Use cases

From West-Life
Jump to: navigation, search

See also the service interaction table

ARP-wARP.png

1 2 3 4 5 6 7
Title End-to-end xtal structure solution with ARP/wARP End-to-end xtal structure solution with PDB-REDO Send to PDB-REDO Use PDB_REDO databank models Round trip between Scipion and Haddock Submit to Haddock FANTEN to Haddock (title of next use case here)
Description The entry point is CCP4online services for structure solution by Molecular Replacement (MR). The entry point here is a target sequence and scaled and merged diffraction data. Mr. Bump can select the so called “search model” (most similar to sequence) and the output of this step is an atomic model which, hopefully, phases the data satisfactorily. This model has in most cases the wrong sequence and is often incomplete. It can also be that it does not phase the data satisfactorily. We envisage two exit points.

• The first one goes to ARP/wARP that can build a new model based on the current solution: that model can have the right sequence and will likely be incomplete. For search models with low similarity to the target sequence, this is the preferred option.

The entry point is CCP4online services for structure solution by Molecular Replacement (MR). The entry point here is a target sequence and scaled and merged diffraction data. Mr. Bump can select the so called “search model” (most similar to sequence) and the output of this step is an atomic model which, hopefully, phases the data satisfactorily. This model has in most cases the wrong sequence and is often incomplete. It can also be that it does not phase the data satisfactorily. We envisage two exit points.

• The second one, goes to PDB_REDO which refines and partially rebuilds the current atomic model. It also returns a validation report for that model. By linking these two services, users will get an end-to-end structure solution. For search models with high similarity to the target sequence, this is the preferred option.

Several services provide output that is suitable for submission to PDB-REDO Many services use PDB entries. It should also be easy to use PDB-REDO entries instead. Examples are:

• Haddock refinement of FANTEN rigid body docking

Many services provide an output suitable for submission to Haddock Haddock refinement of FANTEN rigid body docking
Step1
Inputs Reflection data as integrated and merged reflections.

Target sequence.

Reflection data as integrated and merged reflections.

Target sequence.

PDB file PDB ID PDB file; paramagnetic NMR data (file types in)
Outputs Coordinate data as PDB file (MR model). Coordinate data as PDB file (MR model). Revised PDB file PDB_REDO file with coordinates PDB file; optimized tensor parameters (file types out)
Program Molecular replacement servers (Morda, AMPLE, Balbes, Auto-Rickshaw, MrBump),

or experimental phasing server (Auto-rickshaw)

MrBump PDB_REDO Databank FANTEN
Portal https://www.ccp4.ac.uk/ccp4online/ https://www.ccp4.ac.uk/ccp4online/ PDB_REDO Databank (soon @NKI)
Step2
Inputs Target sequence .AND. Diffraction data .AND. MR model Diffraction data .AND. (MR model .OR. ARP/wARP model) PDB files; paramagnetic restraints; optimized tensor parameters (file types in)
Outputs ARP/wARP model PDB_REDO model, Validation report Optimized PDB of complex (file types out)
Program ARP/wARP molrep PDB_REDO Haddock
Portal EMBL-HH services https://xtal.nki.nl/PDB_REDO/ https://xtal.nki.nl/PDB_REDO/ Haddock Haddock and Scipion Haddock and DISVIS/POWERFIT FANTEN and Haddock
Effort to Develop Need to be able to login into CCP4 and PDB_REDO services from the VRE (this currently doesn’t work from CCP4online embedded in the portal). Need to be able to upload files to the CCP4 services e.g. from a West-Life virtual folder. Need to retrieve the output and pass automatically to PDB_REDO. Need to retrieve the output of PDB_REDO. Need AJAX query to determine whether an entry exists Package the output of FANTEN, including paramagnetic restraints and optimized tensors parameters and push it directly to the Haddock portal (time estimate)
Comments These points can also be used in parallel if the sequence similarity criteria suggests it, and PDB_REDO also takes ARP/wARP output.

The exit point for both options is a new PDB file. This is presented to the user for making a choice.

We could connect CCP4 directly to PDB_REDO, as we do with ARP/wARP. I deliberately specify going via the virtual folder to demonstrate the use of a collated storage area. In future, we could look at staging one or both steps on Grid/Cloud resources. The AMBER or GROMACS portals for MD could be used here (comments on next use case here)