Joint Meeting of the Research Center Directors and the PSI Advisory Committee (PSIAC)

PSI Annual Meeting

National Institute of General Medical Sciences
National Institutes of Health

December 5-6, 2007

The Protein Structure Initiative is now in the third year of its second phase. Four large-scale centers have been established to focus on the acquisition of a large collection of protein structures and the continued improvement of structure-determination pipelines. Six specialized centers have also been charged with tackling problem areas that currently limit high-throughput structure determination of "difficult" proteins. A knowledgebase has recently been established to help better disseminate results from the PSI. Two centers focus on protein modeling and a PSI materials repository has been established. The directors (Principal Investigators) of all of the centers assembled at the annual meeting on December 5-6, 2007, to review accomplishments and outline plans for the future. The PSI Advisory Committee met during the afternoon of December 6 in closed session to evaluate the reports and discuss future plans.

Scope of the Present Report

During the spring of 2007 representative members of the Advisory Committee, sometimes with "outside" experts, conducted site visits to all of the large-scale and specialized centers. Reports on each center have already been submitted to Council. (Overall, the visiting groups were very encouraged by the progress to date.) In addition, on September 24, 2007, NIGMS conducted an assessment of the PSI. Materials from this review will also be available to Council.

Given the above materials, the present report will be brief.

Large-scale Research Centers

Progress in all four of the large-scale centers, reviewed by Principal Investigators Stephen Burley, Andrezj Joachimiak, Guy Montelione and Ian Wilson, continues to be impressive. Each group currently expects to determine approximately 150-190 structures this year. Also, there is continued technology development on many different fronts. The four large-scale center directors also described progress and goals of their biomedical theme projects, which aim at structural coverage of protein families of high biomedical interest. These include: cancer network proteins, complete coverage of a small organism (Thermatoga maritima), pathogens, ocean genome survey proteins, and phosphatases.

Specialized Centers

The directors of the specialized centers, or their representatives, Lance Stewart, John Markley, Mike Malkowski (for George DeTitta), Robert Stroud, Thomas Terwilliger, and Wayne Hendrickson, summarized recent progress in their respective centers. Perhaps one of the most encouraging developments has been the ability to determine the structures of at least a handful of intrinsic membrane proteins. Other developments include the development of methods for the production of SeMet proteins in a yeast expression system, chaperone-assisted crystallization, soluble protein engineering, construction of a compact light source and cell-free protein production, among others. These early successes provide a foundation on which to build more robust methods.

Homology Modeling Centers

The directors of the two homology modeling centers, Roland Dunbrack and Adam Godzik, summarized their plans and initial progress on this important component that could increase the leverage of the PSI structures.

Materials Repository

Joshua LaBaer described the goals of the new center for the collection, processing, analysis, and distribution of plasmids from PSI experimental work to the scientific community.

Accessibility to PSI-generated Information: The Knowledgebase

The Advisory Committee has been concerned for some time at the need to make the work of the PSI Centers (both large-scale and specialized) known and accessible to the broad biomedical community. This need is crucial and was discussed at length. A key component is the dissemination of information via the PSI Knowledgebase.

Helen Berman became director of the Knowledgebase only a few months ago, but has already organized several components. The Knowledgebase will consist of eight modules, including target selection, materials repository, models, annotations, metrics, technology, and outreach. It should be an important step forward in making the PSI useful to biologists who are not experienced in structural biology. At the same time, other ways to make better use of the wealth of the information that is being accumulated would be welcome.

PSI-2 Network Steering Committee: Target Selection

The Chair of the PSI-2 Network Steering Committee, Wayne Hendrickson, organized a series of lively talks on target selection with contributions from Andrzej Joachimiak, Guy Montelione, Thomas Terwilliger, John Moult, and Adam Godzik. The most appropriate way in which targets for structure determination should be selected by the PSI is an active subject for discussion. Initially it seemed most appropriate to focus on representative proteins from "BIG" gene families for which no representative structure was already in the PDB. As more gene sequences have become available, however, it has become apparent that there are not only very large ("MEGA") families but even "SUPER MEGA" families with over 10,000 members. In such cases it is appropriate to determine the structures of additional representatives to provide reasonable coverage within such large groups. Other MEGA families will be chosen for their functional and biomedical relevance. The possible impact that metagenomics may have on target selection ("META" families) was discussed in presentations by Claire Fraser-Liggett and Adam Godzik.


Overall, the Advisory Committee remains highly impressed with the progress, coordination, and productivity of the large-scale centers. The continued technology development in both the large-scale and specialized centers is also extremely impressive.

In general the Advisory Committee feels that the current choice of targets by the PSI is good. The way in which representatives of the four large-scale centers apportion targets in a mutually equitable fashion is highly commendable. One member of the Advisory Committee commented that there could be a greater awareness of biological relevance in target selection. Another member commented that it would be better to have a consistent overall policy for the nomination of targets by the community. Overall, the Advisory Committee does not have a clear consensus on the details of target selection. It is a complicated and challenging task. If any changes are to be made they should be in the nature of fine tuning, not radical alterations.

The Advisory Committee also discussed at length the need to make the results of the PSI (including structures, techniques and materials) more available to the community at large. The Knowledgebase and the Materials Repository should help substantially but other approaches would be welcome.

Report submitted by Protein Structure Initiative Advisory Committee Chair Brian Matthews on January 18, 2008.