XCA Memory error when aligning 700+ structures

[DrAliEbrahim]

When aligning 700+ A71EV2A (lb32627-66) datasets in XCA v3, the aligner falls over with the following memory error:

ERROR: Unexpected fatal error occurred when running aligner
Traceback (most recent call last):
  File "/dls/science/groups/i04-1/software/xchem-align-staging/src/xchemalign/aligner.py", line 866, in main
    a.run()
  File "/dls/science/groups/i04-1/software/xchem-align-staging/src/xchemalign/aligner.py", line 257, in run
    new_meta = self._perform_alignments(input_meta)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/src/xchemalign/aligner.py", line 543, in _perform_alignments
    updated_fs_model = update(
  File "/dls/science/groups/i04-1/software/xchem-align-staging/src/ligand_neighbourhood_alignment/update.py", line 459, in up
date
    _save_ligand_neighbourhood_transforms(fs_model, ligand_neighbourhood_transforms)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/src/ligand_neighbourhood_alignment/io.py", line 33, in _save_l
igand_neighbourhood_transforms
    yaml.safe_dump(dic, f)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/env_xchem_align/lib/python3.10/site-packages/yaml/_init_.py"
, line 269, in safe_dump
    return dump_all([data], stream, Dumper=SafeDumper, **kwds)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/env_xchem_align/lib/python3.10/site-packages/yaml/_init_.py"
, line 241, in dump_all
    dumper.represent(data)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/env_xchem_align/lib/python3.10/site-packages/yaml/representer.
py", line 28, in represent
    self.serialize(node)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/env_xchem_align/lib/python3.10/site-packages/yaml/serializer.p
y", line 54, in serialize
    self.serialize_node(node, None, None)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/env_xchem_align/lib/python3.10/site-packages/yaml/serializer.p
y", line 108, in serialize_node
    self.serialize_node(value, node, key)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/env_xchem_align/lib/python3.10/site-packages/yaml/serializer.p
y", line 108, in serialize_node
    self.serialize_node(value, node, key)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/env_xchem_align/lib/python3.10/site-packages/yaml/serializer.p
y", line 98, in serialize_node
    self.serialize_node(item, node, index)
  File "/dls/science/groups/i04-1/software/xchem-align-staging/env_xchem_align/lib/python3.10/site-packages/yaml/serializer.p
y", line 83, in serialize_node
    self.serialized_nodes[node] = True
MemoryError

Please send this information to the XCA developers:
Log file: aligner.log
Working dir location: /dls/labxchem/data/lb32627/lb32627-66/processing/analysis/xca_v3_20260120-AE/upload-v3

I am processing in /dls/labxchem/data/lb32627/lb32627-66/processing/analysis/xca_v3_20260120-AE/upload-v3 and have attached the log file.

xca_v3_20260120-AE_upload-v3_aligner.log

[DrAliEbrahim]

Solved by running on slurm using a bash script. Memory allocation needs to be looked into for large jobs (is this a memory inefficency issue? @ConorFWild). Results of this need to be added to the XCA documentation (i.e. due to XXX if running over XXX alignments use the cluster) along with example bash script.

xchemalign_staging_aligner.sh

[phraenquex]

@DrAliEbrahim update docs, link up with @ConorFWild for best practice instructions.

(@ConorFWild says no quick win.)(

[DrAliEbrahim]

I have added information explaining how to run the aligner on Diamonds Slurm cluster to the XCA readthedocs. The pull request can be found here:
xchem/xchem-align#64