IHM Meeting Summary Sep 20 2018

IHM Meeting Summary Sep 20, 2018

Attended by Ben Webb, Mikael Trellet, Brian Jimenez, Tom Goddard and Brinda Vallat

1. Mikael is continuing work on incorporating support for the IHM-dictionary within Haddock using the python-ihm library. Brian will continue to attend the IHM meetings so that he is ready to takeover if required.

2. The python-ihm library uses "residue range" as the default value for the _ihm_feature_list.feature_type data item. Mikael will create an issue on GitHub and Ben will address the same.

3. There may be other restraints in Haddock that may either be provided by users or calculated by Haddock during modeling. Eg: Restraints between centers of mass of two subunits that have already been fit to a 3DEM map. Mikael will send examples to Brinda.

4. Issue 66: One way to address the issue is to create a child category of ihm_dataset_group that allows datasets to be combined using boolean operators such as and, or & not. This would allow datasets to be combined in a generic way and these combinations can be used in different steps in the modeling protocol. However, this may not work for the NPC case because the experimental data which provides the geometric object restraints is not available and therefore is not included in ihm_dataset_list and ihm_dataset_group. The alternative is to group crosslink, geometric object and derived distance restraints separately and apply the boolean operators to these categories individually. Neither option is ideal. But it is clear that it is necessary to address the requirement of combining restraints using boolean opearators. Homework for everyone: Think about how to address this issue in a manner that is least disruptive to the existing data model.

5. Next meeting is on Thursday, September 27th at at 9AM PT / 12PM ET / 6PM CEST