[BUG] RFD3 is missing training data artifacts

[amorehead]

Describe the bug
RFD3's training resources contain most of the required materials for others to reproduce its results. However, a few key data artifacts are currently missing, which makes exact reproducibility unattainable for the time being.

Expected behavior
In no particular order, the following data artifacts (i.e., data files or datasets) should be publicly available for download, to make RFD3's training fully reproducible.

1. foundry/models/rfd3/configs/datasets/train/pdb/na_complex_distillation.yaml

   Line 12 in cee116d

   data: /projects/ml/prot_dna/transcriptionFactor_distillation_rf3.newDL.csv

2. foundry/models/rfd3/configs/datasets/train/rfd3_monomer_distillation.yaml

   Line 21 in cee116d

   data: ${paths.data.monomer_distillation_parquet_dir}/af2_distillation_facebook.parquet

3. foundry/models/rfd3/configs/datasets/val/val_examples/bcov_ppi_easy_medium_with_ori.yaml

   Lines 4 to 151 in cee116d

   	insulinr:
   	input: /projects/ml/aa_design/benchmarks/bcov_af3_ppi_benchmark/insulin_target.pdb
   	contig: 100-100,/0,B1-150
   	contig_atoms: '{}'
   	length: 250-250
   	redesign_motif_sidechains: false
   	atom_level_hotspots:
   	B59:
   	CG,CZ: 1
   	B83:
   	CG,CZ: 1
   	B91:
   	CG,CZ: 1
   	pdl1:
   	input: /projects/ml/aa_design/benchmarks/bcov_af3_ppi_benchmark/5o45_pdl1.pdb
   	contig: 100-100,/0,B1-115
   	contig_atoms: '{}'
   	length: 215-215
   	redesign_motif_sidechains: false
   	atom_level_hotspots:
   	B40:
   	CG,CZ: 1
   	B99:
   	CG,SD: 1
   	B107:
   	CG,CZ: 1
   	vegfr:
   	input: /projects/ml/aa_design/benchmarks/bcov_af3_ppi_benchmark/vegfr_2x1w_and_af2_B.pdb
   	contig: 100-100,/0,B1-200
   	contig_atoms: '{}'
   	length: 300-300
   	redesign_motif_sidechains: false
   	atom_level_hotspots:
   	B13:
   	CG1,CG2: 1
   	B15:
   	CG,CZ: 1
   	B43:
   	CG,CZ: 1
   	B75:
   	CG,SD: 1
   	B89:
   	CD1,CG2: 1
   	B91:
   	CG1,CG2: 1
   	B187:
   	CG,CD1: 1
   	rbd:
   	input: /projects/ml/aa_design/benchmarks/bcov_af3_ppi_benchmark/COVID19_target.pdb
   	contig: 100-100,/0,B1-195
   	contig_atoms: '{}'
   	length: 295-295
   	redesign_motif_sidechains: false
   	atom_level_hotspots:
   	B89:
   	CG,CZ: 1
   	B121:
   	CG,CZ: 1
   	B123:
   	CG,CD1: 1
   	B124:
   	CG,CZ: 1
   	B141:
   	CG,CZ: 1
   	B157:
   	CG,CZ: 1
   	B163:
   	CG,CZ: 1
   	B165:
   	CG,CZ: 1
   	B173:
   	CG,CZ: 1
   	cd28:
   	input: /projects/ml/aa_design/benchmarks/bcov_af3_ppi_benchmark/cd28_1yjd_B.pdb
   	contig: 100-100,/0,B1-118
   	contig_atoms: '{}'
   	length: 218-218
   	redesign_motif_sidechains: false
   	atom_level_hotspots:
   	B51:
   	CG,CZ: 1
   	B61:
   	CG,CZ: 1
   	B99:
   	CG,SD: 1
   	B104:
   	CG,CZ: 1
   	il2ra:
   	input: /projects/ml/aa_design/benchmarks/bcov_af3_ppi_benchmark/il2ra_1z92_B.pdb
   	contig: 100-100,/0,B1-122
   	contig_atoms: '{}'
   	length: 222-222
   	redesign_motif_sidechains: false
   	atom_level_hotspots:
   	B3:
   	CG,CD1: 1
   	B26:
   	CG,SD: 1
   	B43:
   	CG,CD1: 1
   	B44:
   	CG,CZ: 1
   	B46:
   	CG,CD1: 1
   	il10ra:
   	input: /projects/ml/aa_design/benchmarks/bcov_af3_ppi_benchmark/il10rb_1lqs_B.pdb
   	contig: 100-100,/0,B1-207
   	contig_atoms: '{}'
   	length: 307-307
   	redesign_motif_sidechains: false
   	atom_level_hotspots:
   	B39:
   	CG,CD1: 1
   	B50:
   	CD1,CG2: 1
   	B59:
   	CG,CZ: 1
   	B63:
   	CA,CB: 1
   	B64:
   	CG1,CG2: 1
   	B66:
   	CG,CD1: 1
   	tie2:
   	input: /projects/ml/aa_design/benchmarks/bcov_af3_ppi_benchmark/tie2_2gy5_official_B.pdb
   	contig: 100-100,/0,B1-188
   	contig_atoms: '{}'
   	length: 288-288
   	redesign_motif_sidechains: false
   	atom_level_hotspots:
   	B132:
   	CG1,CG2: 1
   	B134:
   	CG,CZ: 1
   	B135:
   	CG,CD: 1
   	B139:
   	CG,CZ: 1
   	B140:
   	CD1,CG2: 1
   	B154:
   	CG1,CG2: 1
   	B156:
   	CG,CD1: 1
   	B167:
   	CG,CZ: 1
   	B172:
   	CD1,CG2: 1

4. foundry/models/rfd3/configs/datasets/val/dna_binder_design5.yaml

   Line 7 in cee116d

   data: ${paths.data.design_benchmark_data_dir}/dna_binder.json

5. foundry/models/rfd3/configs/datasets/val/dna_binder_long.yaml

   Line 7 in cee116d

   data: ${paths.root_dir}/tests/dna.json

6. foundry/models/rfd3/configs/datasets/val/dna_binder_short.yaml

   Line 7 in cee116d

   data: ${paths.root_dir}/rfd3/tests/test_data/dna.json

7. foundry/models/rfd3/configs/datasets/val/indexed.yaml

   Line 7 in cee116d

   data: ${paths.data.design_benchmark_data_dir}/indexed.json

8. foundry/models/rfd3/configs/datasets/val/mcsa_41_short_rigid.yaml

   Line 9 in cee116d

   data: ${paths.data.design_benchmark_data_dir}/mcsa_41_short_rigid_new.json

9. foundry/models/rfd3/configs/datasets/val/mcsa_41.yaml

   Line 7 in cee116d

   data: ${paths.data.design_benchmark_data_dir}/mcsa_41.json

10. foundry/models/rfd3/configs/datasets/val/sm_binder_hbonds_short.yaml

    Line 7 in cee116d

    data: ${paths.data.design_benchmark_data_dir}/sm_binder_hbonds_sampled.json

11. foundry/models/rfd3/configs/datasets/val/sm_binder_hbonds.yaml

    Line 6 in cee116d

    data: ${paths.data.design_benchmark_data_dir}/sm_binder_hbonds.json

12. foundry/models/rfd3/configs/datasets/val/unconditional_deep.yaml

    Line 7 in cee116d

    data: ${paths.data.design_benchmark_data_dir}/unconditional_deep.json

13. foundry/models/rfd3/configs/datasets/val/unconditional.yaml

    Line 6 in cee116d

    data: ${paths.data.design_benchmark_data_dir}/monomer.json

14. foundry/models/rfd3/configs/datasets/val/unindexed.yaml

    Line 7 in cee116d

    data: ${paths.data.design_benchmark_data_dir}/unindexed.json

15. foundry/models/rfd3/configs/paths/data/default.yaml

    Lines 6 to 7 in cee116d

    	monomer_distillation_data_dir: /squash/af2_distillation_facebook/
    	monomer_distillation_parquet_dir: /projects/ml/datahub/dfs/distillation/af2_distillation_facebook

16. foundry/models/rfd3/configs/paths/data/default.yaml

    Lines 12 to 13 in cee116d

    	design_benchmark_data_dir: /projects/ml/aa_design/benchmarks
    	design_model_weight_dir: /projects/ml/aa_design/models

17. foundry/models/rfd3/configs/paths/data/default.yaml

    Line 18 in cee116d

    cif_cache_dir: /net/tukwila/ncorley/cifutils/cache

18. (Probably optional):

    foundry/models/rfd3/configs/paths/data/default.yaml

    Line 16 in cee116d

    residue_cache_dir: /net/tukwila/lschaaf/datahub/MACE-Egret-3-noH/mace_embeddings

    
    Additional context
    Thank you for open-sourcing RFD3's code and configs! These are significant resources from which the research community has already begun benefitting.