Critical Assessment of Automated Structure Determination
of Proteins from NMR data
The CASD-NMR manifesto has been published in Nature Methods.
Scope
CASD-NMR is a rolling community-wide experiment involving developers of software tools / protocols for the automated calculation of protein structures from NMR data. The goal of CASD-NMR is to help advance the relevant methodology in order to reach the level of quality and reliability required for direct structure deposition in the PDB. CASD-NMR will also produce extensive data sets that will be useful to develop better methods for NMR structure validation.
How does it work?
CASD-NMR collects and makes available to the participants NMR data sets that can be successfully used for protein structure determination. All data sets in CASD-NMR lead to a satisfactory structure through the traditional manually curated procedures. They are released as blind data sets, i.e., the corresponding protein structure is not publicly available at the time of release. The participants are given 8 weeks to generate a structure using fully automated methods as if they would directly deposit them into the PDB. The coordinates are stored in a central database at the Magnetic Resonance Center in Florence. The results will be analyzed through various validation tools and the corresponding outputs will be made publicly available. After this time, the corresponding manually solved structure (the “reference structure”) will normally be released from the PDB. CASD-NMR is a rolling experiment: data sets will be released regularly over one year time and then a global assessment of the results will be performed. At the beginning, we envisage to release one data set per month.
How can you contribute?
A minimum number of data sets will be provided by the Northeast Structural Genomics consortium of the NIH Protein Structure Initiative (PSI). These experimental data will be available from the database at the Magnetic Resonance Center in Florence and through the PSI Structural Genomics Knowledgebase. We invite any NMR group that is about to deposit (or has just deposited) a structure in the PDB to contribute a blind test case to CASD-NMR on the basis that the reference structure is kept on hold for at least 8 weeks after data submission. To register as a data contributor please check the instructions first (also addressing how your data will be handled).
How can you join in?
Just register at this link. You will be notified by email as soon as a new data set is available as well as when the reference structure will be released in the public domain. You can join in any time: all the data sets remain available, also after the reference structure is publicly available. Therefore, at any point in time you can work also on previous data sets; the corresponding structures will be flagged as late submissions. Check also the instructions for structure submission at the bottom of the page.
Who joined already?
Click here for a list of the participants who registered already.
Please click on the links below for details
- Instruction for data submission
- Instruction for structure submission
- Validation tools: CING, PSVS
Click here to go to the data
The CASD-NMR manifesto has been published in Nature Methods.