AIDA: Ab Initio Domain Assembly Server

Announcement(09-12-2016): Recently a few jobs were terminated due to a server problem. Now the problem is fixed, please resubmit your jobs. For users who submitted jobs with emails, your jobs were resubmitted and you will receive the result by email once the job is finished. Sorry for the inconvenience!

A large portion of proteins are multi-domain proteins, where individual domain structures often interact with each other. Using knowledge-based ab initio potential, AIDA attempts to find the best domain arrangements while maintaining chain connectivity. The structures themselves are kept rigid during the energy minimization simulation, treating only the link regions as completely flexible. The AIDA server is able to work with proteins containing discontinuous domains by fixing the relative positions of the discontinuous domains. The server also supports structure assembly from sequence only. To do this uncovered sequence domains are iteratively split to align with the PDB template found by the FFAS-3D fold recognition program. Furthermore, AIDA also supports restraint-guided simulation producing a final assembly satisfying user-specified inter-domain distance restraints.

1).  Please paste the full amino-acid sequence in fasta format.(sample input)

    .   10    .   20    .   30    .   40    .   50    .   60    .   70    .   80    .   90    .  100    .  110    .  120    .  130    .  140    .  150

2).  Please upload compressed domain models (pmodel1.pdb, pmodel2.pdb...pmodeln.pdb) in one file (*tar.bz2, *.tar.gz, *.zip, *.bz2 or *.gz) (optional). (sample input)

Coordinates of few residues in the two ends of each domain structure could be deleted to make the assembly more flexible. No gap is allowed in the middle of each domain structure.

Numbering of each residue in domain structures is its position in the full-length sequence. Download the package and run "renumber" program to handle it.

3).   check this box only if you upload three domain models and pmodel1.pdb+pmodel3.pdb constitute a discontinuous domain. explanation

4).  Please upload Cα distance restraints to assist assembly (optional). (sample input)

5). Email: (optional, where result will be sent to)

Click [help page] to see the detailed explanation of input and output.

Click [example1] to see an example of assembly of 4 continuous domains.

Click [example2] to see an example of assembly of 1 continuous domain with 1 discontinuous domain.

Click [example3] to see the automated prediction of a single-domain protein.

Click [example4] to see the automated prediction of a protein with 2 continuous domains.

Click [example5] to see the automated prediction of a protein with 1 continuous domain and 1 discontinuous domain.

The AIDA downloadable package is released here:

Download executable files (41MB)

Download benchmark set (19MB)

Selected papers from Godzik Lab
Ying Zhang, Ines Thiele, Dana Weekes, Zhanwen Li, Lukasz Jaroszewski, Krzysztof Ginalski, Ashley Deacon, John Wooley, Scott Lesley, Ian Wilson, Bernhard Palsson, Andrei Osterman, Adam Godzik. Three-Dimensional Structural View of the Central Metabolic Network of Thermotoga maritima. Science. (2009) 325(5947):1544-9.

Dong Xu, Lukasz Jaroszewski, Zhanwen Li, Adam Godzik. FFAS-3D: Improving fold recognition by including optimized structural features and template re-ranking. Bioinformatics (2013) doi: 10.1093.