OSCR

Predicting progression of Alzheimer's disease using blood-based multi-omics data.

Code ↔ Paper

9 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 9 matches
  1. [1] § Methods › DNA methylation data pre-processing ↔ methylation_minfi.R, lines 42 to 92 · score 0.95 · getBeta, preprocessQuantile, read.metharray.exp, rgSet, probes failing, quantile normalization · Report a mismatch
  2. [2] § Methods › DNA methylation data pre-processing ↔ feature_selection_gene_cpg.py, lines 1 to 41 · score 0.76 · Log2 fold change, log2fc, pMCI, sMCI, CpG, Methylation · Report a mismatch
  3. [3] § Methods › Lipid data pre-processing ↔ feature_selection_type.py, lines 4 to 11 · score 0.71 · SelectBySingleFeaturePerformance, SelectFromModel, Feature selection, RFE · Report a mismatch
  4. [4] § Results › Model performance across integration strategies ↔ Early_Late_integration.py, lines 114 to 171 · score 0.64 · L1 regularized logistic, neural network, random forest, metrics, NN, regression · Report a mismatch
  5. [5] § Methods › Model training and feature importance analysis ↔ feature_selection_gene_cpg.py, lines 1 to 41 · score 0.58 · pMCI, sMCI, Python, CpG, binary, selections · Report a mismatch
  6. [6] § Results › Model performance across integration strategies ↔ Early_Late_integration.py, lines 1 to 43 · score 0.58 · single model, XGBoost, concatenated, NN, SVM, RF · Report a mismatch
  7. [7] § Methods › Lipid data pre-processing ↔ feature_selection_gene_cpg.py, lines 108 to 161 · score 0.55 · Feature selection, Mutual Information, MI, filtered, zero · Report a mismatch
  8. [8] § Results › Model performance across integration strategies ↔ Early_Late_integration.py, lines 1 to 43 · score 0.54 · Neural network, XGBoost, SVM, RF, classifier, L1 · Report a mismatch
  9. [9] § Methods › Gene expression data pre-processing ↔ feature_selection_gene_cpg.py, lines 61 to 72 · score 0.53 · avoid division, log2FC, zero, gene · Report a mismatch

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 · 165 lines · 6.1 KB · CC-BY-NC-4.0 · 4 matches

  1. #!/usr/bin/env python3
  2. """Gene / CpG feature-selection (public repo version).
  4. This script keeps a runnable example for GitHub using the bundled toy dataset,
  5. while preserving the key paper-style step of selecting features by effect size:
  7. selected_* = log2fc[abs(log2fc) >= 0.25]
  9. Behavior
  10. - Loads a CSV with a binary `Phenotype` column.
  11. - Detects gene expression columns by suffix `_gene` and CpG methylation columns
  12. by suffix `_cpg` (if present).
  13. - For each feature set (genes / CpGs), computes class-wise means and log2 fold
  14. change (pMCI vs sMCI), then filters by |log2FC| >= threshold (default 0.25).
  15. - Ranks the filtered features with mutual information and writes top-N lists.
  16. - Always writes an overall ranking as a fallback.
  18. Notes
  19. - Toy data may not contain `_cpg` columns; the CpG branch will be skipped.
  20. - For methylation beta values, log2FC is a simple effect-size proxy; you may
  21. prefer mean difference instead. This implementation keeps the original step
  22. you requested for parity.
  24. Usage
  25. python code/feature_selection/feature_selection_gene_cpg.py \
  26. --csv data/toy/toy_complete_data.csv \
  27. --out out/feature_selection \
  28. --top 50 \
  29. --log2fc-thresh 0.25
  30. """
  32. from __future__ import annotations
  34. import argparse
  35. from pathlib import Path
  37. import numpy as np
  38. import pandas as pd
  39. from sklearn.feature_selection import mutual_info_classif
  41. LABEL_COL = "Phenotype"
  44. def load_xy(csv_path: Path) -> tuple[pd.DataFrame, np.ndarray]:
  45. df = pd.read_csv(csv_path)
  46. if LABEL_COL not in df.columns:
  47. raise ValueError(f"Expected a '{LABEL_COL}' column.")
  48. y_raw = df[LABEL_COL].astype(int)
  49. uniq = sorted(y_raw.unique().tolist())
  50. if uniq == [1, 2]:
  51. # sMCI=1, pMCI=2 -> 0/1
  52. y = (y_raw - 1).astype(int).to_numpy()
  53. elif len(uniq) == 2:
  54. y = y_raw.map({uniq[0]: 0, uniq[1]: 1}).astype(int).to_numpy()
  55. else:
  56. raise ValueError(f"Expected binary labels, got: {uniq}")
  57. X = df.drop(columns=[LABEL_COL])
  58. return X, y
  61. def compute_log2fc(X: pd.DataFrame, y: np.ndarray, eps: float = 1e-9) -> pd.Series:
  62. """Compute log2 fold change of feature means: class1 / class0.
  64. log2FC = log2( mean(y=1) / mean(y=0) )
  65. eps avoids division by zero.
  66. """
  67. X0 = X.loc[y == 0]
  68. X1 = X.loc[y == 1]
  69. m0 = X0.mean(axis=0)
  70. m1 = X1.mean(axis=0)
  71. log2fc = np.log2((m1 + eps) / (m0 + eps))
  72. return pd.Series(log2fc, index=X.columns, name="log2fc")
  75. def rank_and_save(
  76. X: pd.DataFrame,
  77. y: np.ndarray,
  78. out_csv: Path,
  79. top: int,
  80. seed: int = 42,
  81. ) -> pd.DataFrame:
  82. if X.shape[1] == 0:
  83. raise ValueError("No features to rank (empty feature matrix).")
  84. mi = mutual_info_classif(X.to_numpy(), y, random_state=seed)
  85. res = (
  86. pd.DataFrame({"feature": X.columns, "score": mi})
  87. .sort_values("score", ascending=False)
  88. .head(top)
  89. )
  90. out_csv.parent.mkdir(parents=True, exist_ok=True)
  91. res.to_csv(out_csv, index=False)
  92. return res
  95. def log2fc_filter(
  96. X: pd.DataFrame,
  97. y: np.ndarray,
  98. thresh: float,
  99. eps: float = 1e-9,
  100. ) -> tuple[pd.DataFrame, pd.Series]:
  101. """Filter features by absolute log2FC threshold."""
  102. log2fc = compute_log2fc(X, y, eps=eps)
  103. selected = log2fc[log2fc.abs() >= thresh] # <-- requested key step
  104. X_sel = X[selected.index]
  105. return X_sel, log2fc
  108. def main() -> None:
  109. ap = argparse.ArgumentParser()
  110. ap.add_argument("--csv", type=str, default=None, help="CSV path (default: toy dataset)")
  111. ap.add_argument("--out", type=str, default=None, help="Output dir (default: repo_root/out/feature_selection)")
  112. ap.add_argument("--top", type=int, default=50, help="Top N features to write per ranking")
  113. ap.add_argument("--seed", type=int, default=42, help="Random seed for MI")
  114. ap.add_argument("--log2fc-thresh", type=float, default=0.25, help="Abs(log2FC) threshold")
  115. ap.add_argument("--eps", type=float, default=1e-9, help="Small constant to avoid division by zero")
  116. args = ap.parse_args()
  118. here = Path(__file__).resolve()
  119. repo_root = here.parents[2]
  120. csv_path = Path(args.csv).expanduser().resolve() if args.csv else (repo_root / "data" / "toy" / "toy_complete_data.csv")
  121. out_dir = Path(args.out).expanduser().resolve() if args.out else (repo_root / "out" / "feature_selection")
  122. out_dir.mkdir(parents=True, exist_ok=True)
  124. X, y = load_xy(csv_path)
  126. gene_cols = [c for c in X.columns if c.endswith("_gene")]
  127. cpg_cols = [c for c in X.columns if c.endswith("_cpg")]
  129. # Genes
  130. if gene_cols:
  131. Xg = X[gene_cols].apply(pd.to_numeric, errors="coerce").fillna(0.0)
  132. Xg_sel, log2fc_g = log2fc_filter(Xg, y, thresh=args.log2fc_thresh, eps=args.eps)
  134. # Save full log2FC table + selected list
  135. log2fc_g.to_csv(out_dir / "gene_log2fc.csv", header=True)
  136. log2fc_g[log2fc_g.abs() >= args.log2fc_thresh].to_csv(out_dir / "gene_selected_by_log2fc.csv", header=True)
  138. if Xg_sel.shape[1] > 0:
  139. rank_and_save(Xg_sel, y, out_dir / "gene_selected_log2fc_then_mi_top.csv", top=args.top, seed=args.seed)
  140. else:
  141. # If threshold selects nothing, still provide MI ranking for all genes
  142. rank_and_save(Xg, y, out_dir / "gene_mutual_information_top.csv", top=args.top, seed=args.seed)
  144. # CpGs
  145. if cpg_cols:
  146. Xc = X[cpg_cols].apply(pd.to_numeric, errors="coerce").fillna(0.0)
  147. Xc_sel, log2fc_c = log2fc_filter(Xc, y, thresh=args.log2fc_thresh, eps=args.eps)
  149. log2fc_c.to_csv(out_dir / "cpg_log2fc.csv", header=True)
  150. log2fc_c[log2fc_c.abs() >= args.log2fc_thresh].to_csv(out_dir / "cpg_selected_by_log2fc.csv", header=True)
  152. if Xc_sel.shape[1] > 0:
  153. rank_and_save(Xc_sel, y, out_dir / "cpg_selected_log2fc_then_mi_top.csv", top=args.top, seed=args.seed)
  154. else:
  155. rank_and_save(Xc, y, out_dir / "cpg_mutual_information_top.csv", top=args.top, seed=args.seed)
  157. # Fallback: overall MI ranking (all features)
  158. X_all = X.apply(pd.to_numeric, errors="coerce").fillna(0.0)
  159. rank_and_save(X_all, y, out_dir / "all_features_mutual_information_top.csv", top=args.top, seed=args.seed)
  161. print(f"Wrote outputs to: {out_dir}")
  164. if __name__ == "__main__":
  165. main()

feature_selection_gene_cpg.py at commit 18c7d2f, under CC-BY-NC-4.0 · at the source

Overview

Authors: Yashu Vashishath1,2,3, Bizhan Alipour Pijani1,2,3, Neha Goud Baddam1,2,3, Fahad Saeed4, Serdar Bozdag1,2,3,5, for the Alzheimer’s Disease Neuroimaging Initiative
ORCID iDs: Yashu Vashishath, Fahad Saeed, Serdar Bozdag
  1. Department of Computer Science and Engineering, University of North Texas, Denton, TX 76203, United States
  2. Center for Computational Life Sciences, University of North Texas, Denton, TX 76203, United States
  3. BioDiscovery Institute, University of North Texas, Denton, TX 76203, United States
  4. School of Computing and Information Sciences, Florida International University, Miami, FL 33199, United States
  5. Department of Mathematics, University of North Texas, Denton, TX 76203, United States
Institutions: University of North Texas (United States); Florida International University (United States)
Journal: Bioinformatics advances, volume 6, issue 1, article vbag085
Dates: received 12 December 2025; accepted 28 February 2026; published online 28 March 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1093/bioadv/vbag085 · PMID 42109582 · PMCID PMC13157226 · OpenAlex W7142678317
Open access: gold, a free copy (OpenAlex)
Status: code verified (how the code is found and verified)
Categories: genetics / omics (modality), human (organism), Alzheimer's / dementia (population), methods / tools (subfield)
Methods: Spectral & time-frequency, Statistics, Machine learning, Connectivity
Topic: Genetic Associations and Epidemiology (Genetics, Biochemistry, Genetics and Molecular Biology), according to OpenAlex
Citations: not cited yet (Europe PMC); 35 references in the paper

Abstract

Motivation: Early prediction of Alzheimer’s disease (AD) progression from mild cognitive impairment (MCI) remains a major challenge, particularly when relying on non-invasive biomarkers. Identifying individuals with progressive MCI (pMCI) before conversion to AD could improve intervention strategies and clinical management. In this study, we developed a machine learning (ML) framework integrating blood-based multi-omics and demographic data to distinguish pMCI from stable MCI (sMCI).

Results: We trained ML models using combinations of single nucleotide polymorphism (SNP), DNA methylation, gene expression, lipid, and bile acid metabolite data using both early and late data integration strategies. Late integration consistently outperformed early integration, with the L1-regularized logistic regression model achieving the highest F1 score of 90.7% when combining SNP and lipid data. Feature interpretability analyses using LIME and SHAP identified reproducible biomarkers across omic layers, including SNPs in MYH11, FOXP1, MAPK10, and SYN3; methylation changes in CDX2 and DHX58; and altered ceramide lipid CER.D19.1.24.0 levels, all previously associated with AD-related pathology. These findings demonstrate that combining multi-omics and demographic data can improve early prediction of AD progression and support the feasibility of blood-based, interpretable biomarkers for precision diagnostics.

Availability and implementation: The data analyzed in this study were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (https://adni.loni.usc.edu/) and are available to qualified researchers upon application. The code used in this study, together with instructions and a toy dataset, is available at: https://github.com/bozdaglab/AD_sMCI_vs_pMCI.

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 9 matches between paragraphs and lines of code.

bozdaglab/AD_sMCI_vs_pMCI

License: CC-BY-NC-4.0
State: the link answers, verified on 30 September 2026
Evidence: files inventoried
Commit: 18c7d2fc3ee61d0272db41b81e96e3bf5f35262c, 26 February 2026
Languages: Python (5), Shell (1), R (1)
Size: 13 files, 7 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README, license file
Not found: CITATION.cff, environment file, tests, continuous integration, documentation
Tools: NumPy (4 files), pandas (4 files), scikit-learn (4 files), SHAP (2 files), XGBoost (1 file)
Availability: 1 check, the latest on 30 September 2026: the link answers
  • 30 September 2026: the link answers
9 files
At the source: github.com/bozdaglab/AD_sMCI_vs_pMCI

Availability and implementation

The data analyzed in this study were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (https://adni.loni.usc.edu/) and are available to qualified researchers upon application. The code used in this study, together with instructions and a toy dataset, is available at: https://github.com/bozdaglab/AD_sMCI_vs_pMCI.

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

Code availability

The code used in this study, along with instructions and a toy dataset, is available at https://github.com/bozdaglab/AD_sMCI_vs_pMCI.

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

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;
  • 7 scripts, each with its path and the digest of its content;
  • 9 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.

Data availability

The data analyzed in this study were obtained from the ADNI database (https://adni.loni.usc.edu/) and are available to qualified researchers upon application.

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, 6 authors, 26 references.

Cite

This paper

Vashishath, Y., Pijani, B. A., Goud Baddam, N., Saeed, F., Bozdag, S., & for the Alzheimer’s Disease Neuroimaging Initiative. (2026). Predicting progression of Alzheimer's disease using blood-based multi-omics data. Bioinformatics advances, 6(1), vbag085. https://doi.org/10.1093/bioadv/vbag085

BibTeX

@article{vashishath2026predicting,
author = {Vashishath, Yashu and Pijani, Bizhan Alipour and Goud Baddam, Neha and Saeed, Fahad and Bozdag, Serdar and {for the Alzheimer’s Disease Neuroimaging Initiative}},
title = {{Predicting progression of Alzheimer's disease using blood-based multi-omics data}},
journal = {Bioinformatics advances},
year = {2026},
month = mar,
volume = {6},
number = {1},
pages = {vbag085},
publisher = {Oxford University Press},
issn = {2635-0041},
doi = {10.1093/bioadv/vbag085},
url = {https://doi.org/10.1093/bioadv/vbag085},
pmid = {42109582},
pmcid = {PMC13157226}
}

RIS

TY - JOUR
AU - Vashishath, Yashu
AU - Pijani, Bizhan Alipour
AU - Goud Baddam, Neha
AU - Saeed, Fahad
AU - Bozdag, Serdar
AU - for the Alzheimer’s Disease Neuroimaging Initiative
TI - Predicting progression of Alzheimer's disease using blood-based multi-omics data
T2 - Bioinformatics advances
J2 - Bioinform Adv
PY - 2026
DA - 2026/03/28
VL - 6
IS - 1
SP - vbag085
SN - 2635-0041
PB - Oxford University Press
DO - 10.1093/bioadv/vbag085
UR - https://doi.org/10.1093/bioadv/vbag085
LA - en
ER -

CSL-JSON

{
"id": "10.1093/bioadv/vbag085",
"type": "article-journal",
"title": "Predicting progression of Alzheimer's disease using blood-based multi-omics data",
"container-title": "Bioinformatics advances",
"author": [
{
"family": "Vashishath",
"given": "Yashu"
},
{
"family": "Pijani",
"given": "Bizhan Alipour"
},
{
"family": "Goud Baddam",
"given": "Neha"
},
{
"family": "Saeed",
"given": "Fahad"
},
{
"family": "Bozdag",
"given": "Serdar"
},
{
"literal": "for the Alzheimer’s Disease Neuroimaging Initiative"
}
],
"container-title-short": "Bioinform Adv",
"volume": "6",
"issue": "1",
"page": "vbag085",
"DOI": "10.1093/bioadv/vbag085",
"PMID": "42109582",
"PMCID": "PMC13157226",
"ISSN": "2635-0041",
"publisher": "Oxford University Press",
"URL": "https://doi.org/10.1093/bioadv/vbag085",
"language": "en",
"issued": {
"date-parts": [
[
2026,
3,
28
]
]
}
}

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/bioinformatics/btad249 [code]
PPAD: a deep learning architecture to predict progression of Alzheimer's disease.
Journal: Bioinformatics (Oxford, England)
In common: scikit-learn, pandas, NumPy, methods / tools, Alzheimer's / dementia, 2 authors
[2] doi:10.1038/s41597-022-01695-7 [code]
An analysis-ready and quality controlled resource for pediatric brain white-matter research.
Journal: Scientific data
In common: SHAP, XGBoost, scikit-learn, 2 other tools, methods / tools, 2 references
[3] doi:10.1002/alz.14600 [code]
Machine learning prediction of tau-PET in Alzheimer's disease using plasma, MRI, and clinical data.
Journal: Alzheimer's & dementia : the journal of the Alzheimer's Association
In common: SHAP, XGBoost, scikit-learn, 2 other tools, Alzheimer's / dementia, 2 references
[4] doi:10.7554/elife.81067 [code]
Biological brain age prediction using machine learning on structural neuroimaging data: Multi-cohort validation against biomarkers of Alzheimer's disease and neurodegeneration stratified by sex.
Journal: eLife
In common: SHAP, XGBoost, scikit-learn, 2 other tools, Alzheimer's / dementia, 1 reference
[5] doi:10.1093/bioinformatics/btae264 [code]
TA-RNN: an attention-based time-aware recurrent neural network architecture for electronic health records.
Journal: Bioinformatics (Oxford, England)
In common: scikit-learn, pandas, NumPy, Alzheimer's / dementia, author Serdar Bozdag
[6] doi:10.1016/j.mex.2025.103574 [code]
MLSPred-bench: Transforming electroencephalography (EEG) datasets into machine learning-ready epileptic seizure prediction benchmarks.
Journal: MethodsX
In common: scikit-learn, pandas, NumPy, methods / tools, author Fahad Saeed
[7] doi:10.1038/s41598-026-48613-0 [code]
An snRNA-seq aging clock for the fruit fly head sheds light on sex-biased aging.
Journal: Scientific reports
In common: SHAP, XGBoost, scikit-learn, 2 other tools, genetics / omics, 2 references
[8] doi:10.1038/s43856-025-01083-3 [code]
A deep learning model to predict glioma recurrence using integrated genomic and clinical data.
Journal: Communications medicine
In common: SHAP, XGBoost, scikit-learn, 2 other tools, 2 references
[9] doi:10.1016/j.bpj.2026.06.005 [code]
Curvature-based machine-learning method for automated segmentation of dendritic spines.
Journal: Biophysical journal
In common: SHAP, XGBoost, scikit-learn, 2 other tools, 2 references
[10] doi:10.1038/s43587-026-01149-4 [code]
The blood metabolome of brain health in midlife and influences of genes, microbiome and exposome.
Journal: Nature aging
In common: scikit-learn, pandas, NumPy, Alzheimer's / dementia, genetics / omics, 3 references

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/10.1093/bioadv/vbag085.svg)](https://openscicode.org/badge/doi/10.1093/bioadv/vbag085)
<!-- oscr-badge:end -->
reStructuredText
.. oscr-badge:start

.. image:: https://openscicode.org/badge/doi/10.1093/bioadv/vbag085.svg
   :target: https://openscicode.org/badge/doi/10.1093/bioadv/vbag085
   :alt: OSCR

.. oscr-badge:end
HTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.1093/bioadv/vbag085"><img src="https://openscicode.org/badge/doi/10.1093/bioadv/vbag085.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.