DIALS core meeting 2020-10-15

Previous Actions

• MG+ND: write a proposal for ‘overall architecture discussion’ and ‘new installer’
  • MG+ND: Implement DC1 – per-module setup.py in a pull request
    • setup.py requires sources to be in a subdirectory, so to do this we need to change the source layout at the same time. We will therefore test this with xia2.
  • MG+ND: Come up with a proposal to move away from all code being in header files and consolidate into a single library
• ND: update CONTRIBUTING in a (separate) pull request to include a soft-imposition of stable master, dials#1353.
• MG: organise a typing intro lecture.
• GW: move cbflib_adaptbx’s compress and uncompress functions into dxtbx in a PR.
  • broke building on Mac completely, and looked like it broke building on Linux but may have been a coincidence. ND looking at it.
  • ASB to ask BKP whether fundamentally scons could run cmake under the hood for a module

Agenda

New folks in the room

Randy, Andrey, (maybe) Peter coming along? Intros; discussion of how things work at the moment; possibly schedule specific meetings for longer form stuff. Questions of degrees-of-freedom for individual reflections & lattice translocation raised recently.

Randy Read: translocation vs NCS, interested in running CryoEM-2D-classification algorithms on found spots to distinguish between fuzzy and non-fuzzy spots.

Peter Zwart: Interested in better likelihood targets, would like to know the DoF for each reflection. (As weighted sum of observations it is not N(obs)-1)

AP: JBE to set up meeting with Andrey/Randy/Peter with DIALS-E folks at a suitable date & time.

Peter Zwart: Have a look at xicam regarding the image. https://xi-cam.readthedocs.io/en/latest/

AP: ASB to organise a meeting on the image viewer.

cctbx conda-forge package

We may want to evaluate moving the dxtbx/DIALS Azure builds to use the now available cctbx conda-forge package, once issue #4 is resolved, which prevents the package being used as a build platform.

• ABS: That would be rad.
• We haven’t appreciated the fundamental perspective shift yet though.
• ASB: Will bring this up the next CCI-all meeting.
• GW: We should bring our heads round the idea that all different components become different conda packages.

Benefits

• Faster builds
• Making use of clearer API boundaries and versioning

Drawbacks

• API boundaries and versioning
  • In other words: We would start treating cctbx as a software package. This restricts dxtbx/DIALS code to only use functionality that is part of the relevant cctbx release. From time to time, when cctbx releases new versions, we would update our dependencies, which would then allow the use of new cctbx features.

Outcome

• ASB: Was discussed at the CCI-all meeting, and the outcome was: this is fine. I need to be able to continue using cctbx as a development installation. Apart from that it is fine to treat cctbx as a versioned software package, ie. run Azure against the cctbx-base conda-forge package.
• The drawbacks noted above will thus apply.
• AP: MG: Make this happen
• ASB: Please don’t wait for cbflib for a dxtbx conda-forge package.

dxtbx → xfel dependencies

At some point we made a somewhat conscious decision that dxtbx should not depend on dials. While setting up some Azure testing of dxtbx (#226) without dials/xia2 and their dependencies present some test failures were flagged because dxtbx has dependencies into xfel.

How should we deal with this? A few options:

• Disable the code so it stays in dxtbx but is not tested/missing dependencies do not trigger errors?
• Cleanest option would be to move xfel-dependent format classes into the xfel package
• Eliminate the dependencies

Discussion:

• GW: I have the opinion that we shouldn’t have circular dependencies. Ever.
• DGW: has electron-diffraction format classes outside of dxtbx. Would love to put SMV classes into a separate box as they are too promiscuous.
• ASB: having format classes stashed together is an easy mental model

Previous outcomes:

• ASB needs to consider the future of the xfel module
• In principle the format classes could go into xfel
• The commit linked above has gone in as part of dxtbx#226
• Will discuss next time

There are a couple of places there dxtbx → xfel

• format/FormatXTC.py
• format/FormatPYmultitile.py
• format/cbf_writer.py
• format/FormatXTCJungfrau.py
• format/FormatXTCRayonix.py
• format/FormatXTCCspad.py
• command_line/radial_average.py
• command_line/detector_superpose.py

AP: ASB to create a ticket in dxtbx to discuss this so that we can get everyone’s opinion.

NIAC Nexus

Nexus as working format for DIALS?

DIALS has an existing data model (past: pickle, now: msgpack). Internal names in that model correspond to the names used inside DIALS for development, eg. xyzobs.px.value has a clear definition.

• internal names can be kept using soft/hardlinks

No objections for a dials.export $file.refl format=nexus, but do not use nexus as a format to interchange data within DIALS programs.

What is the underlying benefit of using Nexus?

• main benefit is putative interoperability

AP: ASB to create a transformation tool DIALS → Nexus → DIALS to demonstrate the feasibility of this approach, so that we have a concrete proposal to discuss. AP: MG to put this into a ticket “Demonstrate round trip of DIALS data to Nexus data with no loss of signal” instead of meeting minutes

Pull request When file indices are provided,…

not discussed in this meeting

ASB: This is dxtbx#210. The FormatSaclaHDF5 class isn’t lazy, which led to that problem. Will fix this in a separate PR, but the PR above is still a good PR independently of that.

In discussion we identified a new potential issue. We need to ensure during deserialisation this loop isn’t hit for classes that don’t have scans and goniometers.

Definition of trusted_range

not discussed in this meeting

Discuss Definition of trusted_range (dxtbx#182)

image_range vs. array_range baked in off-by-one errors

not discussed in this meeting

Discuss image_range vs. array_range baked in off-by-one errors (dxtbx#186)

• DW made a proposal and started work on that, but it proved messy
• There is a new, simpler, proposal but this version kills off the image_range concept rather than array_range
• Discuss, again, what we want here and whether the new proposal is useful

DIALS documentation build

not discussed in this meeting

Should the DIALS documentation build outside a cctbx environment?

• Currently you can only build the DIALS documentation inside a cctbx environment. This means a remote documentation build needs to build cctbx first, so we can’t easily run the documentation build on every commit. When things break they are difficult to fix. We can’t use readthedocs.
• MG: I believe the main obstacles are three cctbx Sphinx plugins that may need to be extracted or removed from DIALS documentation, and some phil parsing logic.
  • MG: the cctbx conda-forge package may be useful here

dxtbx json/msgpack performance

not discussed in this meeting

ASB reports json being problematic when importing e.g. 50,000 experiments, as everything is stored in a single file, so everything needs to be read at once.

• GW: cf. Load before heat death (dxtbx#118). At the core of it are bad design choices in dxtbx.
• Future discussion topics:
  1. treatment of large serial datasets
  2. using an HDF5 backend for reflection files
  3. dials#1407 specific proposal
• ASB: I want check_format to go away.

Previous discussion outcomes:

• ASB:
  • General consensus that this is needed.
  • Make models and store them
  • Load them in parallel with minimal memory requirements
  • Version all models and rewind them
    • This needs a lot of thought and at this time we don’t have agreement here.
  • Not need to copy data like shoeboxes when they aren’t being changed, operate by reference

Remove tntbx dependency

not discussed in this meeting

Appears not to be used anywhere outside of cctbx_project/xfel, but xfel does not declared is as a dependency. xfel only uses the svd() function. Maybe change xfel to use numpy.linalg.svd instead and drop tntbx entirely?

Peter Zwart: My favorite SVD tools lives in scipy: scipy.sparse.linalg.svds (https://docs.scipy.org/doc/scipy/reference/generated/scipy.sparse.linalg.svds.html#scipy.sparse.linalg.svds). This routine allows calculation of the top K vectors in an SVD, it is a massive time saver if you know the rank of the matrix. The code works for dense as well as sparse matrices.

renaming master branch → main

not discussed in this meeting

• Nothing much will happen on this any time soon - either way we’ll wait for GitHub support on this one
• General assent that this is useful
• from 1st of October new GitHub repositories will have main as the default
• proposed outcome: Agree to migrate to main. Turn this into dxtbx/dials/xia2 issues and move once GitHub provides a migration path. Use xia2 as a test-bed and move dials/dxtbx two weeks later.

Next meeting

November 5th, 4pm UK time, 8am PDT.