SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.04313 · ICLR · 2023

Geometry-Complete Diffusion for 3D Molecule Generation and Optimization

Alex Morehead, Jianlin Cheng

arXiv · PDF · Open in the Atlas

Code that ran

7 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 4 failures of our instrument; 3 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2302.04313.json · badge: /api/ran/2302.04313.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
bioinfomachinelearning/gcdm-sbdd canonical (classifier label)
classifier basis not recorded on the link
7 ran · 2 no recorded run
bioinfomachinelearning/gcpnet canonical (classifier label)
classifier basis not recorded on the link
0 ran · 1 no recorded run
FunctionStatusWhere it lives
filter_and_flatten Ran
ran on a synthesized input
bioinfomachinelearning/gcdm-sbdd/process_bindingmoad.py
code served (permissive licence) · get_code("7f62ceaa80a2427c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import random
from collections import defaultdict

# Build a ligand_dict structure that matches what the function expects
# The function iterates: for c in ligand_dict -> for p in ligand_dict[c] -> for m in ligand_dict[c][p]
# Each m is a tuple where:
#   - m[0] contains "ligand_name:ligand_chain:ligand_resi" format
#   - m[1] is a status string (e.g., 'valid')
#   - m[3] is a QED score (float)

ligand_dict = {
    'compound_1': {
        'protein_1': [
            ('aspirin:A:123', 'valid', 'extra_data', 0.85),  # m[0], m[1], m[2], m[3]
            ('aspirin:A:123', 'valid', 'extra_data', 0.92),
            ('ibuprofen:B:456', 'valid', 'extra_data', 0.78),
            ('ibuprofen:B:456', 'invalid', 'extra_data', 0.65),
        ],
        'protein_2': [
            ('paracetamol:C:789', 'valid', 'extra_data', 0.88),
            ('aspirin:A:123', 'valid', 'extra_data', 0.91),
        ],
    },
    'compound_2': {
        'protein_3': [
            ('naproxen:D:101', 'valid', 'extra_data', 0.80),
            ('ibuprofen:B:456', 'valid', 'extra_data', 0.77),
        ],
    },
}

qed_thresh = 0.75
max_occurences = 2
seed = 42

ARGS = {
    'ligand_dict': ligand_dict,
    'qed_thresh': qed_thresh,
    'max_occurences': max_occurences,
    'seed': seed,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_bb_transform Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
bioinfomachinelearning/gcdm-sbdd/geometry_utils.py
code served (permissive licence) · get_code("c81e137b6d163914")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create small batch of backbone atom coordinates
# n=2 (batch size), 3 coordinates per atom
n_samples = 2

# N atom positions (nitrogen)
n_xyz = np.array([
    [1.5, 0.5, 0.2],
    [-0.3, 1.2, 0.8]
], dtype=np.float32)

# CA atom positions (alpha carbon) 
ca_xyz = np.array([
    [0.0, 0.0, 0.0],
    [0.5, 0.5, 0.5]
], dtype=np.float32)

# C atom positions (carbonyl carbon)
c_xyz = np.array([
    [-1.2, 0.8, -0.3],
    [1.1, -0.4, 0.2]
], dtype=np.float32)

ARGS = {
    "n_xyz": n_xyz,
    "ca_xyz": ca_xyz,
    "c_xyz": c_xyz,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_n_nodes Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
bioinfomachinelearning/gcdm-sbdd/process_crossdock.py
code served (permissive licence) · get_code("be813c0c5a49cea2")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (5, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from scipy.ndimage import gaussian_filter

# Create sample masks
# lig_mask and pocket_mask should have the same unique indices
# Each value represents a node index, and counts represent the number of nodes
np.random.seed(42)

# Create simple masks with matching indices
# Let's use indices 0, 1, 2 for 3 different ligand-pocket pairs
lig_mask = np.array([0, 0, 0, 1, 1, 2, 2, 2, 2])
pocket_mask = np.array([0, 0, 1, 1, 1, 2, 2, 2, 2])

# Verify they have the same unique indices
unique_lig = np.unique(lig_mask)
unique_pocket = np.unique(pocket_mask)
assert np.array_equal(unique_lig, unique_pocket), "Masks must have matching unique indices"

ARGS = {
    "lig_mask": lig_mask,
    "pocket_mask": pocket_mask,
    "smooth_sigma": None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (5, 5) — shape (5, 5) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
ligand_list_to_dict Ran
ran on a synthesized input
bioinfomachinelearning/gcdm-sbdd/process_finetuning_set.py
code served (permissive licence) · get_code("667d353a0628f4a1")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned defaultdict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from collections import defaultdict

# Analyze the function: it unpacks tuples from ligand_list as (_, p, m)
# and groups m values by p key into a defaultdict(list)
# So ligand_list should be a list of 3-tuples where:
# - First element (ignored via _)
# - Second element p is used as a key
# - Third element m is appended to lists

# Create a small ligand_list with distinct identifiable values
ligand_list = [
    ("ignored_1", "protein_A", "molecule_1"),
    ("ignored_2", "protein_A", "molecule_2"),
    ("ignored_3", "protein_B", "molecule_3"),
    ("ignored_4", "protein_B", "molecule_4"),
    ("ignored_5", "protein_C", "molecule_5"),
]

ARGS = {
    "ligand_list": ligand_list,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type defaultdict — a defaultdict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a defaultdict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_label_file Ran
ran on a synthesized input
bioinfomachinelearning/gcdm-sbdd/process_bindingmoad.py
code served (permissive licence) · get_code("a05fbc2a9d5ee01c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tempfile
import os

# Create a temporary CSV file with BindingMOAD label file format
temp_dir = tempfile.mkdtemp()
csv_path = os.path.join(temp_dir, "every.csv")

# Write a sample label file matching the expected format
with open(csv_path, 'w') as f:
    # EC number line
    f.write("EC1.1.1.1,,,,,,,,,\n")
    # Protein line (row[2] has the PDB ID)
    f.write(",,1ABC,,,,,,,\n")
    # Ligand line (row[3] = ligand name, row[4] = validity, row[9] = SMILES)
    f.write(",,,ligand1,valid,,,,,CC(=O)O\n")
    f.write(",,,ligand2,invalid,,,,,C1=CC=CC=C1\n")
    # Another protein
    f.write(",,2DEF,,,,,,,\n")
    f.write(",,,ligand3,valid,,,,,CCO\n")
    # Another EC number
    f.write("EC2.2.2.2,,,,,,,,,\n")
    f.write(",,3GHI,,,,,,,\n")
    f.write(",,,ligand4,valid,,,,,CC\n")

ARGS = {
    "csv_path": csv_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
reverse_tensor Ran
ran; our contract draft was wrong, not the code
bioinfomachinelearning/gcdm-sbdd/utils.py
code served (permissive licence) · get_code("f69f2334c7ce9784")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function takes a tensor x and reverses it along dimension 0
# It uses x.size(0) to get the batch size and x[...] for indexing
# We need a simple 2D tensor with batch size 2

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rotation_matrix Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
bioinfomachinelearning/gcdm-sbdd/geometry_utils.py
code served (permissive licence) · get_code("cf32ddf40be29175")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 3, dtype family float, shape (2, 3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Construct arguments for rotation_matrix function
angle = np.array([0.5, 1.2])  # (n,) - array of 2 angles
axis = 0  # axis parameter: 0=x, 1=y, 2=z

ARGS = {
    "angle": angle,
    "axis": axis,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (2, 3, 3) — shape (2, 3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_GCP_with_custom_cfg Not yet run
unverified (harvested, no recorded run)
bioinfomachinelearning/gcpnet/src/models/components/gcpnet.py
code served (permissive licence) · get_code("b07de0bf12fe1cf0")
How it ran

no invocation recorded.

get_grad_norm Not yet run
unverified (harvested, no recorded run)
bioinfomachinelearning/gcdm-sbdd/utils.py
code served (permissive licence) · get_code("975b957da34b7db7")
How it ran

no invocation recorded.

residues_to_atoms Not yet run
unverified (harvested, no recorded run)
bioinfomachinelearning/gcdm-sbdd/utils.py
code served (permissive licence) · get_code("04268029bfb9fddc")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 7 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 0 the same month, 4 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
get_bb_transform 2 other papers
1 dated earlier, 1 later
2210.13695 (2022-10)
Structure-based Drug Design with Equivariant Diffusion Models
2605.17693 (2026-05)
Fine-tuning Pocket-Aware Diffusion Models via Denoising Policy Optimization
get_n_nodes 2 other papers
1 dated earlier, 1 later
2210.13695 (2022-10)
Structure-based Drug Design with Equivariant Diffusion Models
2605.17693 (2026-05)
Fine-tuning Pocket-Aware Diffusion Models via Denoising Policy Optimization
reverse_tensor 2 other papers
1 dated earlier, 1 later
2210.13695 (2022-10)
Structure-based Drug Design with Equivariant Diffusion Models
2608.31009 (2026-08)
Language-Informed Flow Matching for Trend-Guided Structure-Based 3D Molecular Generation
rotation_matrix 2 other papers
1 dated earlier, 1 later
2210.13695 (2022-10)
Structure-based Drug Design with Equivariant Diffusion Models
2605.17693 (2026-05)
Fine-tuning Pocket-Aware Diffusion Models via Denoising Policy Optimization
filter_and_flatten 1 other paper
1 dated earlier
2210.13695 (2022-10)
Structure-based Drug Design with Equivariant Diffusion Models
get_grad_norm 1 other paper
1 dated earlier
2210.13695 (2022-10)
Structure-based Drug Design with Equivariant Diffusion Models
read_label_file 1 other paper
1 dated earlier
2210.13695 (2022-10)
Structure-based Drug Design with Equivariant Diffusion Models

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Motivation: Generative deep learning methods have recently been proposed for generating 3D molecules using equivariant graph neural networks (GNNs) within a denoising diffusion framework. However, such methods are unable to learn important geometric properties of 3D molecules, as they adopt moleculeagnostic and non-geometric GNNs as their 3D graph denoising networks, which notably hinders their ability to generate valid large 3D molecules. Results: In this work, we address these gaps by introducing the Geometry-Complete Diffusion Model (GCDM) for 3D molecule generation, which outperforms existing 3D molecular diffusion models by significant margins across conditional and unconditional settings for the QM9 dataset and the larger GEOM-Drugs dataset, respectively. Importantly, we demonstrate that GCDM's generative denoising process enables the model to generate a significant proportion of valid and energetically-stable large molecules at the scale of GEOM-Drugs, whereas previous methods fail to do so with the features they learn. Additionally, we show that extensions of GCDM can not only effectively design 3D molecules for specific protein pockets but can be repurposed to consistently optimize the geometry and chemical composition of existing 3D molecules for molecular stability and property specificity, demonstrating new versatility of molecular diffusion models. Availability: Code and data are freely available on GitHub.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2302.04313")
get_code_for_paper("2302.04313")
have("2302.04313")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2302.04313.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2302.04313.svg)](https://syntology.ai/paper/2302.04313)

Connect an agent — have() is free.