Bergmann. Like other sessions in the day, it included illustrative hands-on exercises for attendees to try on their laptops. The day was closed with a tutorial by Nicolas Rodriguez on JSBML. Following the tutorial sessions, the HARMONY meeting officially opened with a simultaneous buffet reception and evening poster session. In all, 15 posters were presented. They covered a wide variety of topics��everything from databases that can export data in the various formats discussed at HARMONY, to new software tools and emerging standards, including BioPAX and CellML. Day 2, plenary session on SBGN The first plenary session focused on SBGN. It was chaired by Falk Schreiber and focused on unresolved issues in the 3 SBGN sublanguages: Process Description (SBGN-PD), Entity Relationship (SBGN-ER) and Activity Flow (SBGN-AF). The session started with an overview and history of the SBGN standard from Nicolas Le Nov��re. This was followed by a report on the status of SBGN-PD by Stuart Moodie and discussion of items that were scheduled to be included in the Level 1 Version 2.0 release of SBGN-PD, but which still remained contentious. These problematic topics included the rules of subunit naming, the exact semantics of the ��Empty Set�� glyph, resolution of the community vote about reversible arcs, and the use of the SBGN Unit of Information to describe the cardinality of multimers and the material type of EPNs. A record of these SBGN-PD issues is available online at [91]. The next session of the morning concerned SBGN-ER and was led by Nicolas Le Nov��re. Discussions focused on several issues, particularly the semantics of an ��entity�� and whether different outcome glyphs were required to describe occurrent (e.g. an interaction itself) and continuant (e.g. a complex resulting from the interaction). Huayu Mi then presented the status of SBGN-AF, which was nearing a maintenance release (Level 1 Version 1.1). The major topic of discussion was whether a separate phenotype and perturbing agent activity were required, and whether SBGN-AF needed different glyphs. In addition, the outcome of the vote on how the type of activity (macromolecule, complex etc.) was indicated in AF [92]. Day 3, plenary session on BioPAX The plenary session of the third day was devoted to BioPAX. Gary Bader began with an overview of the planned activities, then introduced the projects and issues that were of interest within the community. The goal was nucleation of interested parties to begin discussions and work. This was followed by the session on specification and data. It was chaired by Emek Demir, who gave an overview of data integration and normalization and discussed progress since his last presentation on the topic during COMBINE 2010. Arman Aksoy then introduced the Patch algorithm and outlined the goal to test data integration at the meeting. He reported that several data providers tested integration with specific data sets.