Bioinformatics Advance Access originally published online on April 26, 2005
Bioinformatics 2005 21(14):3174-3175; doi:10.1093/bioinformatics/bti464
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Dynamite extended: two new services to simplify protein dynamic analysis
Laboratory of Molecular Biophysics, Department of Biochemistry, University of Oxford Oxford OX1 3QU, UK
*To whom correspondence should be addressed.
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONSERVICE OVERVIEWREFERENCES |
|---|
Summary: We describe two additional services now available as part of the previously described Dynamite protein dynamics web service. Dynatraj provides principle component analysis and visualization of modes of motion for a user's own ensemble of protein structures, e.g. from Molecular Dynamics, NMR or experimental ensembles. Dynapocket predicts probable configurations of a protein pocket from a single known structure. Both have been provided in response to requests from users for additional functionality from the Dynamite server. Like Dynamite, both are available free of charge to all users.
Availability: Free of charge at http://dynamite.biop.ox.ac.uk/dynamite
Contact: dynadmin{at}biop.ox.ac.uk
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONSERVICE OVERVIEWREFERENCES |
|---|
We have previously (Barrett et al., 2004) described Dynamite, a free web service that allows non-dynamics specialists to get a prediction of the likely equilibrium motions of their protein of interest. For that service, the user submits a structure in the Protein Data Bank (PDB) format (Berman et al., 2002) to a web page and in return receives a set of predictions about protein dynamics and variations in internal rigidity. The Dynamite protocol makes use of Concoord (de Groot et al., 1997) to generate an ensemble which is subsequently subjected to principal component analysis (PCA) (Amadei et al., 1993; Garcia, 1992), the conclusions of which are the basis of an interactive three-dimensional (3D) representation visualized using VMD (Humphrey et al., 1996) (Fig. 1A). As an ensemble generator, Concoord has the advantage of being fast and capable of fully exploring a subset of the available configuration space for the protein. It has some disadvantages however; it overestimates the significance of non-covalent bonds limiting the overall configuration space available for exploration; it does not easily lend itself to the examination of a specific motion or activity of a protein; neither does it offer the degree of control over conditions, such as temperature, ionic concentration, and so on, which is achieved by using Molecular Dynamics (MD) (Moraitakis et al., 2003). A common request from users of Dynamite, who have also conducted their own MD simulations, is for a service that allows the submission of a MD trajectory or NMR ensemble which is then analysed according to the Dynamite protocol. Dynatraj has been implemented to provide this service.
|
A second common request from users of Dynamite was for a generator of configurations of a protein binding pocket from a single starting structure. This request has been provided for in the service Dynapocket.
|
SERVICE OVERVIEW |
|---|
|
TOPAbstractINTRODUCTIONSERVICE OVERVIEWREFERENCES |
|---|
Detailed instructions for the use of both Dynatraj and Dynapocket on the web page referenced above:
Dynatraj. This service requires submission of two files. One is a PDB file containing the first frame of the trajectory to be analysed. The second is a ‘XTC’ format file containing the rest of the trajectory of interest. The PDB file is needed because the XTC file contains no data on atomic types, residue names, etc. Trajectory files tend to be large and, therefore, we encourage users to be conservative in the files that they submit. We have found that 500 frames selected from a trajectory will be plenty to allow a representative analysis to be made, as long as the frames are sensibly distributed throughout the interval under study. Users may find better results by sampling periods of gross conformational change with higher frequency than in periods of stasis. In return, the user receives the same set of analyses that would normally be provided with a Dynamite submission, i.e. porcupine plots (Fig. 1B), animations and covariance webs indicating regions of internal rigidity. This service has been running for some time in beta-version and results have been published in works, such as that of Haider et al. (2005).
As noted above, it is also possible to take a collection of PDB files and concatenate them into a trajectory file for submission to Dynatraj. For example, we have used this approach (Barrett et al., 2005) to identify the chief modes of variation within a group of X-ray structures for CDK2 taken from the PDB (Fig. 1C).
Dynapocket. Dynamite does not offer detailed side chain motion predictions; instead, it focuses on the global motions of a protein. However, for rational drug design detailed predictions of a binding pocket's conformations are of use. For example, Flex-E [Claussen et al., 2001] is a docking program that can use an ensemble of protein pocket structures to dock into instead of a single structure. This is a convenient way of providing for protein flexibility. Unfortunately, it is often the case that only one or two protein structures have been experimentally determined. Dynapocket can take a single structure and provide an ensemble of likely structures upon which methods such as that embodied in Flex-E can work, as represented in Figure 1D. The user submits a single PDB together with a list of the atoms defining the protein pocket. Dynapocket returns 100 configurations of that pocket as a PDB file. Note that, this is not simply an exploration of side chain rotamers; the method accounts for the global motions of the protein such as interdomain flexing. We highlight that Dynapocket, in contrast to the more sophisticated Dynamite and Dynatraj, can be considered as a web-based front end to the existing Concoord program. It is, however, a convenient and sought-after service that we felt worthproviding.
In conclusion, users can now access three services at the URL (http://dynamite.biop.ox.ac.uk/dynamite): Dynamite as a way of predicting protein thermal motions at equilibrium and regions of internal rigidity starting with only a PDB; Dynatraj as a means of extracting dominant motions from MD, NMR and other user-supplied ensembles; and Dynapocket, which simplifies the process of providing an ensemble of likely pocket configurations starting with a single structure.
|
Acknowledgments |
|---|
We thank the authors of CONCOORD, VMD, DSSP and GROMACS for their permission to use these programs to support the services, although the responsibility for any shortcomings of the services is our own. C.P.B. is supported by the BBSRC through a grant to the Oxford Centre for Molecular Sciences.
Received on March 31, 2005; revised on April 21, 2005; accepted on April 22, 2005
|
REFERENCES |
|---|
|
TOPAbstractINTRODUCTIONSERVICE OVERVIEWREFERENCES |
|---|
Amadei, A., et al. (1993) Essential dynamics of proteins. Proteins, 17, 412–425[CrossRef][Web of Science][Medline].
Barrett, C.P. and Noble, M.E. (2005) Molecular motions of human cyclin dependent kinase 2. J. Biol. Chem., 280, 13993–14005
Barrett, C.P., et al. (2004) Dynamite: a simple way to gain insight into protein motions. Acta Crystallogr. D. Biol. Crystallogr., 60, 2280–2287[CrossRef][Medline].
Berman, H.M., et al. (2002) The Protein Data Bank. Acta Crystallogr. D. Biol. Crystallogr., 58, 899–907[CrossRef][Medline].
Claussen, H., et al. (2001) FlexE: efficient molecular docking considering protein structure variations. J. Mol. Biol., 308, 377–395[CrossRef][Web of Science][Medline].
de Groot, B.L., et al. (1997) Prediction of protein conformational freedom from distance constraints. Proteins, 29, 240–251[CrossRef][Web of Science][Medline].
Garcia, A.E. (1992) Large-amplitude nonlinear motions in proteins. Phys. Rev. Lett., 68, 2696–2699[CrossRef][Web of Science][Medline].
Haider, S., et al. (2005) Modelling and simulation studies of the intracellular domains of the inwardly rectifying K+ channels. Biophys. J., 88, 3310–3320[CrossRef][Medline].
Humphrey, W., et al. (1996) VMD: visual molecular dynamics. J. Mol. Graph., 14, 33–38 27–38[CrossRef][Web of Science][Medline].
Moraitakis, G., et al. (2003) Simulated dynamics and biological macromolecules. Rep. Prog. Phys., 66, 383–406[CrossRef].
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Kim, J. Suh, D. Romano, M. H. Truong, K. Mullin, B. Hooli, D. Norton, G. Tesco, K. Elliott, S. L. Wagner, et al. Potential late-onset Alzheimer's disease-associated mutations in the ADAM10 gene attenuate {alpha}-secretase activity Hum. Mol. Genet., October 15, 2009; 18(20): 3987 - 3996. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||