We are grateful to the selected predictors for taking the time to answer this questionnaire.
Results of these questionnaire are HERE
Please give a concise summary of the method you tended to use for NF or NF/FR targets.
1. Would you classify your method as exclusively fold recognition (i.e always using an alignment to a template that covers most of the target sequence)?
Some methods incorporate a fold recognition step, or use both ab initio and fold recognition protocols to predict protein structure. We do not consider these 'exclusively' fold recognition, but they are noted in the detailed table on our website.
2. Did you alter your prediction method based on some prior-classification of targets (considering only those above)?
If 'YES', could you let us know how you classified the targets.
Some groups treated targets differently, depending on whether they could find a suitable template, etc. In this question, we wanted to identify methods that changed depending on the target.
3. Does your method use secondary structure prediction?
If 'YES', what method did you use? (eg. PSIpred, PHD, etc.)
4. Does your method have any manual intervention (eg. adjusting alignments, inspecting models, etc.)?
If 'YES', please describe how much manual intervention was used.
5. Does your method use homologous sequence information for the target sequence?
If 'YES', please describe how.
6. Did you split target sequences into domains before predicting structure?
If 'YES', what method did you use to define domains?
Some groups started with the full-length target sequence, and only separated into domains when (for example) a threading method identified more than one template spanning the length of the protein. In this case, we do not consider this a dedicated domain-identification step. This is noted in the detailed table.
7. Is your method a fragment-based approach to predicting structure?
If 'YES', what fragment library did you use?
8. Did you use any publically-available server to assist in your predictions?
If 'YES', which servers did you use? What were they used for? (ie. screening targets, finding templates/fragments, etc.).
Most groups used servers to assist in some part of their prediction, from sequence search tools (BLAST, PSI-BLAST, etc.) to secondary structure prediction (PSIPred, JPred, etc.). With this question, we were more concerned with groups that used other group's fold prediction servers (ie. not methods developed in their own labs) to assist in the prediction process. In most cases, this is done to choose good templates, or to remove targets that are obviously comparative modelling or fold recognition targets. This is where metaservers could list the servers upon which their method is based. Other publically-available servers such as SCOP, PROSITE, etc., need not be included here.
9. Does your method include lattice-based representations of proteins?
10. Does your method include threading-type potentials (e.g. pair-potentials, etc.)?
If 'YES', please describe the potentials used.
Some groups used some kind of scoring function that might not necessarily be considered a threading-type potential. These are also noted in the more detailed table on the website.
11. Does your method include steps for relaxation, optimization, minimization?
If 'YES', please describe.
12. Please add any other information about your method that you feel is important.
Many groups supplied extensive information about their methods, much of which can be found in the abstracts document on the Lawrence Livermore CASP5 site.