trRosetta

Introduction

trRosetta is an algorithm for fast and accurate protein structure prediction. It builds the protein structure based on direct energy minimizations with a restrained Rosetta. The restraints include inter-residue distance and orientation distributions, predicted by a deep residual neural network. Homologous templates are included in the network prediction to improve the accuracy for easy targets. In benchmark tests on CASP13 and CAMEO derived sets, trRosetta outperforms all previously described methods. Read more about trRosetta...

The major results returned include (click here for an example):

Submit

  • Provide the protein data (mandatory)
  • Input protein sequence below in FASTA format: (Click for an example input)
    Or upload the protein squence file:


  • Other information (optional)
  • Email: (Optional, where the results will be sent to)

    Target name: (Optional, your given name to this target)


    Do not use templates (check this box if you DO NOT want to use any PDB templates, for some special purposes, such as benchmark.)

    Keep my results private (check this box if you want to keep your job private. A key will be assigned for you to access the results.)


    Update news

  • 12/06/2020, trRosetta-based methods (BAKER, BAKER-ROSETTASERVER, Yang-Server) were assessed in the 14th CASP experiment. The BAKER group was ranked as the second Human group after AlphaFold2; while the BAKER-ROSETTASERVER was ranked as the fourth Server group. The Yang-Server from our lab was ranked among the top 10 Server groups. The major difference between BAKER-ROSETTASERVER and Yang-Server is the former applied an additional step of large-scale refinement on the trRosetta models.
  • 11/23/2020, An automated template detection module was included in the server to improve the modeling accuracy for easy targets.
  • 11/23/2020, A new deep neural network was used, which is about 3% more accurate than the previous version.
  • 09/25/2020, A script 'pdb2npz.py' included to calculate and visualize the interresidue distance and orientation from the input of a PDB structure.
  • 08/27/2020, A script 'top_prob.py' the average probablity of the top L predicted long+medium range contacts.
  • 07/22/2020, As requested by many users, a new script (npz_2_dist.py) was included in the trRosetta package to convert the distance distribution into distance and contact maps.
  • 06/08/2020, Automated MSA selection was included in the server to improve the accuracy by 2-5%.
    ( >> read more ... ).

    Reference

  • J Yang, I Anishchenko, H Park, Z Peng, S Ovchinnikov, D Baker, Improved protein structure prediction using predicted interresidue orientations, PNAS, 117: 1496-1503 (2020).