Joint Meeting of the Research Center Directors and the Protein Structure Initiative Advisory Committee (PSIAC)

PSI Annual Meeting

National Institute of General Medical Sciences
National Institutes of Health

December 8-9, 2005

The Protein Structure Initiative has now entered its second phase (PSI-2). Four large-scale centers have been selected 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 selected to tackle problem areas that currently limit high-throughput structure determination of "difficult" proteins. The PIs of all of the centers assembled on December 8, 2005, to review accomplishments to date and outline plans for the future. The PSI Advisory Council met in closed session the following day to review the information that had been presented.

PSI Supplementary Activities

Following an overview from John Norvell of the PSI plans and goals, summaries were presented of a number of activities which are intended to support and enhance the overall objectives. Cathy Lewis summarized the development of a proposed materials repository that will focus on cloned sequences and expression clones. An RFA for this initiative has just been sent out. Jerry Li discussed progress toward the development of a "knowledge-based" data base as well as improved approaches for comparative modeling. The need to make the results from the PSI more readily available and more relevant to the biomedical community in general is critically important and it is hoped that these initiatives will help address this concern. He also described the RFA for centers to improve homology modeling. Charles Edmonds discussed protein production and crystallization workshops. Workshops of this type continue to be very effective not only in sharing information between the different PSI centers, but also in fostering a sense of unity for the initiative as a whole. He also described the three new state-of-the-art beamlines being developed at the APS by NIGMS and NCI. The PSI centers will have some dedicated beamtime on these.

Protein Data Bank

Helen Berman discussed recent activities at the Protein Data Bank. Dr. Berman emphasized that she would like to devote additional effort on functional annotation of structures derived from the PSI should funds be made available. Dr. Berman's group has established an outstanding track record in catering to the needs of the structural biology community and it is very encouraging that she is interested in better serving the needs of a broader constituency. She was questioned about the need to make the biomedical community more aware of the accomplishments of the PSI.

Large-Scale Research Centers

Ian Wilson, Andrzej Joachimiak, Stephen Burley and Gaetano Montelione discussed progress in the four PSI-2 research centers. These four presentations were, in total, most impressive and bode well for the overall success of this part of PSI-2. To date well over 1300 structures have been determined within the PSI and deposited in the PDB, with over 65% unique. Methods to clone, express, purify, crystallize and determine structures continue to be better automated and made more efficient.

The Target Selection Committee, which includes the PIs of the four large-scale centers, has met and assigned approximately 300 distinct targets to each of the large-scale centers as an initial set of assignments. These targets were chosen from the approximately 1260 largest Pfam families and should go a long way toward addressing one of the main objectives of PSI-2. It is recognized that this division of targets may not be "perfect" but it is very much a step in the right direction. In particular it will provide the large-scale centers with an immediate list of targets on which they can focus during the next year or so. This will allow additional time for "fine-tuning" the list of proposed targets. It also needs to be made clear that the PSI is not attempting to "lay claim" to these targets. Rather, any help from other groups in tackling the large Pfam families will be welcome.

Specialized Centers

The PIs of the six specialized centers gave brief reports. Since these groups have only been in place for a few months the reviews tended to focus on overall objectives rather than actual accomplishments. The creation of these centers adds to the PSI a number of individuals with substantial experience in technology development and in structure determination of highly interesting but technically intractable proteins. It is hoped that future workshops and other PSI-related activities will create a sense of cohesiveness between these groups as occurred for the pilot centers in Phase 1.

Dissemination of Information from the PSI

It is in the best interest of both the PSI and the biomedical community at large that information about PSI structures be made widely and conveniently available. There seems to be general agreement that deposition in the PDB is insufficient, because many in the biomedical community find this database foreign and even intimidating. A modest fraction of all structures are already written up as short papers, e.g. in Proteins and Protein Science, and are therefore searchable through Medline. Medline is used by the whole community, and it seems reasonable to try to find a mechanism whereby all PSI structures would be published in a medium indexed in Medline, so that all structures would become searchable.

One superficially attractive idea is that the PDB should itself become a kind of on line journal, so that, if requested by the authors, structures posted in the PDB would be referenced in Medline. The PDB has just recently started to assign DOI numbers to all submissions, and the on line journal designation seems a logical next step.

There are many reasons why this proposal may not be realistic. PDB may not have the resources to initiate or continue the on line process. Medline may not be interested. Other obstacles not obvious to outsiders may obtrude. But the proposal does serve as a starting point for discussion on how the goal of Medline searchability for all structures can be achieved.

Submitted by Brian Matthews (Chair) on behalf of the PSI Advisory Committee January 13, 2006