I am working on K-Post Translational Modification of protein. I have gone thorugh a lot of papers. But there is no detailed view of how to extract feature like PseAAC. Any python code /package is requested. Additionally, after window splitting into 9,13,15,17 etc.(mine is 13) sequence would be like (R-13, R-12,..., R-1, K, R+1,.., R+12, R+13) something like that. My question is, during applying the Pseudo Amino Acid Composition, should I extract feature separately for the left sequence stream(which is of length 13) and right sequence stream(length of 13)? Then merge the output together for classification? Or I just input the whole windowed sequence of length 27?