Brigitte Laliberté welcomed all the participants on behalf of the IPGRI Regional Office for Europe and ECPGR and congratulated all the partners in EPGRIS for having produced such a successful EU project proposal. EPGRIS is an important model that should inspire other regional PGR networks to centralize or catalogue information on ex situ collections. The main impact expected out of EPGRIS is the development of national inventories and methodologies to develop such national coordination mechanisms. For the IPGRI Regional Office for Europe, EPRGIS is an important tool to bring in associated states and countries and build capacity in documentation. This project has been made known to IPGRI’s partners during several visits namely to Albania and Armenia where commitments have been made to support documentation activities and participation in EPGRIS. IPGRI has recently revised a Memorandum of Understanding between with VIR in Russia. Article 3 of the MoU on documentation and information does not explicitly mentioned EPGRIS but it has clear intentions to bring in Russia into EPGRIS and ensure that national inventory is documented and completed. The Search Catalogue EURISCO together with other databases and catalogues such as SINGER, GRIN and many others will facilitate the access to information on some of the major ex situ collection around the world. Facilitating access to information is the first step to facilitating the use of the material and developing appropriate conservation strategies. EPGRIS is also a very important output for ECPGR and support to the work of the crop working groups without established Central Crop Databases. Additionally, in the context of a recent project proposal for EU funding in situ conservation of European crop wild relatives, EPGRIS has been mentioned as a possible model/approach to building national inventories.

Theo van Hintum, coordinator of EPGRIS also welcomed the participants and reiterated the importance of this project. The documentation of PGR in Europe is in the stage that the access to the individual sources can be streamlined. He thanked the partners for their constructive collaboration in the preparation phase of the project. It quite unusual that a project is accepted by the EU without any amendments. In this case it was possible thanks to the clear communication between the partners, amongst others during the project preparation meeting held already in 1999, also in Rome hosted by IPGRI. Finally, he thanked IPGRI for the efficient preparation of the meeting.

The EU scientific officer for this project Dr. Baig, has been invited to participate in the meeting, but could unfortunately not attend. In correspondence with Dr. Baig, it was agreed that the project could effectively start December 2000, although the contract was signed in September and the project could therefor start in October. If necessary, it would be possible to extend the duration of the contract with an extra few months at the end of the project (obviously without additional funding).

The participants agreed Theo van Hintum to chair the meeting.

The chairman presented the agenda, which was approved by the group. The agenda had been structured following the order of the workpackages (WP) which were to start in the first 18 months. A suggestion to include a session of 15 minutes on the BIG web site of ZADI was proposed and scheduled in the agenda of the Monday (see revised agenda in Annex I).

All participants introduced themselves and their respective institutes, summarized their particular tasks in EPGRIS and expressed their interest in this project.

Theo van Hintum presented an overview of the EPGRIS project. Objective, expected achievements, project work plan, workpackages, project partners, main role of the participants, the details of the 10 different work packages and the relationships between the workpackages (for details refer to the project technical annex). The first achievement expected of EPGRIS are the national inventories and the second is the search catalogue, EURISCO with its estimated 70% of European PGR ex situ accessions by the end of the three-year project. Other important achievements expected are the improved collaboration between genebanks and the Clearing House Mechanism (CHM) of the CBD and improved capacity to manage PGR documentation and information.

The second coordination meeting will be organized by INIA in Portugal in the summer of 2002. The final project meeting is planned to be held in Prague in autumn 2003.

During the general discussion of the project, several issues came up that were to be considered during the detailed discussion of each WP, such as:

• How will EURISCO be defined: as a physical database or as a search catalogue? Its maintenance and updating will have implications for the setup of WP-B1 (Definition of data and standards).
• To what extent do national inventories already exist? In some countries such as Germany, France and Russia (VIR) they already exist.
• The building up of national inventories through EPGRIS is an important contribution and a clear commitment to the CBD. This should be clearly communicated.
• The National Programs will determine what information they want to link to / include in EPGRIS. The project should not dictate what information is to be included but should guide national programs in their decision. Some national inventory only includes information on national germplasm and not on the rest of the ex situ material - is this a problem ?
• The future of the European Central Crop Databases (ECCDB) has been questioned and should be discussed within the context of EPGRIS.
• The issue of standards/descriptors and the adoption or adaptation of the IPGRI/FAO Multi-Crop Passport Descriptors (MCPD) was raised as an important issues to resolve in the course of this meeting.
• Procedures for updating of the information were also raised as an important issue.

Brigitte Laliberté introduced WP-A1. In each European country the ECPGR national coordinator will be approached and asked to propose a National PGR Inventory Focal Person in liaison with the national Clearing House Mechanism focal point. This Focal Person will define the countries needs (help, training, etc.) and be responsible for establishing the National PGR Inventory, and providing the data to the European Search Catalogue. The focus of the project will be on the EU and associated states, but also other European countries will be invited to join the initiative (on a self-funding basis).

The discussion mainly concerned the mandate of the Focal Person (FP), how technical that person should be and the support that EPGRIS will provide to those FP and the national programs in general.

It was agreed that the invitation letter should clearly indicate what is expected of the FP. It is then up to the national coordinator to nominate who they think would be most appropriate. It should be made clear that a commitment should be made towards this project and in particular, data on a national inventory will be expected, even if a national inventory is at a very early stage of development. It was suggested to take some of the wording used in the original letter that was sent to the National Coordinators two years ago for the ECPGR survey on EPGRIS regarding the contribution to the Clearing House Mechanism.

A letter will be sent to all ECPGR National Coordinators, including an extract of the technical annex (brief summary of the project) with the objectives, excluding any financial information. The mechanism of sub-regional meetings, inventories of training needs and technical assistance will also be explained.

It was suggested to consult the list of FAO – WIEWS focal points (available on-line) as a possibility to suggest that it be the same person to enhance complementarily with WIEWS.

It was also agreed that ECPGR countries would be considered but also the other European countries. The letter to the later could be simply addressed to the Ministry of Agriculture. There should be 2 types of letters. The first for the countries of the EU and Associated States eligible for funding and the second letter to the non-EU inviting them to nominate a FP but with the knowledge that the participation would be on a self-funding basis.

It was agreed that a draft letter (2 types, EU, non-EU) would be prepared by the ECPGR Coordinator (Lorenzo Maggioni) and circulated to the project partners for feedback. EPGRIS partners would be given a week to comment. A revised letter would then be sent to all ECPGR National coordinators and to EU countries not in ECPGR. (Action: L. Maggioni, by end of January 2001)

Iva Faberova introduced WP-A2. Her presentation covered the different sub-regions agreed on during the meeting in Rome in August 1999, and the estimated number of participants for each meeting considering the funded partners and the possibility of self-funded partners to attend. A very effective colored map of the different sub-regions was produced to visualize the division. Furthermore an overview of the European ex situ PGR collections with an estimated 750-800,000 accessions was presented. The ECCDB containing information on an estimated 424,812 accessions (July 2000). It is believed that there remains a large proportion of accession that is not included in any ECCDB.

The prerequisites of WP-A2 are that focal persons be nominated and this person should have clear duties and responsibilities for data supply (deliverable of WP-A1). Furthermore, data standards should be proposed (deliverable of WP-B1).

The length of the first sub-regional meetings was discussed and it was agreed that 2 days for the biggest groups should be sufficient, taking into consideration the country report of each participant. For small meetings, one day could be sufficient.

Concerns over the financial amount allocated to the organization of such meetings was expressed and it was felt that the funds could be insufficient in the light of continuously increasing airfares and other costs. It was proposed that, if there would be serious problems, money could possibly be moved from other activities. After the first round of regional meetings the necessity will be considered, but for the time being the coordinator strongly advised to try to stick to the budget. (Budget: for the hosting partner 1 person-month and €2000 for local costs per meeting, and €1000 per expected paid participant, for 'attending partners' (ZADI or CGN) .25 person month and €1000.)

The original timing of the meeting was discussed and it was proposed to postpone the date of the second meeting to allow some feedback from the second coordination meeting. The proposed dates for the first round of sub-regional meetings would be July 2001 (month 8) and the second round would be held in October 2002 (month 23).

A letter of invitation to attend these sub-regional meetings will be sent by the hosting partners. Two types of letters will be sent; one to those funded and the other to those on a self-funding basis. (Action: Organizers sub-regional meeting, at least 3 months before the meetings). It will be then up to IPGRI regional office for Europe and ECPGR to contact non funded FPs individually to provide some additional support. This will be done at their discretion but in consultation with the EPGRIS partners. A draft a letter of invitation (2 types) would be prepared and circulated to the group for comments. (Action: I. Faberova, at the latest May 2001)

It was proposed that a common agenda for all the sub-regional meetings be developed. The agenda of the first meeting could include following points:

• Overview of the project. This could include material from Theo's presentation. (Action: T. van Hintum, draft circulated at the latest May 2001)

• A light demo of EURISCO could be presented to demonstrate how it could look like, keeping in mind that this structure is open for changes. Examples of data from ZADI, BRG, RICP or CGN. (Action: S. Gaiji, prototype circulated at the end of April 2001)

• Country reports, to be delivered by each FP. These report should include information about the current situation concerning the documentation of PGRFA, the envisaged way to establish a National Inventory, and the bottle necks in this process.

• Discussion on how to achieve national inventories. This issue will be discussed via e-mail amongst the partners prior to the first meetings. Some elements could be:
• Current status of a national inventory.
• Nomination of an institute that could house and maintain such a national inventory.
• Merging of the ECCDBs and production of a national "profile" from the ECCBD as an initial set of data.
• Validation of the 'national' data from the ECCDBs.
• Identification of other collections in the country.

• The type of information that partners are planning to provide
• Methods for centralizing, maintaining and updating this information.

• Technical discussion on the uploading mechanism (including the descriptors).

Frank Begemann introduced WP-A3. There are resources available to take away bottlenecks in the development of national inventories: direct financial support for investments and staff exchanges with ZADI, NGB or CGN.

There need to be clear criteria for the selection of candidates for staff exchange or individual support. These could include:

• Level of expertise (performance, who is doing the best)
• Needs (who needs it most)
• Equality (everybody should be treated the same)
• Quantity (number of accessions) and quality of ex situ genetic resources and their importance to the European inventory
• Tangible output (data coming out of the exchange)
• Geographical balance

These criteria should be established before the next meeting. (Action: F. Begemann & T. van Hintum, draft circulated at the latest June 2001) Own knowledge of situation through contacts, country visits and meeting would be also used to select potential candidates, in consultation with the ECPGR National Coordinators. But these criteria could then be used to justify the selection. The decision would then be taken by the three hosting institutes, ZADI, CGN and NGB, to arrange for bilateral invitation procedures.

The question was raised whether non-EU countries could benefit from funding even if indirectly through staff exchange. Possibilities will be checked and results will be circulated. (Action: T. van Hintum, at the latest April 2001)

The chairman introduced the new participants (Steve Katz from FAO and Lorenzo Maggioni the ECPGR Coordinator) and gave a brief introduction to the project.

Steve Katz gave an overview of the activities of the FAO, especially in regards to information and knowledge management (the World Agricultural Information Centre - WAICENT), and put this in the light of the World Food Summit (November 1996). He elaborated on FAO's principles of information management: strategic use of metadata and content repositories, separation of content, presentation and processing, controlled vocabularies, categorization schemes, and the use of international and non-proprietary standards such as XML. The technical structure behind the FAO web-site was presented. This site had in November 2000 and impressive number of 11.5 million hits with over 500.000 user sessions! Also the problem of finding material on the WWW was discussed, and the possible solution provided by the Dublin Core (an agreed set of core metadata) was presented.

Frank Begemann briefly presented BIG, the integrated system developed at ZADI, to give access to a wide spectrum of genetic resources maintained in Germany. The system uses a distributed database approach in which a user defined query is transferred to several locally maintained databases, and all results are translated in one form using the XML protocol.

The use of this kind of approach in EPGRIS is considered an option that can be developed; it would provide the partners to get acquainted with the technology, and since it is generally considered to be the 'next technology', it would prepare EURISCO for the future. However, since the objective of EPGRIS is not to develop new technology, but to develop a new infrastructure, EPGRIS will be conservative in applying it in the 'production database'.

Morten Hulden introduced WP-B1, showing what standards need to be set, and what issues need to be considered when setting these. He gave a comprehensive overview of options available.

A brief discussion on the type of data to include was concluded by leaving this decision to the national partners and be inclusive as opposed to being exclusive at this stage, within the range of PGRFA. If PGRFA information is available and there is a unique identifier (such as accession number), then it can be included.

The data format generated more discussion, as this is the crucial element of the project. The main points of discussion were on the following:

• Data format: Language and character set

• Export file format

• Transport

• Field format (descriptor list)

• A new version of the MCPD will be drafted and circulated to the EPGRIS partners. (Action: Morten Hulden, by the end of January 2001)
• The EPGRIS partners will provide their feedback. (Action: all partners, by mid-February 2001)
• A revised version will be produced. (Action: Morten Hulden, by the end of February 2001)
• If the draft is acceptable to IPGRI, this will be circulated by IPGRI to a wider audience, including FAO. IPGRI is to consolidate comments and produce a final version by June 2001, to be ready for the first sub-regional meetings of July 2001.

Two main models of data conversion were presented: (1) conversion at the central catalogue – "centralized conversion" where one place does the conversion of more than 40 files and (2) conversion at the provider’s side – "decentralized conversion" where 40 people do their own conversion. The advantages and disadvantages of each option was discussed and are summarized below:

It was then agreed that a decentralized conversion would be preferable, mainly because of the resource requirement and therefore low sustainability of the centralized option.

Samy Gaiji introduced WP-B2, presenting the requirements, and options for creating the database system. SINGER will provide the technical support to the IPGRI regional office for Europe to implement the EURISCO search catalogue.

The requirements for the development of EURISCO are:

• Agreement on data standards: passport data, authoritative standards for taxonomy, others such as genebank/institute ID, country codes etc.
• Agreement on data replication formats
• Agreement on search features/options
• Agreement on "identity" and presentation
• Agreement on timeframe

The inclusion of ancillary data such as GIS, map standards, climatic and soil data sets and references (URLs) should be considered as a second step to EURISCO. The ultimate objective of EURISCO is to access information on a particular accession and continue the search on the national database.

One of the obstacles for effective use of the catalogue is the taxonomy; taxonomic references should be used. SINGER uses a range of them such as Species2000, GRIN. BIG used BIGTax as a solution. It was agreed that the taxonomic authority should not be required; it should be strongly encouraged but at this stage should not be mandatory. This should be taken into consideration for data quality improvement and could be done as a second step to the project, part of the public awareness.

Over the last year, SINGER has been involved in developing the SINGER Tool Kit (TK), a software to web enable databases, particularly designed for genebank management databases but can be used for a wide range of databases. The TK replicator has therefore been developed based on SQL/ODBC allowing replication between data sources (MS SQL7). Data integrity is automatically checked. If data is rejected, then a message is generated with the reasons for rejection. If corrected, the data is re-submitted. The front-end design would be left for the partners but the TK design forms and queries.

A prototype of EURISCO could be tested on the EPGRIS Intranet web site (private site accessible only by the project partners. Address: http://141.ipgri.cgiar.org/EPGRIS Username: EPGRIS. Password: EPGRISpartner.)

It was suggested that data be accepted in both formats, pipe-delimited and XML to test the procedures. SINGER would be ready to start in February 2001 with national inventories already available such as from ZADI, RICP, BRG, INIA and CGN. ZADI would provide the format in XML to test the physical transfer. It can provide data for Germany but also for Russia (VIR). This trial would also feedback information for the needs for 'standards'.

The establishment of a project Internet site to be used as a public awareness tool was discussed. It was proposed that the information about the project would be the responsibility of the project coordinator. The front-end should include:

• introduction to EURISCO
• information about the partners
• links to national inventories be included
• legal issues

The issue of replication and the archiving of old files was discussed. It was agreed that the complete set of data from a national inventory should be replaced with the new set and the old files be archived. This would avoid the problem of updating record by record and the problem of deleting redundant records would be solved this way. Downloading protocols should be established for the "distributed" databases. To facilitate the searches, it would be preferable to search data stored locally. Frequent updates is not considered to be problematic (could be done as an automatic procedure daily).

The agreed timeframe for the development of EURISCO is:

Theo van Hintum gave some additional information about the management of the project:

• The project can be extended for a few months without extra funding. The coordinator will start the procedures for requesting this extension when appropriate (after month 20).
• The first lump of money (40% of the total budget) has been received by the coordinators institute. The transfer of this money was ordered and the coordinator will verify that the money has been transferred.
• It was recommended by the coordinator to have the contracts checked by each partners legal department. Any issues of ownership of software, data should be raised to the coordinator as soon as possible. Knowledge generated is owned by the partner generating this knowledge: software developed would remain in the property of IPGRI in the case of EURISCO. Should there be some intellectual property issues, a consortium could be formed to agree on ownership.
• Financial statements and reporting obligations: consult annex II of the contract.
• The coordinator will file the annual reports. The partners will be asked to provide details and information for these reports. If, for the reporting, the division in workpackages as made in the technical annex is followed, this should not cause any difficulty.
• Every year also a financial statement has to be produced. For institutes without experience this might be difficult. The format for this Cost Statement is included in annex II. The first year is from the 01/10/00 to the 01/10/01. The coordinator will prompt the partners for such statements.
• The second coordination meeting has been proposed for the 14-15 June 2002, in Portugal, hosted by INIA.

The project should create a web Intranet site with the following information (Action: B. Laliberté, by end of January 2001)

• objectives
• partners
• project info
• activities
• link to demo site

The issue of the future of the ECCDB was raised. The ECCDB managers are preempting that EPGRIS might take over the roles of the DB managers. It was felt that the ECCDB can bring value to the data and the crop collections. These database can be linked to particular projects and or expertise. Furthermore, they can be used to identify duplicates. A discussion paper could be produced to explain the objectives of EPGRIS and the complementarily with the ECCDBs. The new web site could be used to publish this information. A letter could be sent to all DB managers to also clarify this issue, including a diagram of where all the components fit in together. DB managers should be encouraged to concentrate efforts into characterization and evaluation. T. van Hintum agreed to produce such a letter, since he is both a DB manager and coordinator of EPGRIS. (Action: T. v. Hintum, by April 2001)

The chairman thanked the participants for their contributions to this productive meeting. Important decisions have been made that were critical for the development of the project. The collaboration of the host and organizers of the meeting IPGRI, and especially Brigitte, was gratefully acknowledged.

Note: the minutes of the Meeting of the ECPGR Internet Advisory Group, Sunday 17/12/00 will be produced as a separate document.