<em>In Silico</em> platform for predicting and designing therapeutic peptides for drug delivery — ASN Events
  1. Schedule
  2. Session 7 - Rapid Fire Hot Topics
  3. In Silico platform for predicting and designing therapeutic peptides for drug delivery

In Silico platform for predicting and designing therapeutic peptides for drug delivery (#50)

Ankur Gautam 1 , Kumardeep Chaudhary 1 , Rahul Kumar 1 , Harinder Singh 1 , Pallavi Kapoor 1 , Atul Tyagi 1 , Raghava G.P.S. 1
  1. Institute of Microbial Technology, Chandigarh, Chd, India

Many therapeutic molecules like DNA, proteins, and small molecules do not reach clinical trials due to poor delivery and low bioavailability. However, small peptides known as cell-penetrating peptides (CPPs) have remarkable ability to traverse biological membranes without significant membrane damage. In addition, CPPs are capable of transporting a wide range of therapeutic molecules into the cell interior, thus offering a great potential as versatile drug delivery vehicles. Keeping in mind the huge therapeutic importance of CPPs, we have developed a database of CPPs-“CPPsite”, a unique repository of its kind , which provides comprehensive information about experimentally validated 843 CPPs. A wide range of user-friendly tools like searching, browsing, analyzing, and mapping tools have been incorporated in CPPsite. Next, we have developed a support vector machine (SVM) based algorithm-“CellPPD”, which is a very useful platform for predicting and designing CPPs. Preliminary analysis revealed that certain residues are dominated, and few residues (e.g. Arg, Lys, Pro, Trp, Leu, and Ala) are preferred at specific locations in CPPs. Therefore, SVM models were developed using amino acid composition, dipeptide composition, and binary profiles as input features. In addition, we have identified various motifs in CPPs, and used this information for developing a hybrid prediction model (Motif + SVM). All models were evaluated using five-fold cross-validation technique. We have achieved maximum accuracy of 97.40% using the hybrid model. CellPPD is the first web server in the public domain where users not only can predict CPPs but also can design better CPP analogs with desired physicochemical properties. CellPPD will generate all possible substitution mutants of each submitted peptide with SVM scores and physicochemical properties and allows user to select the best analogues. In conclusion, both CPPsite and CellPPD will be helpful in designing efficient CPPs for delivery of therapeutic molecules.

Availability:

CPPsite: http://crdd.osdd.net/raghava/cppsite/1 

CellPPD: http://crdd.osdd.net/raghava/cellppd/2 

  1. Gautam A, Singh H, Tyagi A, Chaudhary K, Kumar R, Kapoor P, Raghava GP. (2012). CPPsite: a curated database of cell penetrating peptides. Database (Oxford). 2012:bas015.
  2. Gautam A, Chaudhary K, Kumar R, Sharma A, Kapoor P, Tyagi A; Open source drug discovery consortium, Raghava GP. (2013) In silico approaches for designing highly effective cell penetrating peptides. J Transl Med. 11:74.