PEP2D: Peptide Secondary Structure Prediction using Evolutionary Information

PEP2D: Peptide Secondary Structure Prediction using Evolutionary Information Overview PEP2D is a computational method developed specifically for predicting the secondary structure of peptides using evolutionary information and machine learning approaches. The platform was designed because existing protein secondary structure prediction methods perform poorly on short peptides. PEP2D predicts peptide secondary structures into three major classes: Helix (H) Beta-sheet (E) Coil (C) The method uses: Binary profile features Evolutionary information (PSSM profiles) Random Forest IBK Artificial Neural Networks (ANN) Web Server: https://webs.iiitd.edu.in/raghava/pep2d/ Research Paper Title: Peptide Secondary Structure Prediction using Evolutionary Information Authors:Harinder Singh, Sandeep Singh and Gajendra Pal Singh Raghava Publication Type: bioRxiv Preprint (2019) Correct DOI:https://doi.org/10.1101/558791https:// github.com/Piyushh1104/PEP2D.git Dataset Information PEP2D Dataset The dataset was collected from Protein Data Bank (PDB). Initial dataset: 5778 peptide chains Final dataset: 3107 unique peptides Length range: 5–50 amino acids