RM2TS

RM2TS+ is a composite web suite for predicting the tertiary structure of a protein from input amino acid sequence. Read more..

Current session Id: 1511587168

Use this session id to retrieve the files later.


  • Intial Inputs
    • Input amino-acid sequence 
    • Input secondary structure sequence 
  • Method Selection
    • Quick method 
    • Long method 
Initial Inputs
Click to
upload
or
Please provide a sequence information (single letter amino acid code) in plain text(*.txt) or FASTA(*.fasta) format. Load sample protein sequence and secondary structure data

   

Click to
upload
or
Please upload or provide a secondary strucuture information data(*.txt)
State Stretch
3 state: HEC From To



Secondary structure prediction takes 8-10 extra minutes. Kindly bear with us.


Methods


The input amino acid sequence (of any length)is feeded to the server and then secondary structure prediction algorithm runs on it. After Secondary structure prediction is completed, template search is done using BLASTP algorithm. The x, y, z coordinates for the matched regions are picked up and for each template, the missing regions are predicted based on precompiled libraries of Higher Order Ramachandran Maps ( up to 11-mer libraries are annotated). From the phi, psi dihedrals predicted, the x, y, z coordinates were built and then different fragments were generated. Thus the hybrid based fragment assembly of template and library based stretches were done which yielded about ~500 structures as decoy set. The top 5 (best structures) from the decoy selection was done on the basis of ProtSav scoring. These five structures were then provided to the user as output. The entire algorithm takes about 10-15 minutes on a 250 amino acid sequence.


The user gives amino acid sequence input of any length. Secondary structure prediction is done on that input sequence. Exhaustive template search is done using freely available programs. These methods generate about 500 templates for the query sequence. For each template the missing part is predicted using precompiled libraries of Higher Order Ramachandran Maps. (This is the ab initio segment of the algorithm where 11-mer fragment window size is used for dihedral prediction). Fragment assembly based on comparative modelling as well as ab initio structure generation yields ~20000 structures in the decoy bin. ProTSAV scoring segregates 100 best structures on which energy minimization is done. Further scoring and top 5 structure selection done through ProtSAV cumulative score. These 5 structures are provided to the user. The entire algorithm takes about 40 minutes on a 250 amino acid sequence ( standard sequence length).


Results


Minimized Structure(s)