Description of the experiment
The main goal of CASP is to obtain an in-depth and objective assessment of our current
abilities and inabilities in the area of protein structure prediction.
To this end, participants will predict as much as possible about a set of soon to be known
structures. These will be true predictions, not ‘post-dictions’ made on
already known structures.
CASP8 will particularly address the following questions:
-
Are the models produced similar to the corresponding experimental
structure?
-
Is the mapping of the target sequence onto the proposed structure
(i.e. the alignment) correct?
-
Have similar structures that a model can be based on been identified?
-
Are comparative models more accurate than can be obtained by simply
copying the best template?
-
Has there been progress from the earlier CASPs?
-
What methods are most effective?
-
Where can future effort be most productively focused?
Tertiary structure predictions. As in CASP7, we will use two broad
categories to define modeling difficulty of the targets. The 'Template based modeling'
category will include domains where a suitable template can be identified that
covers all or nearly all of the target. The 'Template free modeling' category will
include models of proteins for which no suitable template can be identified.
High resolution models. This relatively new category (introduced in CASP7)
will include a subset of tertiary structure models in which the backbone is sufficiently
accurate so that the details of the side chains, loops, and active sites can be meaningfully
assessed. In addition, success in refining any models beyond the quality obtained by simply
copying from a single template will also be analyzed. As in CASP7, we will also select a
subset of targets for refinement from among the best models received during the prediction
season, and these best models will be reissued as starting structures for refinement.
A separate assessor will judge these high accuracy modeling cases.
Other predictions. As in previous CASPs, we will be assessing
the ability of predictors to define boundaries of structural domains,
detect residue-residue contacts and identify disordered regions in target proteins.
Function prediction, a new category introduced in CASP6, will be restricted to the prediction
of binding sites. We will also evaluate ability of predictors to judge on the quality of models
(without knowing native structures) and the reliability of predicting certain residues in the
structure. Following recommendations of CASP7 predictor meeting, quaternary structure predictions
(multimers) will not be the part of the CASP8 experiment.
Registration for the experiment will start in the last week of March.
The first prediction targets will be released not earlier than May 5;
the last prediction targets will be released not later than July 18;
prediction season will end not later than August 1.
The CASP meeting will take place on December 3-7, and approximately one
month before that, groups with the most accurate and interesting
predictions will receive invitations to give talks.
There will also be discussion of predictions and methods
on the FORCASP web site.
Participation is open to all. To register for CASP8, you will have to register
with the Prediction Center first, and only then - for CASP8 (the basic information
from the Prediction Center registration will be passed along to the CASP8
registration server). Intending participants, and those interested in receiving mailings
concerning progress of the experiment will have the possibility to register for the
experiment from this web page.
Predictors with servers will be requested to register as soon as the registration
page is available as we are planning on having a dry run for servers in the first
part of April.
For the experiment to succeed, it is essential that we obtain the help of the experimental
community. As in previous CASPs, we will invite protein crystallographers and NMR
spectroscopists to provide details of structures they expect to have made public before
September 1, 2008. A target submission form will be available at this web site in mid-April.
Prediction targets will be made available through this web site. All targets will be assigned
an expiry date, and predictions must be received and accepted before that expiration date.
According to the recommendations of the CASP7 predictors meeting, we will be
releasing about 50 targets for evaluation in the human and server
track and as many targets as we can for evaluation in the server-only track
(including "human and server" targets). The human-expert groups wishing to take the challenge
and predict all of the released targets (including the "server-only" ones) are welcome
to do so but the evaluation accent will be placed on the selected targets. Subject
to the availability, we will give priority to targets containing low homology domains
for inclusion in the human-expert modeling experiment.
Predictions must be submitted to this web site or through the email in
CASP format.
For 3D coordinate predictions, this is a simple PDB-like file with consecutive
numbering of residues 1 -> N and a small number of required headers.
As in previous CASPs, independent assessors will evaluate the predictions.
Assessors will be provided with the results
of numerical evaluation of predictions, and will judge the results
primarily on that basis. They will be asked to focus particularly on the
effectiveness of different methods. Numerical evaluation criteria will
as far as possible be similar to those used in previous CASPs, although
the assessors may be permitted to introduce some additional ones.
There are four assessors, focusing on the following areas of prediction:
Jane Richardson and group (Duke University, North Carolina, USA) - template based modeling and quality assessment prediction
Joel Sussman (The Weizmann Institute of Science, Israel) - for template free modeling and disorder prediction
Ken Dill (University of California, San Francisco, USA) - prediction refinement and physics-based prediction methods
Michael Tress (CNIO, Madrid) - prediction of function, domain and contacts
In accordance with CASP policy, assessors are not directly
involved in the organization of the experiment, nor can they take part
in the experiment as predictors. Predictors must not
contact assessors directly with queries, but rather these should be sent to the
email address. Click here
for the list of previous CASP assessors.
All CASP predictions and evaluations will be made available through
this web site shortly before the meeting.
The proceedings of the meeting will be published.
All participants will also be encouraged to fully report their results
and methods on the
FORCASP web site. Contributions to the site will be discussed and scored
by other predictors, and this material will be taken into account in choosing
some presentations at the meeting.
The meeting to discuss results of the experiment will be held on the island of Sardinia,
Italy, December 3-7, 2008 (starting at 6pm on the 3rd and ending at 12pm on the 7th, tentatively).
The meeting will take place at
Hotel Setar ,
near Cagliari. The hotel is located in Quartu S. Elena, 10 kilometers from Cagliari
and 15 from the Cagliari airport. As customary in CASP, the hotel is on the beach!
A lot of useful information about the meeting venue can be found
here .
The meeting will have limited attendance (about 300 participants)
and precedence will be given to active predictors.
Some financial assistance will be available for the most successful predictors and
students.
John Moult, CARB, University of Maryland, USA
Krzysztof Fidelis, University of California, Davis, USA
Andriy Kryshtafovych, University of California, Davis, USA
Burkhard Rost, Columbia University, New York, USA
Anna Tramontano, University of Rome, Italy
|
