SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.00203 · ICML · 2023

End-to-End Full-Atom Antibody Design

Yang Liu, Wenbing Huang, Xiangzhe Kong

arXiv · PDF · Open in the Atlas

Code that ran

5 of 7 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 1 has no contract verdict; 4 failures of our instrument; 2 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/2302.00203.json · badge: /api/ran/2302.00203.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
THUNLP-MT/dyMEAN canonical (classifier label)
classifier basis not recorded on the link
0 ran · 1 no recorded run
A4Bio/dyAb community (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
FunctionStatusWhere it lives
compute_frenet_frames Ran
ran on a synthesized input
A4Bio/dyAb/models/denoiser.py
code served (permissive licence) · get_code("15f3e94cbbbbb1cc")
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 4, dtype family float, shape (2, 4, 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 torch

torch.manual_seed(0)

# x: shape (batch_size, num_points, 3) - trajectory points in 3D
# mask: shape (batch_size, num_points) - binary mask indicating valid points
batch_size = 2
num_points = 4
point_dim = 3

x = torch.randn(batch_size, num_points, point_dim)
mask = torch.ones(batch_size, num_points, dtype=torch.float32)
# Ensure at least 2 points are valid in each batch for the function to work
mask[0, 3] = 0  # Mark last point as invalid for first batch
mask[1, 2:] = 0  # Mark last 2 points as invalid for second batch

eps = 1e-10

ARGS = {
    "x": x,
    "mask": mask,
    "eps": eps,
}
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 4, dtype family float, shape (2, 4, 3, 3) — shape (2, 4, 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
dihedral Ran
ran; our contract draft was wrong, not the code
A4Bio/dyAb/models/denoiser.py
code served (permissive licence) · get_code("bbaaceadb31583c7")
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, 2) 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 expects:
# p: a tensor with shape (..., 4, 3) where the last two dimensions are 4 points with 3D coordinates
# eps: a small float for numerical stability (default 1e-10)

# Create a small tensor with shape (2, 4, 3): batch of 2, 4 points each, 3D coordinates
p = torch.randn(2, 4, 3)

ARGS = {
    "p": p,
    "eps": 1e-10,
}
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, 2) — shape (2, 2) 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
distance Ran
ran; our contract draft was wrong, not the code
A4Bio/dyAb/models/denoiser.py
code served (permissive licence) · get_code("d7081d3937a74966")
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 1, dtype family float, shape (2,) 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 expects:
# - p: a tensor where p[..., 0, :] and p[..., 1, :] are accessed
#   This means p has shape (..., 2, D) where D is the feature dimension
# - eps: a small float (default 1e-10)

# Create a tensor with shape (batch=2, num_points=2, features=4)
# This allows p[..., 0, :] to get shape (2, 4) and p[..., 1, :] to get shape (2, 4)
p = torch.randn(2, 2, 4)

eps = 1e-10

ARGS = {
    'p': p,
    'eps': eps
}
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 1, dtype family float, shape (2,) — shape (2,) 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
unsorted_segment_mean Ran
ran; our contract draft was wrong, not the code
A4Bio/dyAb/models/modules/dyAb_egnn.py
pointer only (licence: MIT); the stored copy get_code serves, from ManlioWu/ESTAG, is pointer only (licence: NONE) · get_code("4a152e58abf0c317")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ManlioWu/ESTAG, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

unsorted_segment_sum Ran
ran; our contract draft was wrong, not the code
A4Bio/dyAb/models/modules/dyAb_egnn.py
pointer only (licence: MIT); the stored copy get_code serves, from ManlioWu/ESTAG, is pointer only (licence: NONE) · get_code("7a39c2530c1f1d13")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ManlioWu/ESTAG, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

coord2radial Not yet run
unverified (harvested, no recorded run)
A4Bio/dyAb/models/modules/dyAb_egnn.py
code served (permissive licence) · get_code("b27f3917d5805a85")
How it ran

no invocation recorded.

url_get Not yet run
unverified (harvested, no recorded run)
THUNLP-MT/dyMEAN/utils/network.py
code served (permissive licence) · get_code("2d15d2649443d0f6")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 5 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 5 later) and 2 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
unsorted_segment_mean 5 other papers
1 dated earlier, 4 later
2208.06073 (2022-08)
Conditional Antibody Design as 3D Equivariant Graph Translation
2405.12868 (2024-05)
Equivariant Spatio-Temporal Attentive Graph Networks to Simulate Physical Dynamics
aaai_32074 (2025 (year only))
Relation-Aware Equivariant Graph Networks for Epitope-Unknown Antibody Design and Specificity Optimization
and 2 more not named here
unsorted_segment_sum 5 other papers
1 dated earlier, 4 later
2208.06073 (2022-08)
Conditional Antibody Design as 3D Equivariant Graph Translation
2405.12868 (2024-05)
Equivariant Spatio-Temporal Attentive Graph Networks to Simulate Physical Dynamics
aaai_32074 (2025 (year only))
Relation-Aware Equivariant Graph Networks for Epitope-Unknown Antibody Design and Specificity Optimization
and 2 more not named here
url_get 3 other papers
1 dated earlier, 2 later
2208.06073 (2022-08)
Conditional Antibody Design as 3D Equivariant Graph Translation
2402.13555 (2024-02)
Full-Atom Peptide Design with Geometric Latent Diffusion
aaai_32074 (2025 (year only))
Relation-Aware Equivariant Graph Networks for Epitope-Unknown Antibody Design and Specificity Optimization
dihedral 2 other papers
2 later
2306.03117 (2023-06)
Str2Str: A Score-based Framework for Zero-shot Protein Conformation Sampling
2410.18403 (2024-10)
Structure Language Models for Protein Conformation Generation
distance 2 other papers
2 later
2306.03117 (2023-06)
Str2Str: A Score-based Framework for Zero-shot Protein Conformation Sampling
2410.18403 (2024-10)
Structure Language Models for Protein Conformation Generation

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

Antibody design is an essential yet challenging task in various domains like therapeutics and biology. There are two major defects in current learning-based methods: 1) tackling only a certain subtask of the whole antibody design pipeline, making them suboptimal or resourceintensive. 2) omitting either the framework regions or side chains, thus incapable of capturing the full-atom geometry. To address these pitfalls, we propose dynamic Multi-channel Equivariant grAph Network (dyMEAN), an end-to-end fullatom model for E(3)-equivariant antibody design given the epitope and the incomplete sequence of the antibody. Specifically, we first explore structural initialization as a knowledgeable guess of the antibody structure and then propose shadow paratope to bridge the epitope-antibody connections. Both 1D sequences and 3D structures are updated via an adaptive multi-channel equivariant encoder that is able to process protein residues of variable sizes when considering full atoms. Finally, the updated antibody is docked to the epitope via the alignment of the shadow paratope. Experiments on epitope-binding CDR-H3 design, complex structure prediction, and affinity optimization demonstrate the superiority of our endto-end framework and full-atom modeling.

For agents

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

get_harvested_code_for_paper("2302.00203")
get_code_for_paper("2302.00203")
have("2302.00203")

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

curl https://syntology.ai/api/ran/2302.00203.json

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

[![Syntology run record](https://syntology.ai/api/ran/2302.00203.svg)](https://syntology.ai/paper/2302.00203)

Connect an agent — have() is free.