bioml-tools/bio-datasets

Issues

• Write round-trip tests for structure feature encoding

  #79 opened 2 months ago
  0

• Identifier-based split builder

  #76 opened 2 months ago
  0

• Manually inspect release contents

  #75 opened 2 months ago
  0

• Check hatch -inferred version

  #74 opened 2 months ago
  0

• Improve build drawing on biotite

  #73 opened 2 months ago
  1

• Decide on how to store residue dictionary in DatasetCard

  #72 opened 2 months ago
  0

• Print ProteinStructureFeature etc should not print all properties of residue dictionary

  #71 opened 2 months ago
  0

• Separate cache dir from datasets

  #69 opened 2 months ago
  1

• bio_features and features in dataset info seem to be in conflict sometimes - maybe I should change default cache dir to avoid this type of issue?

  #68 opened 2 months ago
  1

• Add pdb examples to readme/fix proteinchain thing

  #67 opened 2 months ago
  0

• Use mmcif metadata to infer expected values for standardisation tests

  #64 opened 2 months ago
  0

• write a load_example method

  #63 opened 2 months ago
  2

• Check that nans don't cause issues for writing

  #62 opened 2 months ago
  0

• Why does load_dataset with features passed lead to re-generation of splits? Whereas manually setting features doesn't

  #61 opened 2 months ago
  0

• Add residue_annotations, atom_annotations fields

  #60 opened 2 months ago
  0

• Add helper option to load_dataset to remove or change feature types if not currently available.

  #59 opened 2 months ago
  0

• Generalise fill missing residues to a structure alignment mapping tool (apply to pinder / afdb)

  #58 opened 2 months ago
  0

• Load bond information with include_bonds

  #57 opened 2 months ago
  0

• PDB upload error

  #56 opened 2 months ago
  0

• Use run length encoding for chain id

  #55 opened 2 months ago
  0

• CIF parsing issues

  #54 opened 2 months ago
  0

• Biotite compression issues

  #53 opened 2 months ago
  0

• Why is encode_example slow with ProteinChain (upload_foldcomp_db)

  #50 opened 2 months ago
  0

• Base atom filtering on residue atoms

  #48 opened 2 months ago
  0

• Add tests for structure feature

  #47 opened 2 months ago
  0

• Upload foldcomp asarray suddenly very slow - check e.g. reorder chains.

  #46 opened 2 months ago
  0

• Graph feature type

  #44 opened 2 months ago
  0

• Check ccd cache

  #42 opened 2 months ago
  0

• Check what id does in Image/Audio features

  #38 opened 2 months ago
  0

• Add mypy, other linters

  #37 opened 2 months ago
  0

• Convert cif to binary cif

  #36 opened 2 months ago
  0

• Infer symmetrised assemblies from raw coords

  #35 opened 2 months ago
  1

• Fix element inference in atom array decoding

  #33 opened 2 months ago
  0

• Check OXT support

  #32 opened 2 months ago
  0

• Full foldcomp compression support - at feature level? or at builder level?

  #31 opened 2 months ago
  0

• Check feature casting works sensibly

  #30 opened 2 months ago
  0

• Write generic standardisation utils for: filling missing atoms, filling missing residues

  #29 opened 2 months ago
  0

• Can we get best of both worlds with h5ad and arrow

  #27 opened 2 months ago
  2

• Think carefully about residue dictionary design - effectively the tokeniser

  #26 opened 2 months ago
  0

• Handle empty chains (e.g. with warning)

  #25 opened 2 months ago
  0

• Store library information separately in metadata

  #24 opened 2 months ago
  1

• Include ccd in release similar to biotite

  #23 opened 2 months ago
  0

• Get component is really slow

  #22 opened 2 months ago
  0

• Does ccd log frequency of usage?

  #21 opened 2 months ago
  1

• Graceful handling of unexpected residues / atoms?

  #20 opened 2 months ago
  0

• Use a dataset builder script to support arbitrary splits of e.g. pdb

  #19 opened 2 months ago
  0

• does load dataset with specified split always download the full dataset anyway?

  #17 opened 2 months ago
  0

• write repr / str methods for ProteinChain etc.

  #13 opened 2 months ago
  0

• Do we need a force download or a way of cleaning cache or a way of checking cached features / metadata match

  #12 opened 2 months ago
  0

• Fix portable caching

  #9 opened 2 months ago
  0