OSCR

Generative cerebral vasculature visualization using spatial transcriptomic data.

Code ↔ Paper

The paper beside its authors' code: matches between them have not been computed for this paper yet.

Paper

Loaded from Europe PMC by your browser; OSCR does not store it: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 326 lines · 12 KB · MIT

  1. import os, math
  3. from typing import Tuple
  4. from torch import distributed
  5. from dataclasses import dataclass
  6. from multiprocessing import get_context
  7. from torch.utils.data import DataLoader
  8. from torch.utils.data.distributed import DistributedSampler
  10. from model import ModelConfig
  11. from model.blocks import ScaleAt
  12. from config_base import BaseConfig
  13. from utils.MBADataset import MBADataset
  14. from diffusion.resample import UniformSampler
  15. from diffusion.diffusion import space_timesteps
  16. from diffusion import SpacedDiffusionBeatGansConfig
  17. from diffusion.base import GenerativeType, LossType, ModelMeanType, ModelVarType, get_named_beta_schedule
  18. from utils.choices import OptimizerType, ModelName, ModelType, Activation, TrainMode, ManipulateMode, ManipulateLossType
  21. @dataclass
  22. class PretrainConfig(BaseConfig):
  23. name: str
  24. path: str
  26. @dataclass
  27. class TrainConfig(BaseConfig):
  28. is_tot: bool = False
  29. method: str = 'ours'
  30. dims: int = 2
  31. # random seed
  32. seed: int = 0
  33. train_mode: TrainMode = TrainMode.diffusion
  34. train_cond0_prob: float = 0
  35. train_pred_xstart_detach: bool = True
  36. train_interpolate_prob: float = 0
  37. train_interpolate_img: bool = False
  38. manipulate_mode: ManipulateMode = ManipulateMode.celebahq_all
  39. manipulate_cls: str = None
  40. manipulate_shots: int = None
  41. manipulate_loss: ManipulateLossType = ManipulateLossType.bce
  42. manipulate_znormalize: bool = False
  43. manipulate_seed: int = 0
  44. accum_batches: int = 1
  45. autoenc_mid_attn: bool = True
  46. batch_size: int = 16
  47. batch_size_eval: int = None
  48. beatgans_gen_type: GenerativeType = GenerativeType.ddim
  49. beatgans_loss_type: LossType = LossType.mse
  50. beatgans_model_mean_type: ModelMeanType = ModelMeanType.eps
  51. beatgans_model_var_type: ModelVarType = ModelVarType.fixed_large
  52. beatgans_rescale_timesteps: bool = False
  53. cfg: bool = False
  54. latent_infer_path: str = None
  55. latent_znormalize: bool = False
  56. latent_gen_type: GenerativeType = GenerativeType.ddim
  57. latent_loss_type: LossType = LossType.mse
  58. latent_model_mean_type: ModelMeanType = ModelMeanType.eps
  59. latent_model_var_type: ModelVarType = ModelVarType.fixed_large
  60. latent_rescale_timesteps: bool = False
  61. latent_T_eval: int = 1_000
  62. latent_clip_sample: bool = False
  63. latent_beta_scheduler: str = 'linear'
  64. beta_scheduler: str = 'linear'
  65. semantic_path: str = ''
  66. data_num: int = 0
  67. data_val_name: str = None
  68. diffusion_type: str = None
  69. dropout: float = 0.1
  70. ema_decay: float = 0.9999
  71. eval_num_images: int = 5_000
  72. eval_every_samples: int = 200_000
  73. eval_ema_every_samples: int = 200_000
  74. fid_use_torch: bool = True
  75. fp16: bool = False
  76. grad_clip: float = 1
  77. # initialize 32 to avoid divisble issue
  78. # caused by groupnorm
  79. rna_tpl: Tuple[int] = tuple(range(32))
  80. rna_num: int = 500
  81. stain: float = 'DAPI'
  82. # img_size: int = 64 # Deprecate
  83. image_size: int = 64
  84. input_size: int = None # expanded image size for equi
  85. gn_sz: int = 4
  86. lr: float = 0.0001
  87. optimizer: OptimizerType = OptimizerType.adam
  88. weight_decay: float = 0
  89. model_conf: ModelConfig = None
  90. model_name: ModelName = None
  91. model_type: ModelType = None
  92. full_model_path: str = ""
  93. patch_size: int = 64
  94. net_attn: Tuple[int] = None
  95. net_beatgans_attn_head: int = 1
  96. # not necessarily the same as the the number of style channels
  97. net_beatgans_embed_channels: int = 512
  98. net_resblock_updown: bool = True
  99. net_enc_use_time: bool = False
  100. net_enc_pool: str = 'adaptivenonzero'
  101. net_beatgans_gradient_checkpoint: bool = False
  102. net_beatgans_resnet_two_cond: bool = False
  103. net_beatgans_resnet_use_zero_module: bool = True
  104. net_beatgans_resnet_scale_at: ScaleAt = ScaleAt.after_norm
  105. net_beatgans_resnet_cond_channels: int = None
  106. net_ch_mult: Tuple[int] = None
  107. net_ch: int = 64
  108. net_enc_attn: Tuple[int] = None
  109. net_enc_k: int = None
  110. # number of resblocks for the encoder (half-unet)
  111. net_enc_num_res_blocks: int = 2
  112. net_enc_channel_mult: Tuple[int] = None
  113. net_enc_grad_checkpoint: bool = False
  114. net_autoenc_stochastic: bool = False
  115. net_latent_activation: Activation = Activation.silu
  116. net_latent_channel_mult: Tuple[int] = (1, 2, 4)
  117. net_latent_condition_bias: float = 0
  118. net_latent_dropout: float = 0
  119. net_latent_layers: int = None
  120. net_latent_net_last_act: Activation = Activation.none
  121. net_latent_net_type: bool = None
  122. net_latent_num_hid_channels: int = 1024
  123. net_latent_num_time_layers: int = 2
  124. net_latent_skip_layers: Tuple[int] = None
  125. net_latent_time_emb_channels: int = 64
  126. net_latent_use_norm: bool = False
  127. net_latent_time_last_act: bool = False
  128. net_num_res_blocks: int = 2
  129. # number of resblocks for the UNET
  130. net_num_input_res_blocks: int = None
  131. net_enc_num_cls: int = None
  132. num_workers: int = 4
  133. parallel: bool = False
  134. postfix: str = ''
  135. sample_size: int = 64
  136. sample_every_samples: int = 125_000
  137. save_every_samples: int = 100_000
  138. semantic_enc: bool = False
  139. style_ch: int = 512
  140. T_eval: int = 1_000
  141. T_sampler: str = 'uniform'
  142. T: int = 1_000
  143. total_samples: int = 10_000_000
  144. warmup: int = 0
  145. pretrain: PretrainConfig = None
  146. continue_from: PretrainConfig = None
  147. eval_programs: Tuple[str] = None
  148. # if present load the checkpoint from this path instead
  149. eval_path: str = None
  150. base_dir: str = 'checkpoints'
  151. use_cache_dataset: bool = False
  152. data_cache_dir: str = os.path.expanduser('~/cache')
  153. work_cache_dir: str = os.path.expanduser('~/mycache')
  154. # to be overridden
  155. name: str = 'test'
  156. output_dir: str = f'{work_cache_dir}/gen_images/{name}'
  157. data_path: str = ""
  158. use_pos: bool = True
  160. def __post_init__(self):
  161. self.batch_size_eval = self.batch_size_eval or self.batch_size
  162. # self.data_val_name = self.data_val_name or self.data_name
  164. def scale_up_gpus(self, num_gpus, num_nodes=1):
  165. self.eval_ema_every_samples *= num_gpus * num_nodes
  166. self.eval_every_samples *= num_gpus * num_nodes
  167. self.sample_every_samples *= num_gpus * num_nodes
  168. self.batch_size *= num_gpus * num_nodes
  169. self.batch_size_eval *= num_gpus * num_nodes
  170. return self
  172. @property
  173. def batch_size_effective(self):
  174. return self.batch_size * self.accum_batches
  176. @property
  177. def fid_cache(self):
  178. return f'{self.work_cache_dir}/eval_images/{self.data_name}_{self.eval_num_images}'
  180. @property
  181. def logdir(self):
  182. if self.name:
  183. return f'{self.base_dir}/{self.name}'
  184. else: return ""
  186. @property
  187. def generate_dir(self):
  188. return self.output_dir
  190. def _make_diffusion_conf(self, T=None):
  191. if self.diffusion_type == 'beatgans':
  192. if self.beatgans_gen_type == GenerativeType.ddpm:
  193. section_counts = [T]
  194. elif self.beatgans_gen_type == GenerativeType.ddim:
  195. section_counts = f'ddim{T}'
  196. else:
  197. raise NotImplementedError()
  199. return SpacedDiffusionBeatGansConfig(
  200. gen_type=self.beatgans_gen_type,
  201. model_type=self.model_type,
  202. betas=get_named_beta_schedule(self.beta_scheduler, self.T),
  203. model_mean_type=self.beatgans_model_mean_type,
  204. model_var_type=self.beatgans_model_var_type,
  205. loss_type=self.beatgans_loss_type,
  206. rescale_timesteps=self.beatgans_rescale_timesteps,
  207. use_timesteps=space_timesteps(num_timesteps=self.T,
  208. section_counts=section_counts),
  209. fp16=self.fp16,
  210. cfg=self.cfg,
  211. is_tot=self.is_tot,
  212. image_size=self.image_size,
  213. input_size=self.input_size or self.image_size,
  214. gn_sz=self.gn_sz,
  215. method=self.method,
  216. rna_tpl=self.rna_tpl
  217. )
  218. else:
  219. raise NotImplementedError()
  221. @property
  222. def model_out_channels(self):
  223. return 3
  225. def make_T_sampler(self):
  226. if self.T_sampler == 'uniform':
  227. return UniformSampler(self.T)
  228. else:
  229. raise NotImplementedError()
  231. def make_diffusion_conf(self):
  232. return self._make_diffusion_conf(self.T)
  234. def make_eval_diffusion_conf(self):
  235. return self._make_diffusion_conf(T=self.T_eval)
  237. def make_dataset(self, path=None, **kwargs):
  238. # 609882 0117 28
  239. if self.method == 'ours':
  240. coef = 2 if self.image_size == 128 else 4
  241. im_sz = coef * self.image_size
  242. use_exl = True if self.image_size == 32 else False
  243. else:
  244. coef = 2 if self.is_tot else 4
  245. im_sz = coef * self.image_size
  246. use_exl = True if self.image_size == 32 else False
  247. return MBADataset(self.data_path, self.rna_num,
  248. self.image_size//self.gn_sz,
  249. im_sz, self.gn_sz//2,
  250. snum=len(self.rna_tpl), stain=self.stain,
  251. transform=True, methd=self.method, use_exl=use_exl)
  253. def make_loader(self,
  254. dataset,
  255. collate_fn,
  256. shuffle: bool,
  257. num_worker: bool = None,
  258. drop_last: bool = True,
  259. batch_size: int = None,
  260. parallel: bool = False):
  261. if parallel and distributed.is_initialized():
  262. # drop last to make sure that there is no added special indexes
  263. sampler = DistributedSampler(dataset,
  264. shuffle=shuffle,
  265. drop_last=True)
  266. else:
  267. sampler = None
  268. return DataLoader(
  269. dataset,
  270. collate_fn=collate_fn,
  271. batch_size=batch_size or self.batch_size,
  272. sampler=sampler,
  273. shuffle=False if sampler else shuffle,
  274. num_workers=num_worker or self.num_workers,
  275. pin_memory=False,
  276. drop_last=drop_last,
  277. multiprocessing_context=get_context('fork'),
  278. )
  280. def make_model_conf(self):
  281. if self.method == 'patch-dm':
  282. from model.unet_patch_dm import BeatGANsUNetConfig
  283. self.model_type = ModelType.ddpm
  284. elif self.method == 'sinf':
  285. from model.unet_sinf import BeatGANsUNetConfig
  286. self.model_type = ModelType.ddpm
  287. elif self.method == 'ours':
  288. from model.unet_ours import BeatGANsUNetConfig
  289. self.model_type = ModelType.autoencoder
  290. elif self.method == 'ours_vis':
  291. from model.unet_attn import BeatGANsUNetConfig
  292. self.model_type = ModelType.autoencoder
  293. rchn = len(self.rna_tpl)
  294. rchn = math.ceil(rchn / 2)
  295. self.model_conf = BeatGANsUNetConfig(
  296. use_pos=self.use_pos,
  297. attention_resolutions=self.net_attn,
  298. channel_mult=self.net_ch_mult,
  299. conv_resample=True,
  300. semantic_path = self.semantic_path,
  301. data_num = self.data_num,
  302. dims=self.dims,
  303. dropout=self.dropout,
  304. embed_channels=self.net_beatgans_embed_channels,
  305. image_size=self.image_size,
  306. input_size=self.input_size or self.image_size,
  307. gn_sz=self.gn_sz,
  308. in_channels=rchn if self.stain != 'all' else rchn * 2,
  309. model_channels=self.net_ch,
  310. num_classes=None,
  311. num_head_channels=-1,
  312. num_heads_upsample=-1,
  313. num_heads=self.net_beatgans_attn_head,
  314. num_res_blocks=self.net_num_res_blocks,
  315. num_input_res_blocks=self.net_num_input_res_blocks,
  316. rna_tpl=self.rna_tpl,
  317. rna_num=self.rna_num,
  318. out_channels=rchn if self.stain != 'all' else rchn * 2,
  319. resblock_updown=self.net_resblock_updown,
  320. semantic_enc = self.semantic_enc,
  321. use_checkpoint=self.net_beatgans_gradient_checkpoint,
  322. use_new_attention_order=False,
  323. resnet_two_cond=self.net_beatgans_resnet_two_cond,
  324. resnet_use_zero_module=self.
  325. net_beatgans_resnet_use_zero_module)
  326. return self.model_conf

config.py at commit 083747d, under MIT · at the source

Overview

Authors: Ingrid Berg1, Jiqing Wu2, Viktor H. Koelzer2,3
  1. Department of Pathology and Molecular Pathology, University Hospital Zurich, University of Zurich,Zurich, Switzerland
  2. Department of Biomedical Engineering, University of Basel,Basel, Switzerland
  3. Institute of Medical Genetics and Pathology, University Hospital Basel,Basel, Switzerland
Institutions: University Hospital Zurich (Switzerland); University of Basel (Switzerland); University Hospital of Basel (Switzerland)
Journal: Scientific reports, volume 16, issue 1, article 15540
Dates: received 9 November 2025; accepted 25 March 2026; published online 31 March 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1038/s41598-026-46455-4 · PMID 41917199 · PMCID PMC13187267 · OpenAlex W7146948067
Open access: gold, a free copy (OpenAlex)
Status: code verified (how the code is found and verified)
Categories: genetics / omics (modality), mouse (organism), cellular / molecular (subfield)
Keywords: Spatial transcriptomics, Generative modeling, Cerebral vasculature, Computational biology and bioinformatics, Neuroscience, Systems biology
MeSH: Brain*, Transcriptome*, Actins, Animals, Claudin-5, Generative Artificial Intelligence, Mice, Spatial Transcriptomics (* major topic)
Topic: Single-cell and spatial transcriptomics (Molecular Biology, Biochemistry, Genetics and Molecular Biology), according to OpenAlex
Funding: Promedica Foundation
Citations: not cited yet (Europe PMC); 12 references in the paper

Abstract

The brain is sustained by an intricate vascular network that provides a continuous supply of nutrients and oxygen essential for its function. Understanding the structural variability and region-specific function of this neurovascular system is essential for interpreting neurological alterations in disease models. Leveraging the spatial mRNA-guided generative model Tera-MIND (simulating Tera-scale Mouse braINs using a patch-based and boundary-aware Diffusion model), we propose to predict the spatial organization of brain vasculature by the co-expression patterns of Cldn5 and Acta2 genes and their learned attention map at cell-level resolution. These findings demonstrate the capability of generative AI models trained on single-cell spatial transcriptomic data to capture biologically meaningful higher-order structures. They highlight the potential of such models as in silico systems - GenAI-based simulations that generate realistic representations of biological architecture from spatial molecular data - for high-throughput exploration of vascular function and its dysregulation in neurological disorders. Taken together, our approach repurposes existing spatial transcriptomics datasets to derive new spatial insights into vascular organization, without the need for additional tissue processing.

Supplementary Information: The online version contains supplementary material available at 10.1038/s41598-026-46455-4.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above.

CTPLab/Tera-MIND

License: MIT
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: 083747d0ddc3d17ccbb31c5556b9f319d375852f, 24 June 2025
Languages: Python (32), Shell (7)
Size: 52 files, 39 scripts
Software Heritage: not archived
Found in: “Data availability”
Holds: README, license file, environment (environment.yml)
Not found: CITATION.cff, tests, continuous integration, documentation
Tools: PyTorch (23 files), NumPy (15 files), pandas (7 files), Pillow (6 files), Matplotlib (4 files), Cellpose (3 files), seaborn (3 files), OpenCV (2 files), PyTorch Lightning (1 file), SciPy (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
41 files
At the source: github.com/CTPLab/Tera-MIND

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 39 scripts, each with its path and the digest of its content;
  • no match between paragraphs and code yet;
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data availability

Spatial transcriptomic and histological data (DAPI and Poly-T) for all three mouse brain atlases can be accessed via 10.35077/g.610. The processed spatial transcriptomic and histological data can be accessed via 10.35077/g.1176. The code and details on model training and testing are made available at the repository github.com/CTPLab/Tera-MIND.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 30 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 3 authors, 6 keywords, 8 MeSH terms, 1 funder, 12 references.

Cite

This paper

Berg, I., Wu, J., & Koelzer, V. H. (2026). Generative cerebral vasculature visualization using spatial transcriptomic data. Scientific reports, 16(1), 15540. https://doi.org/10.1038/s41598-026-46455-4

BibTeX

@article{berg2026generative,
author = {Berg, Ingrid and Wu, Jiqing and Koelzer, Viktor H.},
title = {{Generative cerebral vasculature visualization using spatial transcriptomic data}},
journal = {Scientific reports},
year = {2026},
month = mar,
volume = {16},
number = {1},
pages = {15540},
publisher = {Nature Publishing Group},
issn = {2045-2322},
doi = {10.1038/s41598-026-46455-4},
url = {https://doi.org/10.1038/s41598-026-46455-4},
pmid = {41917199},
pmcid = {PMC13187267}
}

RIS

TY - JOUR
AU - Berg, Ingrid
AU - Wu, Jiqing
AU - Koelzer, Viktor H.
TI - Generative cerebral vasculature visualization using spatial transcriptomic data
T2 - Scientific reports
J2 - Sci Rep
PY - 2026
DA - 2026/03/31
VL - 16
IS - 1
SP - 15540
SN - 2045-2322
PB - Nature Publishing Group
DO - 10.1038/s41598-026-46455-4
UR - https://doi.org/10.1038/s41598-026-46455-4
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41598-026-46455-4",
"type": "article-journal",
"title": "Generative cerebral vasculature visualization using spatial transcriptomic data",
"container-title": "Scientific reports",
"author": [
{
"family": "Berg",
"given": "Ingrid"
},
{
"family": "Wu",
"given": "Jiqing"
},
{
"family": "Koelzer",
"given": "Viktor H."
}
],
"container-title-short": "Sci Rep",
"volume": "16",
"issue": "1",
"page": "15540",
"DOI": "10.1038/s41598-026-46455-4",
"PMID": "41917199",
"PMCID": "PMC13187267",
"ISSN": "2045-2322",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41598-026-46455-4",
"language": "en",
"issued": {
"date-parts": [
[
2026,
3,
31
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1016/j.isci.2026.116355 [code]
Tera-MIND: Tera-scale mouse brain simulation via spatial mRNA-guided diffusion.
Journal: iScience
In common: Cellpose, PyTorch Lightning, OpenCV, 7 other tools, genetics / omics, mouse, cellular / molecular, 5 references
[2] doi:10.1038/s41467-025-64259-4 [code]
Data-driven fine-grained region discovery in the mouse brain with transformers.
Journal: Nature communications
In common: PyTorch Lightning, PyTorch, seaborn, 4 other tools, genetics / omics, mouse, 4 references
[3] doi:10.1038/s41588-024-01906-4 [code]
Custom microfluidic chip design enables cost-effective three-dimensional spatiotemporal transcriptomics with a wide field of view.
Journal: Nature genetics
In common: OpenCV, Pillow, seaborn, 4 other tools, genetics / omics, mouse, cellular / molecular, 4 references
[4] doi:10.1038/s41467-022-32848-2 [code]
An analysis modality for vascular structures combining tissue-clearing technology and topological data analysis.
Journal: Nature communications
In common: OpenCV, Pillow, PyTorch, 5 other tools, mouse, 3 references
[5] doi:10.1073/pnas.2515449122 [code]
A complete spatial map of mouse retinal ganglion cells reveals density and gene expression specializations.
Journal: Proceedings of the National Academy of Sciences of the United States of America
In common: OpenCV, Pillow, PyTorch, 3 other tools, genetics / omics, mouse, cellular / molecular, 4 references
[6] doi:10.1016/j.xgen.2024.100739 [code]
CSI-GEP: A GPU-based unsupervised machine learning approach for recovering gene expression programs in atlas-scale single-cell RNA-seq data.
Journal: Cell genomics
In common: PyTorch Lightning, Pillow, PyTorch, 5 other tools, genetics / omics, mouse, cellular / molecular, 2 references
[7] doi:10.1017/s2633903x24000114 [code]
Bridging the gap: Integrating cutting-edge techniques into biological imaging with deepImageJ.
Journal: Biological imaging
In common: Cellpose, PyTorch Lightning, OpenCV, 6 other tools
[8] doi:10.1371/journal.pcbi.1014106 [code]
A multimodal spatial atlas of transcriptomic, morphological, and electrophysiological cell type densities in the mouse brain.
Journal: PLoS computational biology
In common: OpenCV, seaborn, pandas, 3 other tools, genetics / omics, mouse, cellular / molecular, 4 references
[9] doi:10.1038/s41587-022-01455-3 [code]
Scalable in situ single-cell profiling by electrophoretic capture of mRNA using EEL FISH.
Journal: Nature biotechnology
In common: Cellpose, PyTorch Lightning, OpenCV, 5 other tools, genetics / omics, mouse, cellular / molecular
[10] doi:10.1186/s40478-025-02048-2 [code]
Copper supplementation mitigates Parkinson-like wild-type SOD1 pathology and nigrostriatal degeneration in a novel mouse model.
Journal: Acta neuropathologica communications
In common: Cellpose, OpenCV, Pillow, 6 other tools, mouse

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Add the badge to the code's README

OSCR: the badge of this paper

If you are an author or a maintainer of this paper's code, you can add this badge. It says what OSCR knows of the paper (its code, its validated tracing map) and links the code to this page. Paste one of these snippets under the README's title. Keep the two marks around it, so that the badge can be replaced in place and is never added twice. OSCR changes nothing in your repository.

Markdown
<!-- oscr-badge:start -->
[![OSCR](https://openscicode.org/badge/doi/10.1038/s41598-026-46455-4.svg)](https://openscicode.org/badge/doi/10.1038/s41598-026-46455-4)
<!-- oscr-badge:end -->
reStructuredText
.. oscr-badge:start

.. image:: https://openscicode.org/badge/doi/10.1038/s41598-026-46455-4.svg
   :target: https://openscicode.org/badge/doi/10.1038/s41598-026-46455-4
   :alt: OSCR

.. oscr-badge:end
HTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.1038/s41598-026-46455-4"><img src="https://openscicode.org/badge/doi/10.1038/s41598-026-46455-4.svg" alt="OSCR"></a>
<!-- oscr-badge:end -->

On GitHub, open the README, choose Edit (the pencil), paste the snippet under the title, then "Commit changes".

If you choose to, you can have the badge kept up to date for you, with oscr badge --write from the command line or with the GitHub Action in your repository's workflows. Every other form and option (a card, AsciiDoc, LaTeX, shields.io) is on the badge page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: code errors and mismatches between the code and the paper about this paper and its code, from readers and its authors, beside the paper and its tracing map.

Report a mismatch between the code and the paper · Report a code error (signed in)

Reproductions: reports from readers who ran the authors' code and did not get the paper's result, what they ran, with which environment, commit and data.

Report a reproduction failure (signed in)

Versions of its code: the releases of its code tied to a version of this paper (the preprint, the accepted manuscript, the version of record…), each with the tracing map versioned with it.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.