Meeting Minutes 5 March 2007
Meeting Minutes 5 March 2007
Present: David Cumin, Jonna Terkildsen, Poul Nielsen, Tommy Yu, Sarala Dissanayake, Peter Hunter, Andrew Miller
Poul noted that we still need to check that everything is ready for the CellML workshop. Sarala has spoken to Gareth about the CellML website, but Andre says that the current version is not finalised [errata from Andre: it is just the programme that is not yet up and can't be until we know who can make it. The initial web page for this workshop is ready for the main site and waiting on an appropriate place to be allocated for it.]. Peter thinks, given the short time available, we can only hope to find funding for local accommodation, and require attendees to find their own travel funding. Poul thinks that the 5th floor seminar room is the best place for the meeting, but since the 4-5th April is a Wednesday / Thursday, there will be many classes using the room during this time. The lectures could be moved to the ENGSCI lecture theatres if possible, or to the BME lab.
Andrew noted that he has deferred work on the graph selection because he is waiting to see if Peter will agree with the rest of the group regarding the choice of a combo box to select graphs, as opposed to tabs. Peter now agrees with the group. However, there is still a separate issue of needing to switch between total layouts of the graph. Andre has suggested in an e-mail that we need a separate file that describes all the graphs in a model, and we could then have more than one of these open. This would also allow us to save the total state about which models are open.
There was some discussion about how PCEnv can better validate changes to a model. There is currently no validator available, only the integrator, which will detect some errors. Peter has heard that Steve McKeever's students were going to look at this, but Andrew hasn't heard anything from them. He will mention this next time he is in contact with them.
Tommy has verified that launching models from the repository works. He has been working on re-writing the metadata library, as the previous library is unusable.
Tingting is also working on a new model repository as part of her PhD, with the aim of fixing granularity issues. Poul, Matt, and Gill Dobbie are supervising her. There was concern that it would be better for this to be part of the current repository. Poul agreed that she should be encouraged to attend the CellML meetings.
Andrew briefly updated the group on his PhD project. He noted that there was no obvious way to make use of species rates unless we could be sure that nothing in the black box would also use them. He has been looking at the possibility of using OrganelleDB or similar data to constrain which species can interact, and so build up a topological model from this. Jonna was concerned that the black box could mask problems with the white-box. Andrew agreed, but thinks that if highly trusted white boxes are used, such as conservation laws, and reliable localisation data, this is not an issue.
Poul noted that we still need to check that everything is ready for the CellML workshop. Sarala has spoken to Gareth about the CellML website, but Andre says that the current version is not finalised [errata from Andre: it is just the programme that is not yet up and can't be until we know who can make it. The initial web page for this workshop is ready for the main site and waiting on an appropriate place to be allocated for it.]. Peter thinks, given the short time available, we can only hope to find funding for local accommodation, and require attendees to find their own travel funding. Poul thinks that the 5th floor seminar room is the best place for the meeting, but since the 4-5th April is a Wednesday / Thursday, there will be many classes using the room during this time. The lectures could be moved to the ENGSCI lecture theatres if possible, or to the BME lab.
Andrew noted that he has deferred work on the graph selection because he is waiting to see if Peter will agree with the rest of the group regarding the choice of a combo box to select graphs, as opposed to tabs. Peter now agrees with the group. However, there is still a separate issue of needing to switch between total layouts of the graph. Andre has suggested in an e-mail that we need a separate file that describes all the graphs in a model, and we could then have more than one of these open. This would also allow us to save the total state about which models are open.
There was some discussion about how PCEnv can better validate changes to a model. There is currently no validator available, only the integrator, which will detect some errors. Peter has heard that Steve McKeever's students were going to look at this, but Andrew hasn't heard anything from them. He will mention this next time he is in contact with them.
Tommy has verified that launching models from the repository works. He has been working on re-writing the metadata library, as the previous library is unusable.
Tingting is also working on a new model repository as part of her PhD, with the aim of fixing granularity issues. Poul, Matt, and Gill Dobbie are supervising her. There was concern that it would be better for this to be part of the current repository. Poul agreed that she should be encouraged to attend the CellML meetings.
Andrew briefly updated the group on his PhD project. He noted that there was no obvious way to make use of species rates unless we could be sure that nothing in the black box would also use them. He has been looking at the possibility of using OrganelleDB or similar data to constrain which species can interact, and so build up a topological model from this. Jonna was concerned that the black box could mask problems with the white-box. Andrew agreed, but thinks that if highly trusted white boxes are used, such as conservation laws, and reliable localisation data, this is not an issue.