Secondary structure prediction

PSIPRED (http://bioinf.cs.ucl.ac.uk/psipred/) and RaptorX Property [44] are used for the secondary structure prediction of the peptide aptamer. PSIPRED involves incorporating two feed-forward neural networks in which analysis output was obtained from PSI-BLAST (position-specific iterated-BLAST). A template-free approach using the RaptorX Property web server was further used to predict the secondary structure properties of the peptide aptamer. The server is enhanced by a machine learning model called DeepCNF (deep convolutional neural fields) to simultaneously predict secondary structure, solvent accessibility and disorder regions.