boltz-vsynthes 0.0.10__py3-none-any.whl → 0.0.12__py3-none-any.whl

boltz/data/crop/affinity.py

@@ -105,7 +105,7 @@ class AffinityCropper(Cropper):
                 new_tokens = chain_tokens
             else:
                 # First limit to the maximum set of tokens, with the
-                # neighboorhood on both sides to handle edges. This
+                # neighborhood on both sides to handle edges. This
                 # is mostly for efficiency with the while loop below.
                 min_idx = token["res_idx"] - self.neighborhood_size
                 max_idx = token["res_idx"] + self.neighborhood_size

boltz/data/crop/boltz.py

@@ -245,7 +245,7 @@ class BoltzCropper(Cropper):
                 new_tokens = chain_tokens
             else:
                 # First limit to the maximum set of tokens, with the
-                # neighboorhood on both sides to handle edges. This
+                # neighborhood on both sides to handle edges. This
                 # is mostly for efficiency with the while loop below.
                 min_idx = token["res_idx"] - neighborhood_size
                 max_idx = token["res_idx"] + neighborhood_size

boltz/data/parse/pdb.py

@@ -8,9 +8,6 @@ from Bio.PDB.Polypeptide import PPBuilder
 from Bio.Data.IUPACData import protein_letters_3to1
 from rdkit import Chem
 from rdkit.Chem.rdchem import Mol
-from Bio.SeqUtils import seq1
-from collections import defaultdict
-
 
 from boltz.data.types import Target
 from boltz.data.parse.schema import parse_boltz_schema
@@ -41,35 +38,27 @@ def parse_pdb(
         Dictionary containing sequences and bonds.
     """
     # Read PDB file
-    sequences = []
-    sequence_by_chain = defaultdict(list)
+    parser = PDBParser(QUIET=True)
+    structure = parser.get_structure("protein", str(pdb_path))
+    ppb = PPBuilder()
 
-    # Parse SEQRES records directly
-    with open(pdb_path) as f:
-        for line in f:
-            if line.startswith("SEQRES"):
-                parts = line.split()
-                chain_id = parts[2]
-                residues = parts[4:]
-                for res in residues:
-                    try:
-                        aa = seq1(res)
-                    except KeyError:
-                        aa = 'X'
-                    sequence_by_chain[chain_id].append(aa)
+    # Convert to yaml format
+    sequences = []
+    for model in structure:
+        for chain in model:
+            for pp in ppb.build_peptides(chain):
+                seq = str(pp.get_sequence())
+                if seq:  # Only add if sequence is not empty
+                    sequences.append({
+                        "protein": {
+                            "id": chain.id,
+                            "sequence": seq,
+                            "modifications": [],
+                        }
+                    })
 
-    # Convert to yaml-style list
-    for chain_id, aa_list in sequence_by_chain.items():
-        sequences.append({
-            "protein": {
-                "id": chain_id,
-                "sequence": ''.join(aa_list),
-                "modifications": [],
-            }
-        })
-    print(sequences)
     return {
         "sequences": sequences,
         "bonds": [],
         "version": 1,
-    }
+    } 

boltz/data/parse/pdb_download.py

@@ -8,8 +8,6 @@ from Bio.PDB.PDBParser import PDBParser
 from Bio.PDB.Polypeptide import PPBuilder
 from rdkit import Chem
 from rdkit.Chem.rdchem import Mol
-from Bio.SeqUtils import seq1
-from collections import defaultdict
 
 from boltz.data.types import Target
 from boltz.data.parse.schema import parse_boltz_schema
@@ -81,33 +79,25 @@ def parse_pdb_id(
     # Download PDB file
     pdb_path = download_pdb(pdb_id, cache_dir)
 
-    sequences = []
-    sequence_by_chain = defaultdict(list)
-
-    # Parse SEQRES records directly
-    with open(pdb_path) as f:
-        for line in f:
-            if line.startswith("SEQRES"):
-                parts = line.split()
-                chain_id = parts[2]
-                residues = parts[4:]
-                for res in residues:
-                    try:
-                        aa = seq1(res)
-                    except KeyError:
-                        aa = 'X'
-                    sequence_by_chain[chain_id].append(aa)
+    # Read PDB file
+    parser = PDBParser(QUIET=True)
+    structure = parser.get_structure("protein", str(pdb_path))
+    ppb = PPBuilder()
 
-    # Convert to yaml-style list
-    for chain_id, aa_list in sequence_by_chain.items():
-        sequences.append({
-            "protein": {
-                "id": chain_id,
-                "sequence": ''.join(aa_list),
-                "modifications": [],
-            }
-        })
-    print(sequences)
+    # Convert to yaml format
+    sequences = []
+    for model in structure:
+        for chain in model:
+            for pp in ppb.build_peptides(chain):
+                seq = str(pp.get_sequence())
+                if seq:  # Only add if sequence is not empty
+                    sequences.append({
+                        "protein": {
+                            "id": chain.id,
+                            "sequence": seq,
+                            "modifications": [],
+                        }
+                    })
 
     return {
         "sequences": sequences,

boltz/data/tokenize/boltz2.py

@@ -133,7 +133,7 @@ def tokenize_structure(  # noqa: C901, PLR0915
     chains = struct.chains[struct.mask]
 
     # Ensemble atom id start in coords table.
-    # For cropper and other operations, harcoded to 0th conformer.
+    # For cropper and other operations, hardcoded to 0th conformer.
     offset = struct.ensemble[0]["atom_coord_idx"]
 
     for chain in chains:

boltz/main.py

@@ -9,7 +9,7 @@ from dataclasses import asdict, dataclass
 from functools import partial
 from multiprocessing import Pool
 from pathlib import Path
-from typing import Literal, Optional, List
+from typing import Literal, Optional
 
 import click
 import torch
@@ -18,8 +18,6 @@ from pytorch_lightning.strategies import DDPStrategy
 from pytorch_lightning.utilities import rank_zero_only
 from rdkit import Chem
 from tqdm import tqdm
-import time
-from datetime import datetime
 
 from boltz.data import const
 from boltz.data.module.inference import BoltzInferenceDataModule
@@ -205,21 +203,22 @@ def download_boltz2(cache: Path) -> None:
         The cache directory.
 
     """
-    # Use /tmp if possible for faster local disk I/O
-    if str(cache).startswith("/home") or str(cache).startswith("/mnt"):
-        cache = Path("/tmp/boltz_cache")
-    cache.mkdir(parents=True, exist_ok=True)
-
     # Download CCD
     mols = cache / "mols"
     tar_mols = cache / "mols.tar"
-    if not mols.exists():
+    if not tar_mols.exists():
         click.echo(
-            f"Downloading and extracting the CCD data to {mols}. "
+            f"Downloading the CCD data to {tar_mols}. "
             "This may take a bit of time. You may change the cache directory "
             "with the --cache flag."
         )
         urllib.request.urlretrieve(MOL_URL, str(tar_mols))  # noqa: S310
+    if not mols.exists():
+        click.echo(
+            f"Extracting the CCD data to {mols}. "
+            "This may take a bit of time. You may change the cache directory "
+            "with the --cache flag."
+        )
         with tarfile.open(str(tar_mols), "r") as tar:
             tar.extractall(cache)  # noqa: S202
 
@@ -984,7 +983,6 @@ def predict(  # noqa: C901, PLR0915, PLR0912
     torch.set_grad_enabled(False)
 
     # Ignore matmul precision warning
-    # torch.set_float32_matmul_precision('medium')
     torch.set_float32_matmul_precision("highest")
 
     # Set rdkit pickle logic
@@ -1031,16 +1029,14 @@ def predict(  # noqa: C901, PLR0915, PLR0912
             msg = f"Method {method} not supported. Supported: {method_names}"
             raise ValueError(msg)
 
-    # 1. Before and after process_inputs
-    t_process_inputs = time.time()
+    # Process inputs
     ccd_path = cache / "ccd.pkl"
     mol_dir = cache / "mols"
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Starting process_inputs...")
     process_inputs(
         data=data,
         out_dir=out_dir,
-        ccd_path=cache / "ccd.pkl",
-        mol_dir=cache / "mols",
+        ccd_path=ccd_path,
+        mol_dir=mol_dir,
         use_msa_server=use_msa_server,
         msa_server_url=msa_server_url,
         msa_pairing_strategy=msa_pairing_strategy,
@@ -1048,27 +1044,18 @@ def predict(  # noqa: C901, PLR0915, PLR0912
         preprocessing_threads=preprocessing_threads,
         max_msa_seqs=max_msa_seqs,
     )
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] process_inputs finished in {time.time() - t_process_inputs:.2f} seconds")
 
-    # 2. Before and after load manifest
-    t_manifest = time.time()
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Loading manifest...")
+    # Load manifest
     manifest = Manifest.load(out_dir / "processed" / "manifest.json")
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Manifest loaded in {time.time() - t_manifest:.2f} seconds")
 
-    # 3. Before and after Filter out existing predictions
-    t_filter = time.time()
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Filtering out existing predictions...")
+    # Filter out existing predictions
     filtered_manifest = filter_inputs_structure(
         manifest=manifest,
         outdir=out_dir,
         override=override,
     )
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Filtering finished in {time.time() - t_filter:.2f} seconds")
 
-    # 4. Before and after load processed data
-    t_processed = time.time()
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Loading processed data...")
+    # Load processed data
     processed_dir = out_dir / "processed"
     processed = BoltzProcessedInput(
         manifest=filtered_manifest,
@@ -1088,22 +1075,8 @@ def predict(  # noqa: C901, PLR0915, PLR0912
             (processed_dir / "mols") if (processed_dir / "mols").exists() else None
         ),
     )
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Processed data loaded in {time.time() - t_processed:.2f} seconds")
 
-    # 5. Before and after create prediction writer
-    t_writer = time.time()
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Creating prediction writer...")
-    pred_writer = BoltzWriter(
-        data_dir=processed.targets_dir,
-        output_dir=out_dir / "predictions",
-        output_format=output_format,
-        boltz2=model == "boltz2",
-    )
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Prediction writer created in {time.time() - t_writer:.2f} seconds")
-
-    # 6. Before and after set up trainer
-    t_trainer = time.time()
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Setting up trainer...")
+    # Set up trainer
     strategy = "auto"
     if (isinstance(devices, int) and devices > 1) or (
         isinstance(devices, list) and len(devices) > 1
@@ -1156,16 +1129,13 @@ def predict(  # noqa: C901, PLR0915, PLR0912
         devices=devices,
         precision=32 if model == "boltz1" else "bf16-mixed",
     )
-    print(f"[{datetime.now().strftime('%H:%M:%S')}] Trainer set up in {time.time() - t_trainer:.2f} seconds")
 
     if filtered_manifest.records:
         msg = f"Running structure prediction for {len(filtered_manifest.records)} input"
         msg += "s." if len(filtered_manifest.records) > 1 else "."
         click.echo(msg)
 
-        # 7. Before and after create data module
-        t_datamodule = time.time()
-        print(f"[{datetime.now().strftime('%H:%M:%S')}] Creating data module...")
+        # Create data module
         if model == "boltz2":
             data_module = Boltz2InferenceDataModule(
                 manifest=processed.manifest,
@@ -1186,16 +1156,14 @@ def predict(  # noqa: C901, PLR0915, PLR0912
                 num_workers=num_workers,
                 constraints_dir=processed.constraints_dir,
             )
-        print(f"[{datetime.now().strftime('%H:%M:%S')}] Data module created in {time.time() - t_datamodule:.2f} seconds")
 
-        # 8. Before and after load model
-        t_model = time.time()
-        print(f"[{datetime.now().strftime('%H:%M:%S')}] Loading model...")
+        # Load model
         if checkpoint is None:
             if model == "boltz2":
                 checkpoint = cache / "boltz2_conf.ckpt"
             else:
                 checkpoint = cache / "boltz1_conf.ckpt"
+
         predict_args = {
             "recycling_steps": recycling_steps,
             "sampling_steps": sampling_steps,
@@ -1205,17 +1173,17 @@ def predict(  # noqa: C901, PLR0915, PLR0912
             "write_full_pae": write_full_pae,
             "write_full_pde": write_full_pde,
         }
+
         steering_args = BoltzSteeringParams()
         steering_args.fk_steering = use_potentials
         steering_args.guidance_update = use_potentials
+
         model_cls = Boltz2 if model == "boltz2" else Boltz1
-        device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-        print(f"[{datetime.now().strftime('%H:%M:%S')}] Using device: {device}")
         model_module = model_cls.load_from_checkpoint(
             checkpoint,
             strict=True,
             predict_args=predict_args,
-            map_location=device,
+            map_location="cpu",
             diffusion_process_args=asdict(diffusion_params),
             ema=False,
             use_kernels=not no_kernels,
@@ -1224,17 +1192,13 @@ def predict(  # noqa: C901, PLR0915, PLR0912
             steering_args=asdict(steering_args),
         )
         model_module.eval()
-        print(f"[{datetime.now().strftime('%H:%M:%S')}] Model loaded in {time.time() - t_model:.2f} seconds")
 
-        # 9. Before and after compute structure predictions (predict)
-        t_predict = time.time()
-        print(f"[{datetime.now().strftime('%H:%M:%S')}] Computing structure predictions...")
+        # Compute structure predictions
         trainer.predict(
             model_module,
             datamodule=data_module,
             return_predictions=False,
         )
-        print(f"[{datetime.now().strftime('%H:%M:%S')}] Structure predictions computed in {time.time() - t_predict:.2f} seconds")
 
     # Check if affinity predictions are needed
     if any(r.affinity for r in manifest.records):
@@ -1287,13 +1251,11 @@ def predict(  # noqa: C901, PLR0915, PLR0912
         if affinity_checkpoint is None:
             affinity_checkpoint = cache / "boltz2_aff.ckpt"
 
-        device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-        print(f"[{datetime.now().strftime('%H:%M:%S')}] Using device: {device}")
         model_module = Boltz2.load_from_checkpoint(
             affinity_checkpoint,
             strict=True,
             predict_args=predict_affinity_args,
-            map_location=device,
+            map_location="cpu",
             diffusion_process_args=asdict(diffusion_params),
             ema=False,
             pairformer_args=asdict(pairformer_args),
@@ -1310,5 +1272,6 @@ def predict(  # noqa: C901, PLR0915, PLR0912
             return_predictions=False,
         )
 
+
 if __name__ == "__main__":
     cli()

boltz/model/layers/pairformer.py

@@ -325,13 +325,13 @@ class PairformerNoSeqModule(nn.Module):
                     z,
                     pair_mask,
                     chunk_size_tri_attn,
-                    use_kernels=use_kernels,
+                    use_kernels,
                 )
             else:
                 z = layer(
                     z,
                     pair_mask,
                     chunk_size_tri_attn,
-                    use_kernels=use_kernels,
+                    use_kernels,
                 )
         return z

boltz/model/models/boltz2.py

@@ -593,7 +593,7 @@ class Boltz2(LightningModule):
                     pred_distogram_logits=(
                         dict_out["pdistogram"][
                             :, :, :, 0
-                        ].detach()  # TODO only implemeted for 1 distogram
+                        ].detach()  # TODO only implemented for 1 distogram
                     ),
                     multiplicity=diffusion_samples,
                     run_sequentially=run_confidence_sequentially,

boltz/utils/sdf_to_pre_affinity_npz.py

@@ -0,0 +1,35 @@
+#!/usr/bin/env python3
+"""
+boltz-sdf-to-pre-affinity: Convert a docked SDF file (protein-ligand complex) to Boltz pre_affinity_[ligand_id].npz format.
+
+Usage:
+    boltz-sdf-to-pre-affinity --sdf docked_pose.sdf --ligand_id ligand1 --output pre_affinity_ligand1.npz
+
+This command is available after installing boltz with pip.
+"""
+import argparse
+from pathlib import Path
+
+from boltz.data.parse.sdf import parse_sdf
+from boltz.data.types import StructureV2
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Convert SDF to Boltz pre_affinity_*.npz format.")
+    parser.add_argument('--sdf', type=str, required=True, help='Input SDF file (protein-ligand complex)')
+    parser.add_argument('--ligand_id', type=str, required=True, help='Ligand ID (used in output filename)')
+    parser.add_argument('--output', type=str, required=True, help='Output .npz file path')
+    args = parser.parse_args()
+
+    sdf_path = Path(args.sdf)
+    output_path = Path(args.output)
+
+    # Parse the SDF file to StructureV2
+    structure: StructureV2 = parse_sdf(sdf_path)
+
+    # Save as pre_affinity_[ligand_id].npz
+    structure.dump(output_path)
+    print(f"Saved: {output_path}")
+
+if __name__ == "__main__":
+    main() 

{boltz_vsynthes-0.0.10.dist-info → boltz_vsynthes-0.0.12.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: boltz-vsynthes
-Version: 0.0.10
+Version: 0.0.12
 Summary: Boltz for VSYNTHES
 Requires-Python: <3.13,>=3.10
 Description-Content-Type: text/markdown
@@ -43,8 +43,8 @@ Dynamic: license-file
   <img src="docs/boltz2_title.png" width="300"/>
   <img src="https://model-gateway.boltz.bio/a.png?x-pxid=bce1627f-f326-4bff-8a97-45c6c3bc929d" />
 
-[Boltz-1](https://doi.org/10.1101/2024.11.19.624167) | [Boltz-2](https://bit.ly/boltz2-pdf) |
-[Slack](https://join.slack.com/t/boltz-community/shared_invite/zt-3751cpmn6-kDLgLcQFMOPeUdFIJd4oqQ) <br> <br>
+[Boltz-1](https://doi.org/10.1101/2024.11.19.624167) | [Boltz-2](https://doi.org/10.1101/2025.06.14.659707) |
+[Slack](https://join.slack.com/t/boltz-community/shared_invite/zt-37uc4m8t2-gbbph6ka704ORcDCHLlFKg) <br> <br>
 </div>
 
 
@@ -56,7 +56,7 @@ Dynamic: license-file
 
 Boltz is a family of models for biomolecular interaction prediction. Boltz-1 was the first fully open source model to approach AlphaFold3 accuracy. Our latest work Boltz-2 is a new biomolecular foundation model that goes beyond AlphaFold3 and Boltz-1 by jointly modeling complex structures and binding affinities, a critical component towards accurate molecular design. Boltz-2 is the first deep learning model to approach the accuracy of physics-based free-energy perturbation (FEP) methods, while running 1000x faster — making accurate in silico screening practical for early-stage drug discovery.
 
-All the code and weights are provided under MIT license, making them freely available for both academic and commercial uses. For more information about the model, see the [Boltz-1](https://doi.org/10.1101/2024.11.19.624167) and [Boltz-2](https://bit.ly/boltz2-pdf) technical reports. To discuss updates, tools and applications join our [Slack channel](https://join.slack.com/t/boltz-community/shared_invite/zt-34qg8uink-V1LGdRRUf3avAUVaRvv93w).
+All the code and weights are provided under MIT license, making them freely available for both academic and commercial uses. For more information about the model, see the [Boltz-1](https://doi.org/10.1101/2024.11.19.624167) and [Boltz-2](https://doi.org/10.1101/2025.06.14.659707) technical reports. To discuss updates, tools and applications join our [Slack channel](https://join.slack.com/t/boltz-community/shared_invite/zt-37uc4m8t2-gbbph6ka704ORcDCHLlFKg).
 
 ## Installation
 
@@ -108,7 +108,7 @@ If you're interested in retraining the model, currently for Boltz-1 but soon for
 
 ## Contributing
 
-We welcome external contributions and are eager to engage with the community. Connect with us on our [Slack channel](https://join.slack.com/t/boltz-community/shared_invite/zt-34qg8uink-V1LGdRRUf3avAUVaRvv93w) to discuss advancements, share insights, and foster collaboration around Boltz-2.
+We welcome external contributions and are eager to engage with the community. Connect with us on our [Slack channel](https://join.slack.com/t/boltz-community/shared_invite/zt-37uc4m8t2-gbbph6ka704ORcDCHLlFKg) to discuss advancements, share insights, and foster collaboration around Boltz-2.
 
 Boltz also runs on Tenstorrent hardware thanks to a [fork](https://github.com/moritztng/tt-boltz) by Moritz Thüning.
 
@@ -126,8 +126,8 @@ If you use this code or the models in your research, please cite the following p
   author = {Passaro, Saro and Corso, Gabriele and Wohlwend, Jeremy and Reveiz, Mateo and Thaler, Stephan and Somnath, Vignesh Ram and Getz, Noah and Portnoi, Tally and Roy, Julien and Stark, Hannes and Kwabi-Addo, David and Beaini, Dominique and Jaakkola, Tommi and Barzilay, Regina},
   title = {Boltz-2: Towards Accurate and Efficient Binding Affinity Prediction},
   year = {2025},
-  doi = {},
-  journal = {}
+  doi = {10.1101/2025.06.14.659707},
+  journal = {bioRxiv}
 }
 
 @article{wohlwend2024boltz1,

{boltz_vsynthes-0.0.10.dist-info → boltz_vsynthes-0.0.12.dist-info}/RECORD

@@ -1,13 +1,13 @@
 boltz/__init__.py,sha256=F_-so3S40iZrSZ89Ge4TS6aZqwWyZXq_H4AXGDlbA_g,187
-boltz/main.py,sha256=hLNHO-F3b9jDVQP1NHWhl1m4kAXBtGAwv3qsyoCjROo,42864
+boltz/main.py,sha256=zBLxa6T8hxcBs7gj1BnWfgJSx6uki8iV-QgClvoaiSA,39951
 boltz/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 boltz/data/const.py,sha256=1M-88Z6HkfKY6MkNtqcj3b9P-oX9xEXluh3qM_u8dNU,26779
 boltz/data/mol.py,sha256=maOpPHEGX1VVXCIFY6pQNGF7gUBZPAfgSvuPf2QO1yc,34268
 boltz/data/pad.py,sha256=O4CGOOc5TwFuuWeP7hKjMIIsljdfLj-VJtXQeVXFx8s,2066
 boltz/data/types.py,sha256=4w9brpOCQe16AyByNrxz7pjIzrgzFNihtik3aaHvKaE,21965
 boltz/data/crop/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-boltz/data/crop/affinity.py,sha256=toD0tly_SB2YZgkPS12hK3FRw8RH3jI38D8p1syhsMQ,5733
-boltz/data/crop/boltz.py,sha256=oJSxaalUMXUNcYzoXWUn_CLaqU1LZApBlxnk5UMh73I,9841
+boltz/data/crop/affinity.py,sha256=NqGtvI83iOtWGDmAVwTpXyIzxeSSAHDauGCrmxqvsYs,5732
+boltz/data/crop/boltz.py,sha256=bloP9vk6cmKIZ4zZoO04IXhbouSIEFvKbxBUoLEDQHE,9840
 boltz/data/crop/cropper.py,sha256=BGDgFwNVUPyBZvcqnJwPyH_IfZ1NAYCZ_KMwbQf4HK4,1145
 boltz/data/feature/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 boltz/data/feature/featurizer.py,sha256=yYWsgW3VAH9GOgfitZPt__gsB2bcxrzcvUQjKVeRYhc,44574
@@ -38,8 +38,8 @@ boltz/data/parse/csv.py,sha256=Hcq8rJW2njczahEr8jfd_o-zxLaNSgJ3YIoC9srIqpw,2518
 boltz/data/parse/fasta.py,sha256=taI4s_CqPtyF0XaLJAsVAJHCL0GXm2g1g8Qeccdxikk,3906
 boltz/data/parse/mmcif.py,sha256=25kEXCkx-OuaawAs7cdz0fxdRu5_CCO0AV00u84PrjQ,36822
 boltz/data/parse/mmcif_with_constraints.py,sha256=WHYZckSqUwu-Nb9vmVmxHmC7uxwVrF7AVUeVKsc5wGQ,51473
-boltz/data/parse/pdb.py,sha256=Gnu1UiHSuTb1DWcy6wUbq4o6zIoU-_AObgkLuIdY6qE,1912
-boltz/data/parse/pdb_download.py,sha256=vcgCn6FlFpRduOLnTFFn91pBDBxGUfQa3MOQs3KaZ10,2957
+boltz/data/parse/pdb.py,sha256=iybk4p2UgUy_ABGprDq_xxyPSdm1HAZsGTM0lhxVEwM,1654
+boltz/data/parse/pdb_download.py,sha256=wge-scX-lOatX0q83W1wOsaql99rYp-6uGWSHEc995M,2718
 boltz/data/parse/schema.py,sha256=kNu28U2_MGiecwWNlcxgaDH3WOcO0P-q2LdoSPSb66w,63826
 boltz/data/parse/sdf.py,sha256=fs3MQVClDcCzxJaeVYiDuoh-fUrYc8Tcd5Bz8ws3FKI,2052
 boltz/data/parse/yaml.py,sha256=GRFRMtDD4PQ4PIpA_S1jj0vRaEu2LlZd_g4rN1zUrNo,1505
@@ -50,7 +50,7 @@ boltz/data/sample/random.py,sha256=jmennDNfkuq2UByn55wrQTwNPCOBuu6_jBJT4VraL28,1
 boltz/data/sample/sampler.py,sha256=LXV4FLQPdMC_ja4MjPj0kicErr8UXSWrKehPBF5zOxQ,1095
 boltz/data/tokenize/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 boltz/data/tokenize/boltz.py,sha256=h4rb2Jm9u5SP0xWygg4zE2kJ_RySRvYYDwvwlKqsSa0,6643
-boltz/data/tokenize/boltz2.py,sha256=2xWq72_CMkyzQS8jxC4ffoePAhBoJAFJ7JGNCJ0Xj5A,12821
+boltz/data/tokenize/boltz2.py,sha256=to7E9pJOlC7yVMpgKTNZvH2jSMEEot_hAFrKXzW2bTI,12822
 boltz/data/tokenize/tokenizer.py,sha256=9okkO89A1RvQd5tcW9VLIKZCSvvUALosy2itdN1mDIE,484
 boltz/data/write/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 boltz/data/write/mmcif.py,sha256=zHwe4ulietjCVz5JMzl4_bu4liqwwcUKlgr9orVVd2Q,12177
@@ -66,7 +66,7 @@ boltz/model/layers/dropout.py,sha256=kUr8_a84CtyYryQO1_VK44jGAD51qV4HTUQWXMV9xCc
 boltz/model/layers/initialize.py,sha256=KnK9tUfGvCWkdwbcLvJYoL-Wmlo9YcIcKBtdrXaUG30,2463
 boltz/model/layers/outer_product_mean.py,sha256=AZoWhbaZHvVCmKSRWJ8Jw7xaCiN232VBjFsMclk_SR8,3151
 boltz/model/layers/pair_averaging.py,sha256=XXZipyOlxZv4QR4Uk2cIUe8Ow_Xpr45eECqsWouNAVs,4332
-boltz/model/layers/pairformer.py,sha256=khDRUHbJWIBoO7zJYZ3t5VMA5FoOuFInvhCHnL_zFN0,10326
+boltz/model/layers/pairformer.py,sha256=Oh8mOXOBgjSbFXuc4kjqnOkYb5ZtIRLZTYX5TXi_CyI,10302
 boltz/model/layers/relative.py,sha256=fLOwADG5VDDdH2Gqis-L6Lw8hnWGxCPjfV6t8kYeEIc,1875
 boltz/model/layers/transition.py,sha256=zFClGmqmaUVUIFSd4W4pVY17lvin2NpvaZMKuec4zzc,2290
 boltz/model/layers/triangular_mult.py,sha256=uQahL0rwZ04xPK95UOWjHlB48VypDBSaIXKVDmWCzJw,5996
@@ -85,7 +85,7 @@ boltz/model/loss/distogramv2.py,sha256=dFgMGwpdLK4-skHJwvpERG10KfF3ZUN1T9_hUj-iW
 boltz/model/loss/validation.py,sha256=gYpbag9mulg5HJPXjOUFaMV9XSYX_s2bIQ0iYjiAow0,33501
 boltz/model/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 boltz/model/models/boltz1.py,sha256=x-x0b3VAXiAkPUBePnF56k1aYEPNgX1M6GtNCYVdCso,51718
-boltz/model/models/boltz2.py,sha256=j8UqodTeaMGtpdEzopmfc7JW6td7nC2q6CCA7Fbnzy0,51522
+boltz/model/models/boltz2.py,sha256=3XOWjUWaSJquw8Xdp7JItDUnVDyoC0qtx3q4MFQrd38,51523
 boltz/model/modules/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 boltz/model/modules/affinity.py,sha256=FktI2wrkDqsjGHJOuvzVrZK78MOPjU65QN0l6sB1QPQ,7041
 boltz/model/modules/confidence.py,sha256=sXGymZiiMtfXPkUvHpa2KCCvNY79D8jXXEx9Gz2rNFs,17475
@@ -108,10 +108,11 @@ boltz/model/potentials/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3
 boltz/model/potentials/potentials.py,sha256=vev8Vjfs-ML1hyrdv_R8DynG4wSFahJ6nzPWp7CYQqw,17507
 boltz/model/potentials/schedules.py,sha256=m7XJjfuF9uTX3bR9VisXv1rvzJjxiD8PobXRpcBBu1c,968
 boltz/utils/sdf_splitter.py,sha256=ZHn_syOcmm-fDnJ3YEGyGv_vYz2IRzUW7vbbMSU2JBY,2108
+boltz/utils/sdf_to_pre_affinity_npz.py,sha256=ev0s2pS8NoB-_3BkSjRKk3GMnqjOxTRCH-r9avKYGOg,1212
 boltz/utils/yaml_generator.py,sha256=ermWIG-BE6nNWHFvpEwpk92N9J-YATpGXZGLvD1I2oQ,4012
-boltz_vsynthes-0.0.10.dist-info/licenses/LICENSE,sha256=8GZ_1eZsUeG6jdqgJJxtciWzADfgLEV4LY8sKUOsJhc,1102
-boltz_vsynthes-0.0.10.dist-info/METADATA,sha256=HGhImUlu6jhPROzbqGDPk2K1Ckq3yVqY7ioGTzkDRfs,7171
-boltz_vsynthes-0.0.10.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-boltz_vsynthes-0.0.10.dist-info/entry_points.txt,sha256=R_G5TQWoiTDF3a2n7WQfpMLGufFd1lm5tZXVNaczlss,143
-boltz_vsynthes-0.0.10.dist-info/top_level.txt,sha256=MgU3Jfb-ctWm07YGMts68PMjSh9v26D0gfG3dFRmVFA,6
-boltz_vsynthes-0.0.10.dist-info/RECORD,,
+boltz_vsynthes-0.0.12.dist-info/licenses/LICENSE,sha256=8GZ_1eZsUeG6jdqgJJxtciWzADfgLEV4LY8sKUOsJhc,1102
+boltz_vsynthes-0.0.12.dist-info/METADATA,sha256=XPcHZExavOK9oz8BaZnp31f-acFOXbCKwmDORbDxaP8,7235
+boltz_vsynthes-0.0.12.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+boltz_vsynthes-0.0.12.dist-info/entry_points.txt,sha256=WlXNdHTJF1ahGN-H1QpgMbZm_5SobXY05eiuT8buycw,212
+boltz_vsynthes-0.0.12.dist-info/top_level.txt,sha256=MgU3Jfb-ctWm07YGMts68PMjSh9v26D0gfG3dFRmVFA,6
+boltz_vsynthes-0.0.12.dist-info/RECORD,,

{boltz_vsynthes-0.0.10.dist-info → boltz_vsynthes-0.0.12.dist-info}/entry_points.txt

@@ -1,4 +1,5 @@
 [console_scripts]
 boltz = boltz.main:cli
 boltz-generate-yaml = boltz.utils.yaml_generator:main
+boltz-sdf-to-pre-affinity = boltz.utils.sdf_to_pre_affinity_npz:main
 boltz-split-sdf = boltz.utils.sdf_splitter:main