This feature vector is constructed by fusing consensus sequence (CS), segmented PsePSSM, and
segmented auto-covariance transformation (ACT) based on PSSM. by consensus sequence a 40-dimensional feature
vector is obtained, in segmented PsePSSM group, by dividing PSSM Matrix to 2 and 3 segments a 380-dimensional
feature vector is obtained and in ACT group, similar to the previous group at first PSSM Matrix is divided
to 2 and 3 segments then a feature vector of length 280 is obtained.eventually by fusing these features a
700-dimensional feature vector is obtained.
Usage
CS_PSe_PSSM(pssm_name, vec_name)
Arguments
pssm_name
name of PSSM Matrix file
vec_name
a character that user imports to specify kind of feature vector which it can be varied between
four values
Value
feature vector that its length depends on the vec_name which user imports. vec_name can be one
of "cspssm", "segmented_psepssm", "segmented_acpssm", "total".
Details
If vec_name equals to "segmented_psepssm" then a feature vector of length 380 is obtained.
if vec_name equals to "segmented_acpssm" then a feature vector of length 280 is obtained, and if vec_name
equals to "cspssm" the obtained feature vector would be of length 40 eventually if vec_name equals to
"total" then feature vector would be of length 700.
References
Y. Liang, S. Liu, S. J. C. Zhang, and m. m. i. medicine, "Prediction of protein structural
classes for low-similarity sequences based on consensus sequence and segmented PSSM," vol. 2015, 2015.