OSCR

MIF-MAPMS: Enhancing identification of myelin autoantigenic peptides in multiple sclerosis through multimodal information fusion.

Code ↔ Paper

6 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 6 matches
  1. [1] § MATERIALS AND METHODS › Hybrid representation ↔ Multimodal Feature/BiGRU_encoder..ipynb, lines 1–127 · score 0.74 · generate embeddings, BiGRU, Mol2Vec, bidirectional, backward, optimized
  2. [2] § MATERIALS AND METHODS › Hybrid representation ↔ Feature Desciption/TPC_ESM_Mol2Vec.ipynb, lines 175–244 · score 0.69 · Word2Vec, Mol2Vec, sentences, words, dimensionality, ESM
  3. [3] § MATERIALS AND METHODS › Multimodal information fusion ↔ Multimodal_Representation_Classification/Fusion_Classifier.ipynb, lines 116–118 · score 0.66 · AdamW, logits loss, BCE, hidden, dropout, MLP
  4. [4] § MATERIALS AND METHODS › Molecular fingerprint representation ↔ Multimodal Feature/MLP_encoder.ipynb, lines 53–81 · score 0.66 · PubChem, concatenate, Pharmacophore, bit, MACCS, SMILES
  5. [5] § MATERIALS AND METHODS › Molecular fingerprint representation ↔ Multimodal Feature/MLP_encoder.ipynb, lines 98–117 · score 0.54 · ReLU, MLP encoder, dropout, layer, fingerprints
  6. [6] § MATERIALS AND METHODS › Hybrid representation ↔ Multimodal Feature/BiGRU_encoder..ipynb, lines 1–127 · score 0.52 · BiGRU, backward, dropout, Adam, optimizer, encoder

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

Jupyter notebook · 152 lines · 4.1 KB · no license · 2 matches

  1. # %% [markdown]
  2. # # BiGRU
  4. # %%
  5. # =========================================
  6. # BiGRU on Mol2Vec
  7. # =========================================
  8. import random
  9. import numpy as np
  10. import torch
  11. import pandas as pd
  12. from torch.utils.data import Dataset, DataLoader
  13. import torch.nn as nn
  14. import torch.nn.functional as F
  16. # =========================================
  17. # 0. SEED
  18. # =========================================
  19. SEED = 42
  20. random.seed(SEED)
  21. np.random.seed(SEED)
  22. torch.manual_seed(SEED)
  23. torch.cuda.manual_seed_all(SEED)
  24. torch.backends.cudnn.deterministic = True
  25. torch.backends.cudnn.benchmark = False
  27. # =========================================
  28. # 1. Load Mol2Vec CSV
  29. # =========================================
  30. train_df = pd.read_csv("/kaggle/input/mapms-alt-type1/TPCESM_Train.csv")
  31. test_df = pd.read_csv("/kaggle/input/mapms-alt-type1/TPCESM_Test.csv")
  32. # สมมติ column แรกเป็น ID, column สุดท้ายเป็น label
  33. X_train = train_df.iloc[:, 1:].values.astype(np.float32)
  34. X_test = test_df.iloc[:, 1:].values.astype(np.float32)
  39. # =========================================
  40. # 2. Dataset & DataLoader (augmentation)
  41. # =========================================
  42. def random_dropout(seq, drop_prob=0.1):
  43. """Randomly drop features but keep shape."""
  44. mask = np.random.rand(*seq.shape) >= drop_prob
  45. return seq * mask
  47. class Mol2VecSimCSEDataset(Dataset):
  48. def __init__(self, X):
  49. self.X = X
  51. def __len__(self):
  52. return len(self.X)
  54. def __getitem__(self, idx):
  55. x = self.X[idx]
  56. x1 = random_dropout(x, 0.10)
  57. x2 = random_dropout(x, 0.10)
  58. return torch.tensor(x1, dtype=torch.float32), torch.tensor(x2, dtype=torch.float32)
  60. train_dataset = Mol2VecSimCSEDataset(X_train)
  61. train_loader = DataLoader(train_dataset, batch_size=64, shuffle=True, num_workers=0,
  62. worker_init_fn=lambda worker_id: np.random.seed(SEED))
  64. # =========================================
  65. # 3. BiGRU model
  66. # =========================================
  67. class Mol2VecBiGRU(nn.Module):
  68. def __init__(self, input_dim, hidden_dim=64):
  69. super().__init__()
  70. self.bigru = nn.GRU(input_dim, hidden_dim, batch_first=True, bidirectional=True)
  72. def forward(self, x):
  73. # x: (B, D) -> (B, seq_len=1, D)
  74. x = x.unsqueeze(1)
  75. out, _ = self.bigru(x)
  76. # out: (B, seq_len=1, 2*hidden_dim)
  77. pooled = out.squeeze(1)
  78. return pooled
  80. device = torch.device("cpu")
  81. model = Mol2VecBiGRU(input_dim=X_train.shape[1]).to(device)
  83. # =========================================
  84. # 4. SimCSE contrastive loss
  85. # =========================================
  86. optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
  87. temperature = 0.05
  89. def contrastive_loss(z1, z2, temperature):
  90. batch_size = z1.size(0)
  91. z1 = F.normalize(z1, dim=1)
  92. z2 = F.normalize(z2, dim=1)
  93. sim_matrix = torch.matmul(z1, z2.T) / temperature
  94. labels = torch.arange(batch_size).long().to(device)
  95. return F.cross_entropy(sim_matrix, labels)
  97. # =========================================
  98. # 5. Training
  99. # =========================================
  100. EPOCHS = 100
  101. for epoch in range(EPOCHS):
  102. model.train()
  103. total_loss = 0
  104. for x1, x2 in train_loader:
  105. x1 = x1.to(device)
  106. x2 = x2.to(device)
  107. optimizer.zero_grad()
  108. z1 = model(x1)
  109. z2 = model(x2)
  110. loss = contrastive_loss(z1, z2, temperature)
  111. loss.backward()
  112. optimizer.step()
  113. total_loss += loss.item()
  114. print(f"Epoch {epoch+1}/{EPOCHS} Loss: {total_loss/len(train_loader):.4f}")
  116. # =========================================
  117. # 6. Generate embeddings
  118. # =========================================
  119. model.eval()
  120. with torch.no_grad():
  121. train_emb = model(torch.tensor(X_train, dtype=torch.float32).to(device)).cpu().numpy()
  122. test_emb = model(torch.tensor(X_test, dtype=torch.float32).to(device)).cpu().numpy()
  124. pd.DataFrame(train_emb).to_csv("BiGRU-main_train.csv", index=True)
  125. pd.DataFrame(test_emb).to_csv("BiGRU-main_test.csv", index=True)
  127. print("Embeddings saved.")
  130. # %% [markdown]
  133. # %%
  136. # %% [markdown]
  139. # %%
  142. # %%
  145. # %%
  148. # %%
  151. # %%

BiGRU_encoder..ipynb at commit 7a73f3c, no license · at the source

Overview

Authors: Watshara Shoombuatong1, Nalini Schaduangrat1, Pramote Chumnanpuen2,3, Lawankorn Mookdarsanit4, Pakpoom Mookdarsanit5
  1. Center for Research Innovation and Biomedical Informatics, Faculty of Medical Technology, Mahidol University, Bangkok, Thailand
  2. Department of Zoology, Faculty of Science, Kasetsart University, Bangkok, Thailand
  3. KUSynBio Special Research Incubator Unit, Kasetsart University International College (KUIC), Kasetsart University, Bangkok, Thailand
  4. Computer and Digital Business Technology, Faculty of Management Science, Chandrakasem Rajabhat University, Bangkok, Thailand
  5. Computer Science and Artificial Intelligence, Faculty of Science, Chandrakasem Rajabhat University, Bangkok, Thailand
Institutions: Mahidol University (Thailand); Kasetsart University (Thailand); Chandrakasem Rajabhat University (Thailand)
Journal: Protein science : a publication of the Protein Society, volume 35, issue 8, article e70695
Dates: received 30 March 2026; accepted 15 June 2026; published online 9 July 2026; in print August 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1002/pro.70695 · PMID 42423156 · PMCID PMC13347377 · OpenAlex W7167804373
Open access: hybrid, a free copy (OpenAlex)
Status: code verified
Categories: human (organism), multiple sclerosis (population), cellular / molecular (subfield)
Methods: Machine learning
Keywords: autoantigen, bioinformatics, deep learning, multimodal, multiple sclerosis, myelin
MeSH: Autoantigens*, Multiple Sclerosis*, Myelin Sheath*, Peptides*, Deep Learning, Humans (* major topic)
Topic: vaccines and immunoinformatics approaches (Molecular Biology, Biochemistry, Genetics and Molecular Biology), according to OpenAlex
Funding: National Research Council of Thailand and Mahidol University (N42A660380)
Citations: not cited yet (Europe PMC); 62 references in the paper

Abstract

The abstract is not reproduced here: the paper's license (CC BY-NC-ND) does not allow it. Read it in the paper, at the publisher or on Europe PMC.

Repository

Its files are read in the Code ↔ Paper reader above, with 6 matches between paragraphs and lines of code.

lawankorn-m/MIF-MAPMS

License: none: the authors keep all their rights
State: the link answers, verified on 26 September 2026
Evidence: files inventoried
Commit: 7a73f3c05dbdb08a751ebfc7bf963f61f1cb93f2, 11 April 2026
Languages: Jupyter (5)
Size: 14 files, 5 scripts
Software Heritage: not archived
Found in: “Data Availability Statement”
Holds: README, 5 notebooks
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: NumPy (5 files), pandas (5 files), PyTorch (4 files), Keras (2 files), RDKit (2 files), scikit-learn (2 files), TensorFlow (2 files), Matplotlib (1 file)
Availability: 1 check, the latest on 26 September 2026: the link answers
  • 26 September 2026: the link answers
6 files
At the source: github.com/lawankorn-m/MIF-MAPMS

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;
  • 5 scripts, each with its path and the digest of its content;
  • 6 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.

Code and data availability statement

The paper has a code and data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:

Read it in the paper: doi.org/10.1002/pro.70695.

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, 5 authors, 6 keywords, 6 MeSH terms, 1 funder, 55 references.

Cite

This paper

Shoombuatong, W., Schaduangrat, N., Chumnanpuen, P., Mookdarsanit, L., & Mookdarsanit, P. (2026). MIF-MAPMS: Enhancing identification of myelin autoantigenic peptides in multiple sclerosis through multimodal information fusion. Protein science : a publication of the Protein Society, 35(8), e70695. https://doi.org/10.1002/pro.70695

BibTeX

@article{shoombuatong2026mif,
author = {Shoombuatong, Watshara and Schaduangrat, Nalini and Chumnanpuen, Pramote and Mookdarsanit, Lawankorn and Mookdarsanit, Pakpoom},
title = {{MIF-MAPMS: Enhancing identification of myelin autoantigenic peptides in multiple sclerosis through multimodal information fusion}},
journal = {Protein science : a publication of the Protein Society},
year = {2026},
month = aug,
volume = {35},
number = {8},
pages = {e70695},
publisher = {Wiley},
issn = {0961-8368},
doi = {10.1002/pro.70695},
url = {https://doi.org/10.1002/pro.70695},
pmid = {42423156},
pmcid = {PMC13347377}
}

RIS

TY - JOUR
AU - Shoombuatong, Watshara
AU - Schaduangrat, Nalini
AU - Chumnanpuen, Pramote
AU - Mookdarsanit, Lawankorn
AU - Mookdarsanit, Pakpoom
TI - MIF-MAPMS: Enhancing identification of myelin autoantigenic peptides in multiple sclerosis through multimodal information fusion
T2 - Protein science : a publication of the Protein Society
J2 - Protein Sci
PY - 2026
DA - 2026/08/01
VL - 35
IS - 8
SP - e70695
SN - 0961-8368
PB - Wiley
DO - 10.1002/pro.70695
UR - https://doi.org/10.1002/pro.70695
LA - en
ER -

CSL-JSON

{
"id": "10.1002/pro.70695",
"type": "article-journal",
"title": "MIF-MAPMS: Enhancing identification of myelin autoantigenic peptides in multiple sclerosis through multimodal information fusion",
"container-title": "Protein science : a publication of the Protein Society",
"author": [
{
"family": "Shoombuatong",
"given": "Watshara"
},
{
"family": "Schaduangrat",
"given": "Nalini"
},
{
"family": "Chumnanpuen",
"given": "Pramote"
},
{
"family": "Mookdarsanit",
"given": "Lawankorn"
},
{
"family": "Mookdarsanit",
"given": "Pakpoom"
}
],
"container-title-short": "Protein Sci",
"volume": "35",
"issue": "8",
"page": "e70695",
"DOI": "10.1002/pro.70695",
"PMID": "42423156",
"PMCID": "PMC13347377",
"ISSN": "0961-8368",
"publisher": "Wiley",
"URL": "https://doi.org/10.1002/pro.70695",
"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:10.1038/s42003-026-10957-8 [code]
Brain defence by the extracellular matrix protein Cochlin.
Journal: Communications biology
In common: RDKit, Keras, TensorFlow, 5 other tools, cellular / molecular
[2] 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: RDKit, Keras, TensorFlow, 5 other tools
[3] doi:10.34133/csbj.0184 [code]
Cross-Species Multitask Learning with Molecular and ADME Descriptors for Liver Microsomal Metabolic Stability.
Journal: Computational and structural biotechnology journal
In common: RDKit, PyTorch, scikit-learn, 3 other tools, cellular / molecular, 2 references
[4] doi:10.34133/csbj.0036 [code]
HYG-mol: An Interpretable Multimodal Hypergraph Framework for Molecular Property Prediction.
Journal: Computational and structural biotechnology journal
In common: RDKit, PyTorch, scikit-learn, 3 other tools, cellular / molecular, 2 references
[5] doi:10.1039/d6ra03343a [code]
A benchmark dataset and interpretable deep learning framework for drug-induced developmental neurotoxicity prediction.
Journal: RSC advances
In common: RDKit, Keras, TensorFlow, 4 other tools
[6] doi:10.1038/s41598-026-53415-5 [code]
Computational design and immunoinformatics validation of a T cell multi-epitope vaccine targeting glioblastoma stem cells.
Journal: Scientific reports
In common: RDKit, TensorFlow, PyTorch, 4 other tools, cellular / molecular
[7] doi:10.1038/s41586-026-10391-0 [code]
Cell-type-targeted mitochondrial transplantation rescues cell degeneration.
Journal: Nature
In common: RDKit, TensorFlow, PyTorch, 4 other tools, cellular / molecular
[8] doi:10.1021/acs.biochem.5c00596 [code]
Cargo Recognition of Nesprin-2 by the Dynein Adapter Bicaudal D2 for a Nuclear Positioning Pathway That Is Important for Brain Development.
Journal: Biochemistry
In common: RDKit, TensorFlow, PyTorch, 4 other tools, cellular / molecular
[9] doi:10.3390/ijms27156614 [code]
Candidalysin Inhibits <i>Porphyromonas gingivalis</i> Lipoprotein-Induced IL-1β Production in BV-2 Microglia via Hydrophobic Microbial Interactions.
Journal: International journal of molecular sciences
In common: RDKit, TensorFlow, PyTorch, 4 other tools, cellular / molecular
[10] doi:10.1038/s44320-026-00208-7 [code]
Interpretable deep generative ensemble learning for single-cell omics with Hydra.
Journal: Molecular systems biology
In common: Keras, TensorFlow, PyTorch, 4 other tools, cellular / molecular

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.