CASP 10: Call for targets to assess the state of the art in protein structure modeling

As many of you know, CASP (Critical Assessment of Structure prediction) has been assessing the state of the art in modeling protein structure from sequence since 1994, running a community experiment once every two years. In these experiments, information on soon to be solved structures is collected from the experimental community, and the sequence data are passed to the structure modeling community so that blind predictions of structure and can collected and assessed (1).  Over that period CASP has seen enormous progress in the quality of modeled structures (2), but many problems remain. CASP is only possible because of the generous participation of the experimental community in providing the modeling targets (3). We recently appealed for your help in providing novel fold and membrane protein targets for the ongoing CASP ROLL process. That is very successful, with 57 prediction teams participating and 26 targets already released, thanks to all of you who have contributed.

The prediction season for the next full biannual experiment, CASP 10, begins May 1, and so we are now asking for your help again, this time in reaching our goal of releasing at least varied 100 targets in a the three-month period. We need all sorts of targets, spanning the categories below:

1. (More than ever) novel folds and membrane protein targets – even with the extended collection season provided by CASP ROLL, there will be still a shortage. There are some interesting methods developments for ab initio modeling, and it is important to be able to decisively evaluate their effectiveness.

2. A diversity of comparative modeling targets. Cases where the there is fairly high sequence identity (30-50%) between the target structure and an available template are valuable for testing the degree to which model accuracy can approach that of experiment, particularly in functionally critical regions. Cases with lower sequence identity to template, right down to undetectable, are valuable for testing the ability of the methods to detect remote homologs, to overcome challenging alignment difficulties, to make use of multiple templates, and to build regions of the structure not obviously available from a template.

3. Targets containing significant amounts of disordered structure, so as to test the ability of methods to identify these regions. Disorder identified by absence of density in crystallography is current the main source and very useful, but its important to have cases where the nature of the disorder has been established by other means, such as NMR.

4. Some targets will also be used to test methods of structure refinement. In these cases, the best model received for a target will be released to the refinement community, who will subsequently submit new models. This too is an area where there have been some exciting developments in the last year, so we are hoping for significant progress. Refinement targets are selected based on the nature of the errors in the initial models. Because of the extended process, these targets need to be available for longer before release of the experimental structure.

For those of you who have not provided targets to CASP before, the procedure is simple. There are three ways to submit targets: (1) Go the target submission web page and fill in the easy form; (2) When you submit your co-ordinates to the PDB, tick the ‘CASP HOLD’ box, automatically setting up a target entry; (3) Send an email the prediction center with details. If you have queries, there is a very experienced prediction center staff to deal with them. The key thing is timing: We need a window of at least four weeks between receiving information about a target and the release of your experimental structure. A longer window – up to eight weeks - is often useful, for example so that a target can be used to test refinement methods, but is not essential (note that using the PDB target submission route automatically selects eight weeks). We don’t need your experimental structure in advance of its release by the PDB, provided that will happen by the end of August. Please consult the target submission page for more details: http://www.predictioncenter.org/casp10/targets_submission.cgi

There are already over 200 prediction teams signed for CASP 10, and so any targets provided will receive plenty of attention. If you have a suitable target now, we are ready to receive it. If new targets come up before July 17, we would also love to have them. (After July 17, we will continue with CASP ROLL for novel folds and membrane proteins). So please get in touch whenever a suitable opportunity arises, and help improve modeling methods.

Thanks,
CASP organizing committee
John Moult, IBBR, University of Maryland, USA
Krzysztof Fidelis, University of California, Davis, USA
Andriy Kryshtafovych, University of California, Davis, USA
Torsten Schwede, SIB & Biozentrum University of Basel, Switzerland
Anna Tramontano, University of Rome, Italy

Get in touch: [email protected]
More information: http://www.predictioncenter.org/casp10/index.cgi
CASP Roll targets so far: http://www.predictioncenter.org/casprol/targetlist.cgi
Submit a target: http://www.predictioncenter.org/casp10/targets_submission.cgi

1. Critical assessment of methods of protein structure prediction (CASP)--round IX.
Moult J, Fidelis K, Kryshtafovych A, Tramontano A.
Proteins. 2011;79 Suppl 10:1-5. doi: 10.1002/prot.23200. Epub 2011 Oct 14.

2. CASP9 results compared to those of previous CASP experiments.
Kryshtafovych A, Fidelis K, Moult J.
Proteins. 2011;79 Suppl 10:196-207. doi: 10.1002/prot.23182. Epub 2011 Oct 14

3. Target highlights in CASP9: Experimental target structures for the critical assessment of techniques for protein structure prediction.
Kryshtafovych A et al.
Proteins 2011;79 Suppl 10:6-20. doi: 10.1002/prot.23196. Epub 2011 Oct 21.