Support Processing plan
The processing plan needs to support things like:
1)Which pipelines to use -- any user combination of grenades_fastproc, grenades_parallelproc, EDNAproc, autoProc, XDSapp etc
2)For all pipelines, a common way to specify:
A reference HKL file
Statistics program (MRFANA for now)
High Resolution cutoff
Isotropic or Anisotropic
A user selectable combination of metrics with some logical grouping if possible. e.g. (Completeness >90) AND (I/SigI >1 OR CC0.5>30%). I think for the short -medium term this will only be a simple AND relationship though
metric (I/sigI, CC0.5,Rmeas,% etc), value
Binning:
Number of bins
Number of reflections per bin
Binning mode:
Equal-volume
Equal-observation-number
1/d**2
3)For a selection of pipelines (only fastproc and parallelproc for the moment):
Should we run Molecular replacement from user model
if so - a place for complexes/models like before (this is one of the overlaps with sample information)
Molecular replacement from unit cell
Automatic SAD
If so, anomalous element, how many, proteins sequence (could be from sample information instead of processing plan)
I think the missing bits are:
Multiple selection. Example: pipelines: "grenades_fastproc, grenades_parallelproc, EDNAproc"- PDB/other files: this has definitively be linked to the sample itself
- Reference to file: not sure if it has to be done at the level of the "sample" for each row or at the level of the sample. I presume this can be done as 2)
Edited by Alejandro De Maria Antolinos