data set of heterocomplexes and homocomplexes.
^cType II data set with minor interface as positive samples.