OSCR

Spatial-spectral fusion enables drug repositioning by capturing indirect and long-range associations in biological networks.

Code ↔ Paper

2 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 2 matches
  1. [1] § 3 Experiments › 3.2 Implementation details ↔ train_DDA.py, lines 1–62 · score 0.87 · fold cross validation, weight decay, dropout rate, attention heads, coefficient, dimension
  2. [2] § 3 Experiments › 3.3 Drug-disease link prediction ↔ train_DDA.py, lines 64–123 · score 0.63 · F1 scores, AUPR, accuracy, precision, AUC, recall

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 286 lines · 15 KB · MIT · 2 matches

  1. import timeit
  2. import argparse
  3. import numpy as np
  4. import pandas as pd
  5. import torch.optim as optim
  6. import torch
  7. import torch.nn as nn
  8. import torch.nn.functional as fn
  9. from data_preprocess import *
  10. from Metapath_Augmentation.path_aug_model import Metapath_Augmentation
  11. from Structural_Augmentation.struc_aug import Structural_Augmentation
  12. from pos_contrast import mp_pos, mp_data
  14. from model import SSF
  15. from metric import *
  16. import warnings
  18. warnings.filterwarnings('ignore')
  19. # Use GPU if available, otherwise fall back to CPU.
  20. # To select a specific GPU, run with e.g. `CUDA_VISIBLE_DEVICES=0 python train_DDA.py`.
  21. device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
  23. import random
  25. torch.manual_seed(123)
  26. random.seed(123)
  28. if __name__ == '__main__':
  30. parser = argparse.ArgumentParser()
  31. parser.add_argument('--k_fold', type=int, default=10, help='k-fold cross validation')
  32. parser.add_argument('--epochs', type=int, default=200, help='number of epochs to train')
  33. parser.add_argument('--lr', type=float, default=1e-3, help='learning rate')
  34. parser.add_argument('--weight_decay', type=float, default=1e-5, help='weight_decay')
  35. parser.add_argument('--random_seed', type=int, default=1234, help='random seed')
  36. parser.add_argument('--neighbor', type=int, default=5, help='neighbor')
  37. parser.add_argument('--negative_rate', type=float, default=1.0, help='negative_rate')
  38. parser.add_argument('--dataset', default='F-dataset', help='dataset')
  39. parser.add_argument('--dropout', default='0.2', type=float, help='general dropout rate')
  40. parser.add_argument('--augmentation_path', default=['RDR', 'RPR', 'DRD', 'DPD'], type=list,
  41. help='augmentation_path')
  42. parser.add_argument('--augmentation_intra_graph_num', default=1, type=int, help='augmentation_intra_graph_num')
  43. parser.add_argument('--augmentation_inter_graph_num', default=0, type=int, help='augmentation_inter_graph_num')
  44. parser.add_argument('--resolution', default=1000, type=int, help='resolution of graphon')
  45. parser.add_argument('--graphon_method', default='USVT', help='method of graphon estimation')
  46. parser.add_argument('--threshold_usvt', default='0.1', type=float, help='threshold of usvt')
  47. parser.add_argument('--lam_r', default='0.7', type=float, help='coefficient of mixup')
  48. parser.add_argument('--lam_d', default='0.7', type=float, help='coefficient of mixup')
  49. parser.add_argument('--n', default='2', type=int, help='n power of the adjacency matrix')
  50. parser.add_argument('--p_drug', default='30', type=int, help='threshold of drug node degree')
  51. parser.add_argument('--p_disease', default='15', type=int, help='threshold of disease node degree')
  52. parser.add_argument('--t', default='0.5', type=float, help='threshold of feature similarity')
  53. parser.add_argument('--pos_num', default='5', type=int, help='threshold of positive samples selection')
  55. parser.add_argument('--embedding_dim', default='512', type=int, help='GAE embedding dimension')
  56. parser.add_argument('--hgt_in_dim', default='64', type=int, help='input feature dimension')
  57. parser.add_argument('--hgt_out_dim', default='64', type=int, help='output feature dimension')
  59. parser.add_argument('--spectral_layer', default='1', type=int, help='number of spectral HSGE layers')
  60. parser.add_argument('--spectral_head', default='4', type=int, help='number of attention heads for spectral models')
  61. parser.add_argument('--spectral_sge_layers', default='2', type=int, help='number of spectral SGE layers')
  62. parser.add_argument('--spectral_dropout', default='0.2', type=float, help='dropout rate for spectral models')
  64. parser.add_argument('--freq_enhance_ratio', default='0.1', type=float,
  65. help='frequency enhancement ratio for spectral models')
  66. parser.add_argument('--freq_low_pass_ratio', default='0.25', type=float,
  67. help='frequency low pass ratio for spectral models')
  68. parser.add_argument('--freq_min_components', default='32', type=int,
  69. help='minimum frequency components for spectral models')
  70. parser.add_argument('--freq_max_components', default='64', type=int,
  71. help='maximum frequency components for spectral models')
  73. parser.add_argument('--gwn_time', default='1.0', type=float, help='time parameter for graph wave network')
  74. parser.add_argument('--gwn_dt', default='1.0', type=float, help='time step for graph wave network')
  75. parser.add_argument('--gwn_dropout', default='0.2', type=float, help='dropout rate for graph wave network')
  76. parser.add_argument('--gwn_laplacian', default='sym', type=str, help='laplacian type for graph wave network')
  77. parser.add_argument('--gwn_method', default='exp', type=str, help='method for graph wave network')
  78. parser.add_argument('--gwn_init_residual', default=False, type=bool,
  79. help='use initial residual for graph wave network')
  81. parser.add_argument('--tau_cross', default='0.8', type=float, help='tau in cross-view contrastive loss')
  82. parser.add_argument('--tau_mp', default='0.8', type=float, help='tau_mp in intra-view contrastive loss')
  83. parser.add_argument('--tau_sc', default='0.8', type=float, help='tau_sc in intra-view contrastive loss')
  84. parser.add_argument('--lam_cross', default='0.5', type=float, help='lam in cross-view contrastive loss')
  85. parser.add_argument('--lam_intra', default='0.5', type=float, help='lam in intra-view contrastive loss')
  86. parser.add_argument('--loss_rate', default='0.5', type=float,
  87. help='loss rate of unsupervised learning and training')
  88. parser.add_argument('--n_neg', default='20', type=int, help='number of negative candidates')
  89. parser.add_argument('--pool', default='mean', help='method of pooling in negative sampling')
  91. args = parser.parse_args()
  92. args.data_dir = 'data/' + args.dataset + '/'
  94. data = get_data(args)
  95. args.drug_number = data['drug_number']
  96. args.disease_number = data['disease_number']
  97. args.protein_number = data['protein_number']
  99. data = data_processing(data, args)
  100. data = k_fold(data, args)
  102. drdr_graph, didi_graph, prpr_graph, data = dgl_similarity_graph(data, args)
  105. drdr_graph = drdr_graph.to(device)
  106. didi_graph = didi_graph.to(device)
  107. drug_feature = torch.FloatTensor(data['drug_gae']).to(device)
  108. disease_feature = torch.FloatTensor(data['disease_gae']).to(device)
  109. protein_feature = torch.FloatTensor(data['protein_gae']).to(device)
  111. all_sample = torch.tensor(data['all_drdi']).long()
  113. start = timeit.default_timer()
  115. cross_entropy = nn.CrossEntropyLoss()
  117. Metric = (
  118. 'Epoch\t\tTime\t\tLoss_con\t\tLoss_cls\t\tAUC\t\tAUPR\t\tAccuracy\t\tPrecision\t\tRecall\t\tF1-score\t\tMcc')
  119. AUCs, AUPRs = [], []
  121. precisions, recalls, accuracys, f1_scores = [], [], [], []
  123. print('Dataset:', args.dataset)
  125. for i in range(args.k_fold):
  127. print('fold:', i)
  128. print(Metric)
  130. model = SSF(args)
  131. model = model.to(device)
  133. optimizer = optim.Adam(model.parameters(), lr=args.lr, weight_decay=args.weight_decay)
  134. scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer=optimizer, mode='max', factor=0.8, patience=100,
  135. verbose=True, min_lr=1e-3)
  137. text_file = '{}fold/{}/data_train.csv'.format(args.data_dir, i)
  138. train_samples = np.array(pd.read_csv(text_file).iloc[:, 1:])
  139. train_X = train_samples[:, 0:2]
  140. train_X_pos = [row[:-1] for row in train_samples if row[-1] == 1]
  141. train_X_neg = [row[:-1] for row in train_samples if row[-1] == 0]
  142. train_Y = train_samples[:, -1]
  144. text_file1 = '{}fold/{}/data_test.csv'.format(args.data_dir, i)
  145. test_samples = np.array(pd.read_csv(text_file1).iloc[:, 1:])
  146. test_X = test_samples[:, 0:2]
  147. test_X_pos = [row[:-1] for row in test_samples if row[-1] == 1]
  148. test_X_neg = [row[:-1] for row in test_samples if row[-1] == 0]
  149. test_Y = test_samples[:, -1]
  151. data['X_train'][i] = train_X
  152. data['X_train_p'][i] = np.array([item.tolist() for item in train_X_pos])
  153. data['X_train_n'][i] = np.array([item.tolist() for item in train_X_neg])
  154. data['Y_train'][i] = train_Y
  156. data['X_test'][i] = test_X
  157. data['X_test_p'][i] = np.array([item.tolist() for item in test_X_pos])
  158. data['X_test_n'][i] = np.array([item.tolist() for item in test_X_neg])
  159. data['Y_test'][i] = test_Y
  161. best_auc, best_aupr, best_accuracy, best_precision, best_recall, best_f1, best_mcc = 0, 0, 0, 0, 0, 0, 0
  162. X_train = torch.LongTensor(data['X_train'][i]).to(device)
  163. X_train_p = torch.LongTensor(data['X_train_p'][i]).to(device)
  164. X_train_n = torch.LongTensor(data['X_train_n'][i]).to(device)
  165. Y_train = torch.LongTensor(data['Y_train'][i]).to(device)
  166. X_test = torch.LongTensor(data['X_test'][i]).to(device)
  167. X_test_p = torch.LongTensor(data['X_test_p'][i]).to(device)
  168. X_test_n = torch.LongTensor(data['X_test_n'][i]).to(device)
  169. Y_test = data['Y_test'][i].flatten()
  171. drdipr_graph, meta_paths, edge_types, data = dgl_heterograph(data, data['X_train_p'][i], args)
  173. g_s_drug, feature_graph_drug = Structural_Augmentation(data, args, 'drug', drdipr_graph, edge_types)
  174. g_s_disease, feature_graph_disease = Structural_Augmentation(data, args, 'disease', drdipr_graph, edge_types)
  175. g_m_drug, g_m_disease, g_adj_r, g_adj_d = Metapath_Augmentation(drdipr_graph, meta_paths, edge_types, args)
  177. rdr, drd = mp_data(data['X_train_p'][i], data, args)
  178. pos_r, pos_d = mp_pos(rdr, drd, feature_graph_drug, feature_graph_disease, args)
  180. drdipr_graph = drdipr_graph.to(device)
  181. g_s_drug = g_s_drug.to(device)
  182. g_s_disease = g_s_disease.to(device)
  183. g_m_drug = g_m_drug.to(device)
  184. g_m_disease = g_m_disease.to(device)
  186. pos_r = torch.IntTensor(pos_r.toarray()).to(device)
  187. pos_d = torch.IntTensor(pos_d.toarray()).to(device)
  188. drug_nei2 = searching_2hop(args, data, 'drug', drdipr_graph, drdr_graph, rdr, drd, data['X_train_p'][i])
  189. disease_nei2 = searching_2hop(args, data, 'disease', drdipr_graph, didi_graph, rdr, drd,
  190. data['X_train_p'][i][:, [1, 0]])
  192. for epoch in range(args.epochs):
  194. neg_candidates_train_r = sampling_2hop(args, 'drug', drug_nei2, data['X_train_p'][i], data['X_train_n'][i])
  195. neg_candidates_test_r = sampling_2hop(args, 'drug', drug_nei2, data['X_test_p'][i], data['X_test_n'][i])
  196. neg_candidates_train_d = sampling_2hop(args, 'disease', disease_nei2, data['X_train_p'][i][:, [1, 0]],
  197. data['X_train_n'][i][:, [1, 0]])
  198. neg_candidates_test_d = sampling_2hop(args, 'disease', disease_nei2, data['X_test_p'][i][:, [1, 0]],
  199. data['X_test_n'][i][:, [1, 0]])
  200. model.train()
  201. l_con, train_output, _, _ = model(g_m_drug, g_m_disease, pos_r, pos_d,
  202. drug_feature, disease_feature, protein_feature, X_train_p, X_train_n,
  203. neg_candidates_train_r, neg_candidates_train_d, data['X_train_n'][i])
  205. l_train = cross_entropy(train_output, torch.flatten(Y_train))
  206. l_all = args.loss_rate * l_con + (1 - args.loss_rate) * l_train
  207. optimizer.zero_grad()
  208. l_all.backward()
  209. optimizer.step()
  211. l_con = l_con.detach().cpu().numpy()
  212. l_train = l_train.detach().cpu().numpy()
  214. with torch.no_grad():
  215. model.eval()
  216. _, test_score, _, _ = model(g_m_drug, g_m_disease,
  217. pos_r, pos_d, drug_feature, disease_feature, protein_feature, X_test_p,
  218. X_test_n, neg_candidates_test_r, neg_candidates_test_d, data['X_test_n'][i])
  220. test_prob = fn.softmax(test_score, dim=-1)
  221. test_score = torch.argmax(test_score, dim=-1)
  223. predict_result = test_prob
  224. predict_score = np.zeros((len(test_score), 1))
  226. test_prob = test_prob[:, 1]
  227. test_prob = test_prob.cpu().numpy()
  228. test_score = test_score.cpu().numpy()
  230. AUC, AUPR, accuracy, precision, recall, f1, mcc = get_metric(Y_test, test_score, test_prob)
  232. scheduler.step(AUC)
  234. end = timeit.default_timer()
  235. time = end - start
  236. show = [epoch + 1, round(time, 2), l_con, l_train, round(AUC, 5), round(AUPR, 5), round(accuracy, 5),
  237. round(precision, 5), round(recall, 5), round(f1, 5), round(mcc, 5)]
  238. print('\t\t'.join(map(str, show)))
  239. if AUC > best_auc:
  240. best_epoch = epoch + 1
  241. best_auc = AUC
  242. best_aupr, best_accuracy, best_precision, best_recall, best_f1, best_mcc = AUPR, accuracy, precision, recall, f1, mcc
  243. print('AUC improved at epoch ', best_epoch, ';\tbest_auc:', best_auc)
  245. AUCs.append(AUC)
  246. AUPRs.append(AUPR)
  247. accuracys.append(accuracy)
  248. precisions.append(precision)
  249. recalls.append(recall)
  250. f1_scores.append(f1)
  252. print("***************************************************************")
  253. print('AUC:', AUCs)
  254. AUC_mean = np.mean(AUCs)
  255. AUC_std = np.std(AUCs)
  256. print('Mean AUC:', AUC_mean, '(', AUC_std, ')')
  258. print("***************************************************************")
  259. print('AUPR:', AUPRs)
  260. AUPR_mean = np.mean(AUPRs)
  261. AUPR_std = np.std(AUPRs)
  262. print('Mean AUPR:', AUPR_mean, '(', AUPR_std, ')')
  264. print("***************************************************************")
  265. print('accuracy:', accuracys)
  266. accuracy_mean = np.mean(accuracys)
  267. accuracy_std = np.std(accuracys)
  268. print('Mean accuracy:', accuracy_mean, '(', accuracy_std, ')')
  270. print("***************************************************************")
  271. print('precision:', precisions)
  272. precision_mean = np.mean(precisions)
  273. precision_std = np.std(precisions)
  274. print('Mean precision:', precision_mean, '(', precision_std, ')')
  276. print("***************************************************************")
  277. print('recall:', recalls)
  278. recall_mean = np.mean(recalls)
  279. recall_std = np.std(recalls)
  280. print('Mean recall:', recall_mean, '(', recall_std, ')')
  282. print("***************************************************************")
  283. print('f1:', f1_scores)
  284. f1_mean = np.mean(f1_scores)
  285. f1_std = np.std(f1_scores)
  286. print('Mean f1:', f1_mean, '(', f1_std, ')')

train_DDA.py at commit 7a76031, under MIT · at the source

Overview

Authors: Xiaobo Zhu1, Lei Wang2, Runzhou Tang1, Zhi-An Huang3, Yu-an Huang4, Feng Tan5, Lun Hu1, Zhuhong You4, Pengwei Hu1
ORCID iDs: Lei Wang, Zhi-An Huang, Lun Hu, Pengwei Hu
  1. Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi 810011, China
  2. China University of Mining and Technology, Xuzhou, China
  3. Research Office, City University of Hong Kong (Dongguan), Dongguan 523000, China
  4. School of Computing, Northwestern Polytechnical University, Xi’an 710129, China
  5. AI and Quantum Lab, Darmstadt, Germany
Institutions: Chinese Academy of Sciences (China); Xinjiang Technical Institute of Physics & Chemistry (China); China University of Mining and Technology (China); City University of Hong Kong (Dongguan) (China); Northwestern Polytechnical University (China)
Journal: Bioinformatics (Oxford, England), volume 42, issue 8, article btag553
Dates: received 17 May 2026; accepted 15 July 2026; published online 10 August 2026; in print August 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1093/bioinformatics/btag553 · PMID 42573507 · PMCID PMC13461371 · OpenAlex W7202078187
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: computational modeling (no new data) (modality), human (organism), Alzheimer's / dementia (population), Parkinson's (population)
Methods: Machine learning
MeSH: Computational Biology*, Drug Repositioning*, Algorithms, Alzheimer Disease, Graph Neural Networks, Humans, Molecular Docking Simulation, Parkinson Disease (* major topic)
Journal subjects: Systems Biology
Topic: Bioinformatics and Genomic Networks (Molecular Biology, Biochemistry, Genetics and Molecular Biology), according to OpenAlex
Funding: National Natural Science Foundation of China (62302495, 62373348); Natural Science Foundation of Xinjiang Uygur Autonomous Region (2023D01E15); Xinjiang Tianchi Talents Program (E33B9401); Tianshan Talent Training Program (2023TSYCLJ0021)
Citations: not cited yet (Europe PMC); 31 references in the paper

Abstract

Motivation: Drug repositioning accelerates clinical translation by identifying new therapeutic indications for approved drugs. However, therapeutic associations in biomolecular networks often exist indirectly, through transitive chains and long-range mechanisms, rather than as directly observed links. Shallow methods are confined to direct similarity and miss such indirect associations, whereas deep graph neural networks suffer from over-smoothing and lose discriminative power in highly connected networks.

Results: We propose a spatial-spectral collaborative framework. In the spatial domain, a wave-evolution process propagates similarity from local to global, capturing multi-hop transitive associations while preserving discriminative representations. In the spectral domain, network-specific spectral transforms model global connectivity for long-range dependencies over homogeneous similarity and heterogeneous drug-protein-disease networks, with the two views aligned by contrastive learning. On three benchmarks the method outperforms state-of-the-art baselines on most evaluation metrics; case studies on Alzheimer’s and Parkinson’s disease and molecular docking confirm its ability to recover non-explicit therapeutic associations.

Availability: The source code and data are available at https://github.com/Juniper-cola/BIO_SSF.

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

Repository

Its files are read in the Code ↔ Paper reader above, with 2 matches between paragraphs and lines of code.

Juniper-cola/BIO_SSF

License: MIT
State: the link answers, verified on 26 September 2026
Evidence: files inventoried
Commit: 7a7603103f1c9fdc0140ef4aa3eaf630aaf589d2, 20 July 2026
Languages: Python (12)
Size: 18 files, 12 scripts
Software Heritage: not archived
Found in: “Data availability”
Holds: README, license file, environment (requirements.txt)
Not found: CITATION.cff, tests, continuous integration, documentation
Tools: PyTorch (11 files), NumPy (10 files), pandas (4 files), scikit-learn (4 files), SciPy (3 files), OpenCV (2 files), Matplotlib (1 file), NetworkX (1 file), PyTorch Geometric (1 file), scikit-image (1 file)
Availability: 1 check, the latest on 26 September 2026: the link answers
  • 26 September 2026: the link answers
14 files
At the source: github.com/Juniper-cola/BIO_SSF

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;
  • 12 scripts, each with its path and the digest of its content;
  • 2 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • 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.

Availability

The source code and data are available at https://github.com/Juniper-cola/BIO_SSF.

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

Data availability

The source code and datasets used in this study are available at https://github.com/Juniper-cola/BIO_SSF.

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, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 9 authors, 8 MeSH terms, 4 funders, 27 references.

Cite

This paper

Zhu, X., Wang, L., Tang, R., Huang, Z.-A., Huang, Y.-a., Tan, F., Hu, L., You, Z., & Hu, P. (2026). Spatial-spectral fusion enables drug repositioning by capturing indirect and long-range associations in biological networks. Bioinformatics (Oxford, England), 42(8), btag553. https://doi.org/10.1093/bioinformatics/btag553

BibTeX

@article{zhu2026spatial,
author = {Zhu, Xiaobo and Wang, Lei and Tang, Runzhou and Huang, Zhi-An and Huang, Yu-an and Tan, Feng and Hu, Lun and You, Zhuhong and Hu, Pengwei},
title = {{Spatial-spectral fusion enables drug repositioning by capturing indirect and long-range associations in biological networks}},
journal = {Bioinformatics (Oxford, England)},
year = {2026},
month = aug,
volume = {42},
number = {8},
pages = {btag553},
publisher = {Oxford University Press},
issn = {1367-4803},
doi = {10.1093/bioinformatics/btag553},
url = {https://doi.org/10.1093/bioinformatics/btag553},
pmid = {42573507},
pmcid = {PMC13461371}
}

RIS

TY - JOUR
AU - Zhu, Xiaobo
AU - Wang, Lei
AU - Tang, Runzhou
AU - Huang, Zhi-An
AU - Huang, Yu-an
AU - Tan, Feng
AU - Hu, Lun
AU - You, Zhuhong
AU - Hu, Pengwei
TI - Spatial-spectral fusion enables drug repositioning by capturing indirect and long-range associations in biological networks
T2 - Bioinformatics (Oxford, England)
J2 - Bioinformatics
PY - 2026
DA - 2026/08/01
VL - 42
IS - 8
SP - btag553
SN - 1367-4803
PB - Oxford University Press
DO - 10.1093/bioinformatics/btag553
UR - https://doi.org/10.1093/bioinformatics/btag553
LA - en
ER -

CSL-JSON

{
"id": "10.1093/bioinformatics/btag553",
"type": "article-journal",
"title": "Spatial-spectral fusion enables drug repositioning by capturing indirect and long-range associations in biological networks",
"container-title": "Bioinformatics (Oxford, England)",
"author": [
{
"family": "Zhu",
"given": "Xiaobo"
},
{
"family": "Wang",
"given": "Lei"
},
{
"family": "Tang",
"given": "Runzhou"
},
{
"family": "Huang",
"given": "Zhi-An"
},
{
"family": "Huang",
"given": "Yu-an"
},
{
"family": "Tan",
"given": "Feng"
},
{
"family": "Hu",
"given": "Lun"
},
{
"family": "You",
"given": "Zhuhong"
},
{
"family": "Hu",
"given": "Pengwei"
}
],
"container-title-short": "Bioinformatics",
"volume": "42",
"issue": "8",
"page": "btag553",
"DOI": "10.1093/bioinformatics/btag553",
"PMID": "42573507",
"PMCID": "PMC13461371",
"ISSN": "1367-4803",
"publisher": "Oxford University Press",
"URL": "https://doi.org/10.1093/bioinformatics/btag553",
"language": "en",
"issued": {
"date-parts": [
[
2026,
8,
1
]
]
}
}

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:10.1093/bib/bbag118 [code]
Drug screening for α-synuclein aggregation inhibitors via multimodal graph neural network.
Journal: Briefings in bioinformatics
In common: PyTorch Geometric, NetworkX, PyTorch, 5 other tools, computational modeling (no new data), Parkinson's
[2] doi:10.1038/s41398-026-03965-z [code]
Disentangling individual heterogeneity reveals robust network and molecular signatures of major depressive disorder with suicidal ideation.
Journal: Translational psychiatry
In common: PyTorch Geometric, NetworkX, OpenCV, 6 other tools
[3] doi:10.1002/hbm.70469 [code]
VarCoNet: A Variability-Aware Self-Supervised Framework for Functional Connectome Extraction From Resting-State fMRI.
Journal: Human brain mapping
In common: PyTorch Geometric, NetworkX, OpenCV, 6 other tools
[4] doi:10.1093/nar/gkag706 [code]
scDifformer: diffusion-based post-training for virtual cell modeling across large-scale single-cell data.
Journal: Nucleic acids research
In common: PyTorch Geometric, NetworkX, scikit-image, 6 other tools
[5] doi:10.1038/s41467-026-76045-x [code]
A manufacturability-informed topology framework for AI-guided design of fibrous network materials.
Journal: Nature communications
In common: PyTorch Geometric, NetworkX, OpenCV, 6 other tools
[6] doi:10.21203/rs.3.rs-9676637/v1 [code]
A Comprehensive Benchmarking of Spatial Deconvolution and Domain Detection Methods across Diverse Tissues and Spatial Transcriptomic Technologies
Journal: Research Square (preprint)
In common: PyTorch Geometric, OpenCV, scikit-image, 6 other tools
[7] doi:10.1038/s41467-026-71759-4 [code]
CellNiche represents cellular microenvironments in atlas-scale spatial omics data with contrastive learning.
Journal: Nature communications
In common: PyTorch Geometric, NetworkX, PyTorch, 5 other tools, computational modeling (no new data)
[8] doi:10.1038/s42003-026-10957-8 [code]
Brain defence by the extracellular matrix protein Cochlin.
Journal: Communications biology
In common: NetworkX, OpenCV, scikit-image, 6 other tools
[9] doi:10.1371/journal.pcbi.1014571 [code]
SynAPSeg: A novel dataset and image analysis framework for deep learning-based synapse detection and quantification.
Journal: PLoS computational biology
In common: NetworkX, OpenCV, scikit-image, 6 other tools
[10] doi:10.1126/sciadv.aed3650 [code]
Truthful visualizations for mass spectrometry imaging enable high-spatial-resolution interactive <i>m/z</i> mapping and exploration.
Journal: Science advances
In common: NetworkX, OpenCV, scikit-image, 6 other tools

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.

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: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

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