Coupled cross-sectional and longitudinal non-negative matrix factorization reveals dominant brain aging trajectories in 48,949 individuals.
One block of code here has a permanent identifier (oscr:b:…), listed with the matches below: citable for good, with a page of its own. Select a match and press Cite to copy it. What a block identifier is.
The 1 match
- [1] § Methods › Normative modeling for estimating the cross-sectional deviation map (C-map) ↔ ccl_nmf/aa.py, lines 150 to 201 · score 0.64 · cyclical learning rate, Adam, cycle, gradient, optimizer, batch · block
oscr:b:Z26TEHN8STDRP-M· 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 · 403 lines · 15 KB · MIT · 1 match
- from __future__ import annotations
- import argparse
- import random
- from dataclasses import dataclass
- from pathlib import Path
- from typing import Tuple
- import numpy as np
- import pandas as pd
- import tensorflow as tf
- from tensorflow import keras
- import joblib
- from sklearn.model_selection import StratifiedShuffleSplit
- from sklearn.preprocessing import StandardScaler
- # Project-local imports (must exist in your repo)
- from models import (
- make_encoder_model_v1,
- make_decoder_model_v1,
- make_discriminator_model_v1,
- )
- # ------------------------- Minimal determinism helpers -------------------------
- def set_global_seeds(seed: int) -> None:
- """Seed Python, NumPy, and TensorFlow RNGs."""
- random.seed(seed)
- np.random.seed(seed)
- tf.keras.utils.set_random_seed(seed)
- def set_tf_verbosity(quiet: bool = True) -> None:
- if not quiet:
- return
- import logging, warnings # noqa
- warnings.simplefilter(action="ignore", category=FutureWarning)
- warnings.simplefilter(action="ignore", category=Warning)
- tf.get_logger().setLevel(logging.ERROR)
- def feature_cols_from_csv(path: Path) -> list[str]:
- cols = pd.read_csv(path, nrows=0).columns.tolist()
- return sorted(c for c in cols if c != "participant_id")
- def load_and_merge(features_csv: Path, covariates_csv: Path) -> pd.DataFrame:
- features = pd.read_csv(features_csv)
- covars = pd.read_csv(covariates_csv)
- df = covars.merge(features, on="participant_id").reset_index(drop=True)
- return df
- def stratify_key(df: pd.DataFrame, n_bins: int) -> pd.Series:
- """
- Equal-width age bins over the observed [min(Age), max(Age)].
- Assumes 'Age' has no NaNs (drop earlier if needed).
- """
- ages = df["Age"].to_numpy()
- min_age = float(np.nanmin(ages))
- max_age = float(np.nanmax(ages))
- if not np.isfinite(min_age) or not np.isfinite(max_age):
- raise ValueError("Age contains no finite values for stratification.")
- if max_age <= min_age:
- return pd.Series(["bin0"] * len(df), index=df.index)
- # Equal-width bins
- edges = np.linspace(min_age, max_age, num=n_bins + 1)
- labels = [f"bin{i}" for i in range(n_bins)]
- out = pd.cut(df["Age"], bins=edges, labels=labels,
- include_lowest=True, right=False, duplicates="drop")
- # If duplicates dropped (rare), fallback: label NaNs as last bin
- out = out.astype("object")
- mask_nan = pd.isna(out.values)
- if mask_nan.any():
- out[mask_nan] = labels[-1]
- return pd.Series(out)
- # ------------------------- Config -------------------------
- @dataclass
- class TrainConfig:
- project_root: Path
- features_csv: Path
- covariates_csv: Path
- outdir: Path
- # Splits
- valheldout_size: float = 0.35
- heldout_frac_within_val: float = 0.40
- age_bins: int = 5 # choose 4 or 5
- # Training
- batch_size: int = 200
- epochs: int = 1000
- patience: int = 50
- z_dim: int = 20
- h_dim: Tuple[int, int] = (110, 110)
- seed: int = 0
- # CLR
- base_lr: float = 1e-4
- max_lr: float = 5e-3
- gamma: float = 0.98
- step_size: int | None = None
- # ------------------------- Train -------------------------
- def train(cfg: TrainConfig) -> Path:
- set_global_seeds(cfg.seed)
- set_tf_verbosity(quiet=True)
- # Load & prepare
- df = load_and_merge(cfg.features_csv, cfg.covariates_csv)
- feature_cols = feature_cols_from_csv(cfg.features_csv)
- # Drop rows with missing Age before stratification
- if df["Age"].isna().any():
- dropped = int(df["Age"].isna().sum())
- print(f"[INFO] Dropping {dropped} rows with missing Age before stratified split.")
- df = df[df["Age"].notna()].reset_index(drop=True)
- df_split = pd.DataFrame({"participant_id": df["participant_id"], "Age": df["Age"]})
- strat = stratify_key(df_split, cfg.age_bins)
- # Split TRAIN vs (VAL+HELDOUT)
- sss = StratifiedShuffleSplit(
- n_splits=1, test_size=cfg.valheldout_size, random_state=cfg.seed)
- train_idx, valheld_idx = next(sss.split(df_split["participant_id"], strat))
- df_train = df.iloc[train_idx].copy()
- df_valheld = df.iloc[valheld_idx].copy()
- # Split (VAL+HELDOUT) into VAL and HELDOUT
- sss2 = StratifiedShuffleSplit(
- n_splits=1, test_size=cfg.heldout_frac_within_val, random_state=cfg.seed)
- strat2 = stratify_key(df_valheld[["Age"]].assign(participant_id=df_valheld["participant_id"]), cfg.age_bins)
- val_idx, held_idx = next(sss2.split(df_valheld["participant_id"], strat2))
- # Smaller subset becomes validation, remainder heldout (consistent naming)
- df_val = df_valheld.iloc[held_idx].copy()
- df_held = df_valheld.iloc[val_idx].copy()
- # Save split manifests
- cfg.outdir.mkdir(parents=True, exist_ok=True)
- df_train[["participant_id"]].to_csv(cfg.outdir / "train_participants.csv", index=False)
- df_val[["participant_id"]].to_csv(cfg.outdir / "val_participants.csv", index=False)
- df_held[["participant_id"]].to_csv(cfg.outdir / "heldout_participants.csv", index=False)
- df_train[["participant_id"] + feature_cols].to_csv(cfg.outdir / "train_features.csv", index=False)
- df_val [["participant_id"] + feature_cols].to_csv(cfg.outdir / "val_features.csv", index=False)
- df_held [["participant_id"] + feature_cols].to_csv(cfg.outdir / "heldout_features.csv", index=False)
- # Scale
- scaler = StandardScaler()
- X_train = scaler.fit_transform(df_train[feature_cols].values).astype("float32")
- X_val = scaler.transform(df_val[feature_cols].values).astype("float32")
- # Build models
- n_features = X_train.shape[1]
- encoder = make_encoder_model_v1(n_features, list(cfg.h_dim), cfg.z_dim)
- decoder = make_decoder_model_v1(cfg.z_dim, n_features, list(cfg.h_dim)[::-1])
- discriminator = make_discriminator_model_v1(cfg.z_dim, list(cfg.h_dim)[::-1])
- # Losses
- bce = tf.keras.losses.BinaryCrossentropy(from_logits=True)
- mse = tf.keras.losses.MeanSquaredError()
- def d_loss(real_out, fake_out):
- return bce(tf.ones_like(real_out), real_out) + bce(tf.zeros_like(fake_out), fake_out)
- def g_loss(fake_out):
- return bce(tf.ones_like(fake_out), fake_out)
- # Optimizers (CLR-updated per step)
- ae_opt = keras.optimizers.Adam(learning_rate=cfg.base_lr)
- d_opt = keras.optimizers.Adam(learning_rate=cfg.base_lr)
- g_opt = keras.optimizers.Adam(learning_rate=cfg.base_lr)
- # tf.data pipeline (deterministic ordering and seeded shuffle)
- ds = tf.data.Dataset.from_tensor_slices(X_train)
- ds = ds.shuffle(buffer_size=len(df_train), seed=cfg.seed, reshuffle_each_iteration=True).batch(cfg.batch_size)
- options = tf.data.Options()
- options.experimental_deterministic = True
- ds = ds.with_options(options)
- # ---- Cyclical Learning Rate (CLR) setup ----
- n_samples = X_train.shape[0]
- step_size = cfg.step_size if cfg.step_size is not None else int(2 * np.ceil(n_samples / cfg.batch_size))
- def scale_fn(cycle):
- return cfg.gamma ** cycle
- global_step = 0
- @tf.function
- def train_step(xb, ae_lr, d_lr, g_lr):
- ae_opt.learning_rate.assign(tf.cast(ae_lr, tf.float32))
- d_opt.learning_rate.assign(tf.cast(d_lr, tf.float32))
- g_opt.learning_rate.assign(tf.cast(g_lr, tf.float32))
- # Autoencoder
- with tf.GradientTape() as t_ae:
- z = encoder(xb, training=True)
- xhat = decoder(z, training=True)
- ae_loss = mse(xb, xhat)
- ae_grads = t_ae.gradient(ae_loss, encoder.trainable_variables + decoder.trainable_variables)
- ae_opt.apply_gradients(zip(ae_grads, encoder.trainable_variables + decoder.trainable_variables))
- # Discriminator
- with tf.GradientTape() as t_d:
- z_real = tf.random.normal([tf.shape(xb)[0], cfg.z_dim])
- z_fake = encoder(xb, training=True)
- d_real = discriminator(z_real, training=True)
- d_fake = discriminator(z_fake, training=True)
- loss_d = d_loss(d_real, d_fake)
- d_grads = t_d.gradient(loss_d, discriminator.trainable_variables)
- d_opt.apply_gradients(zip(d_grads, discriminator.trainable_variables))
- # Generator/Encoder
- with tf.GradientTape() as t_g:
- z_fake = encoder(xb, training=True)
- d_fake = discriminator(z_fake, training=True)
- loss_g = g_loss(d_fake)
- g_grads = t_g.gradient(loss_g, encoder.trainable_variables)
- g_opt.apply_gradients(zip(g_grads, encoder.trainable_variables))
- return ae_loss, loss_d, loss_g
- # Training loop with EarlyStopping on val AE loss
- best_val = np.inf
- best_epoch = -1
- history = {"ae_loss": [], "d_loss": [], "g_loss": [], "val_ae_loss": []}
- for epoch in range(cfg.epochs):
- ae_epoch, d_epoch, g_epoch = [], [], []
- for xb in ds:
- global_step += 1
- cycle = np.floor(1 + global_step / (2 * step_size))
- x_lr = np.abs(global_step / step_size - 2 * cycle + 1)
- clr = cfg.base_lr + (cfg.max_lr - cfg.base_lr) * max(0.0, 1.0 - x_lr) * scale_fn(cycle)
- clr32 = tf.constant(clr, dtype=tf.float32)
- ae_l, d_l, g_l = train_step(xb, clr32, clr32, clr32)
- ae_epoch.append(float(ae_l.numpy()))
- d_epoch.append(float(d_l.numpy()))
- g_epoch.append(float(g_l.numpy()))
- # Validation AE loss
- z_val = encoder(X_val, training=False)
- xhat_val = decoder(z_val, training=False)
- val_loss = float(mse(X_val, xhat_val).numpy())
- history["ae_loss"].append(np.mean(ae_epoch))
- history["d_loss"].append(np.mean(d_epoch))
- history["g_loss"].append(np.mean(g_epoch))
- history["val_ae_loss"].append(val_loss)
- print(f"Epoch {epoch:04d} | AE {np.mean(ae_epoch):.4f} | D {np.mean(d_epoch):.4f} "
- f"| G {np.mean(g_epoch):.4f} | VAL {val_loss:.4f} | BEST {best_val:.4f} @ {best_epoch}")
- # Save best
- if val_loss < best_val:
- best_val = val_loss
- best_epoch = epoch
- encoder.save(cfg.outdir / "best_encoder.h5")
- decoder.save(cfg.outdir / "best_decoder.h5")
- discriminator.save(cfg.outdir / "best_discriminator.h5")
- if epoch - best_epoch >= cfg.patience:
- break
- # Save final artifacts
- encoder.save(cfg.outdir / "encoder.h5")
- decoder.save(cfg.outdir / "decoder.h5")
- discriminator.save(cfg.outdir / "discriminator.h5")
- joblib.dump(scaler, cfg.outdir / "scaler.joblib")
- pd.DataFrame(history).to_csv(cfg.outdir / "training_history.csv", index=False)
- print(f"Best VAL AE loss: {best_val:.6f} @ epoch {best_epoch}")
- return cfg.outdir
- # ------------------------- Inference -------------------------
- def infer(models_dir: Path, features_csv: Path, covariates_csv: Path, dataset_name: str = "test", seed: int = 0) -> Path:
- set_global_seeds(seed)
- set_tf_verbosity(quiet=True)
- outdir = Path(models_dir) / dataset_name
- outdir.mkdir(parents=True, exist_ok=True)
- df = load_and_merge(features_csv, covariates_csv)
- feature_cols = feature_cols_from_csv(features_csv)
- # Optional safety check:
- missing = [c for c in feature_cols if c not in df.columns]
- if missing:
- raise ValueError(f"Missing expected feature columns: {missing}")
- X = df[feature_cols].values.astype("float32")
- encoder = keras.models.load_model(Path(models_dir) / "best_encoder.h5", compile=False)
- decoder = keras.models.load_model(Path(models_dir) / "best_decoder.h5", compile=False)
- scaler = joblib.load(Path(models_dir) / "scaler.joblib")
- Xn = scaler.transform(X).astype("float32")
- Z = encoder(Xn, training=False)
- Xhat = decoder(Z, training=False)
- # normalized inputs
- norm_df = pd.DataFrame({"participant_id": df["participant_id"]})
- for i, c in enumerate(feature_cols):
- norm_df[c] = Xn[:, i]
- norm_df.to_csv(outdir / "normalized.csv", index=False)
- # reconstructions
- recon_df = pd.DataFrame({"participant_id": df["participant_id"]})
- xhat_np = Xhat.numpy()
- for i, c in enumerate(feature_cols):
- recon_df[c] = xhat_np[:, i]
- recon_df.to_csv(outdir / "reconstruction.csv", index=False)
- # latent codes
- enc_df = pd.DataFrame({"participant_id": df["participant_id"]})
- z_np = Z.numpy()
- for i in range(z_np.shape[1]):
- enc_df[f"z{i:02d}"] = z_np[:, i]
- enc_df.to_csv(outdir / "encoded.csv", index=False)
- # reconstruction error (per-subject MSE)
- rec_err = np.mean((Xn - xhat_np) ** 2, axis=1)
- pd.DataFrame({
- "participant_id": df["participant_id"],
- "Reconstruction error": rec_err}).to_csv(outdir / "reconstruction_error.csv", index=False)
- print(f"Wrote outputs to {outdir}")
- return outdir
- # ------------------------- CLI -------------------------
- def build_parser() -> argparse.ArgumentParser:
- p = argparse.ArgumentParser(description="AAE pipeline (train + infer) — NO-SEX condition (minimal deterministic)")
- sub = p.add_subparsers(dest="cmd", required=True)
- pt = sub.add_parser("train", help="Train models")
- pt.add_argument("--features_csv", required=True, type=Path)
- pt.add_argument("--covariates_csv", required=True, type=Path)
- pt.add_argument("--outdir", required=True, type=Path)
- pt.add_argument("--z_dim", type=int, default=20)
- pt.add_argument("--h_dim", type=int, nargs=2, default=[110, 110])
- pt.add_argument("--epochs", type=int, default=1000)
- pt.add_argument("--patience", type=int, default=50)
- pt.add_argument("--batch_size", type=int, default=200)
- pt.add_argument("--seed", type=int, default=0)
- pt.add_argument("--valheldout_size", type=float, default=0.35)
- pt.add_argument("--heldout_frac_within_val", type=float, default=0.40)
- # CLR
- pt.add_argument("--base_lr", type=float, default=1e-4)
- pt.add_argument("--max_lr", type=float, default=5e-3)
- pt.add_argument("--gamma", type=float, default=0.98)
- pt.add_argument("--step_size", type=int, default=0, help="0=auto (2 * ceil(N/batch))")
- # Age-bin choice
- pt.add_argument("--age_bins", type=int, choices=[4, 5], default=5)
- pi = sub.add_parser("infer", help="Run inference with trained models")
- pi.add_argument("--models_dir", required=True, type=Path)
- pi.add_argument("--features_csv", required=True, type=Path)
- pi.add_argument("--covariates_csv", required=True, type=Path)
- pi.add_argument("--name", default="test", help="Output subfolder name (e.g., val/heldout/test)")
- pi.add_argument("--seed", type=int, default=0, help="Seed for deterministic inference")
- return p
- def main():
- parser = build_parser()
- args = parser.parse_args()
- if args.cmd == "train":
- cfg = TrainConfig(
- project_root=Path.cwd(),
- features_csv=args.features_csv,
- covariates_csv=args.covariates_csv,
- outdir=args.outdir,
- z_dim=args.z_dim,
- h_dim=(args.h_dim[0], args.h_dim[1]),
- epochs=args.epochs,
- patience=args.patience,
- batch_size=args.batch_size,
- seed=args.seed,
- valheldout_size=args.valheldout_size,
- heldout_frac_within_val=args.heldout_frac_within_val,
- base_lr=args.base_lr,
- max_lr=args.max_lr,
- gamma=args.gamma,
- step_size=(None if args.step_size == 0 else args.step_size),
- age_bins=args.age_bins)
- train(cfg)
- elif args.cmd == "infer":
- infer(
- models_dir=args.models_dir,
- features_csv=args.features_csv,
- covariates_csv=args.covariates_csv,
- dataset_name=args.name,
- seed=args.seed)
- if __name__ == "__main__":
- main()
aa.py at commit 76d8801, under MIT · at the source
Overview
- AI2D Center for AI and Data Science for Integrated Diagnostics, University of Pennsylvania,Philadelphia, PA USA
- School of Electrical and Computer Engineering, National Technical University of Athens,Athens, Greece
- Department of Radiology, University of Pennsylvania,Philadelphia, PA USA
- Laboratory of Behavioral Neuroscience, National Institute on Aging,Baltimore, MD USA
- Department of Biostatistics, Epidemiology, & Informatics, University of Pennsylvania,Philadelphia, PA USA
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.
Repositories
Its files are read in the Code ↔ Paper reader above, with 1 match between paragraphs and lines of code.
cbica/ccl_nmf_prediction
ec021b5aafc9ddf7ed08f0bca1ce4b673e3bfc4f, 28 April 2025Availability: 1 check, the latest on 30 September 2026: the link answers
- 30 September 2026: the link answers
7 files
- ccl_nmf_prediction/
__init__.py , Python, 1 line - ccl_nmf_prediction/
__main__.py , Python, 76 lines - ccl_nmf_prediction/
utils.py , Python, 58 lines - setup.py, Python, 45 lines
- test/
cmd/ , Shell, 7 linesrun_test.sh - LICENSE, License, 21 lines
- README.md, Text, 31 lines
IoannaSkampardoni/CCL-NMF
76d8801e3f9bf821e8a46ebf5f4c6d05349cc487, 19 September 2025Availability: 1 check, the latest on 30 September 2026: the link answers
- 30 September 2026: the link answers
8 files
- ccl_nmf/
C_map_calc.py , Python, 78 lines - ccl_nmf/
L_map_calc.py , Python, 162 lines - ccl_nmf/
aa.py , Python, 403 lines, 1 match - ccl_nmf/
jointNMF.py , Python, 171 lines - ccl_nmf/
models.py , Python, 41 lines - ccl_nmf/
run_jointNMF.py , Python, 124 lines - LICENSE, License, 21 lines
- README.md, Text, 130 lines
Code availability statement
The paper has a code availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:
- it points to the authors' code: cbica/
ccl_nmf_prediction , IoannaSkampardoni/CCL-NMF
Read it in the paper: doi.org/10.1038/s41467-026-72091-7.
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:
- 2 repositories of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
- 11 scripts, each with its path and the digest of its content;
- 1 match 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
Datasets cited
- zenodo:18850194, at Zenodo; found in “Data availability”
Data availability statement
The paper has a data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:
- it points to a dataset: Zenodo 18850194
- it says that the data are available on request
Read it in the paper: doi.org/10.1038/s41467-026-72091-7.
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, 14 authors, 5 keywords, 13 MeSH terms, 1 funder, 58 references.
Cite
This paper
Skampardoni, I., Erus, G., Nasrallah, I. M., Yang, Z., Duggan, M. R., Walker, K. A., Getka, A., Baik, K., Melhem, R., Govindarajan, S. T., Resnick, S. M., Shou, H., Nikita, K., & Davatzikos, C. (2026). Coupled cross-sectional and longitudinal non-negative matrix factorization reveals dominant brain aging trajectories in 48,949 individuals. Nature communications, 17(1), 5709. https://
BibTeX
@article{skampardoni2026
author = {Skampardoni, Ioanna and Erus, Guray and Nasrallah, Ilya M. and Yang, Zhijian and Duggan, Michael R. and Walker, Keenan A. and Getka, Alexander and Baik, Kyunglok and Melhem, Randa and Govindarajan, Sindhuja T. and Resnick, Susan M. and Shou, Haochang and Nikita, Konstantina and Davatzikos, Christos},
title = {{Coupled cross-sectional and longitudinal non-negative matrix factorization reveals dominant brain aging trajectories in 48,949 individuals}},
journal = {Nature communications},
year = {2026},
month = apr,
volume = {17},
number = {1},
pages = {5709},
publisher = {Nature Portfolio},
issn = {2041-1723},
doi = {10.1038/
url = {https://
pmid = {42034902},
pmcid = {PMC13319776}
}
RIS
TY - JOUR
AU - Skampardoni, Ioanna
AU - Erus, Guray
AU - Nasrallah, Ilya M.
AU - Yang, Zhijian
AU - Duggan, Michael R.
AU - Walker, Keenan A.
AU - Getka, Alexander
AU - Baik, Kyunglok
AU - Melhem, Randa
AU - Govindarajan, Sindhuja T.
AU - Resnick, Susan M.
AU - Shou, Haochang
AU - Nikita, Konstantina
AU - Davatzikos, Christos
TI - Coupled cross-sectional and longitudinal non-negative matrix factorization reveals dominant brain aging trajectories in 48,949 individuals
T2 - Nature communications
J2 - Nat Commun
PY - 2026
DA - 2026/
VL - 17
IS - 1
SP - 5709
SN - 2041-1723
PB - Nature Portfolio
DO - 10.1038/
UR - https://
LA - en
ER -
CSL-JSON
{
"id": "10.1038/
"type": "article-journal",
"title": "Coupled cross-sectional and longitudinal non-negative matrix factorization reveals dominant brain aging trajectories in 48,949 individuals",
"container-title": "Nature communications",
"author": [
{
"family": "Skampardoni",
"given": "Ioanna"
},
{
"family": "Erus",
"given": "Guray"
},
{
"family": "Nasrallah",
"given": "Ilya M."
},
{
"family": "Yang",
"given": "Zhijian"
},
{
"family": "Duggan",
"given": "Michael R."
},
{
"family": "Walker",
"given": "Keenan A."
},
{
"family": "Getka",
"given": "Alexander"
},
{
"family": "Baik",
"given": "Kyunglok"
},
{
"family": "Melhem",
"given": "Randa"
},
{
"family": "Govindarajan",
"given": "Sindhuja T."
},
{
"family": "Resnick",
"given": "Susan M."
},
{
"family": "Shou",
"given": "Haochang"
},
{
"family": "Nikita",
"given": "Konstantina"
},
{
"family": "Davatzikos",
"given": "Christos"
}
],
"container-title-short":
"volume": "17",
"issue": "1",
"page": "5709",
"DOI": "10.1038/
"PMID": "42034902",
"PMCID": "PMC13319776",
"ISSN": "2041-1723",
"publisher": "Nature Portfolio",
"URL": "https://
"language": "en",
"issued": {
"date-parts": [
[
2026,
4,
25
]
]
}
}
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/s41467-025-57867-7 [code]
- Machine learning reveals distinct neuroanatomical signatures of cardiovascular and metabolic diseases in cognitively unimpaired individuals.Journal: Nature communicationsIn common: scikit-learn, pandas, NumPy, Alzheimer's / dementia, structural MRI / diffusion, 5 references, 6 authors
- [2] doi:10.1038/s41467-023-44271-2 [code]
- Gene-SGAN: discovering disease subtypes with imaging and genetic signatures via multi-view weakly-supervised deep clustering.Journal: Nature communicationsIn common: scikit-learn, pandas, NumPy, Alzheimer's / dementia, structural MRI / diffusion, 10 references, 4 authors
- [3] doi:10.1038/s41398-024-03121-5 [code]
- Genetic and clinical correlates of two neuroanatomical AI dimensions in the Alzheimer's disease continuum.Journal: Translational psychiatryIn common: scikit-learn, pandas, NumPy, Alzheimer's / dementia, structural MRI / diffusion, 8 references, 4 authors
- [4] doi:10.1038/s41467-021-26703-z [code]
- A deep learning framework identifies dimensional representations of Alzheimer's Disease from brain structure.Journal: Nature communicationsIn common: scikit-learn, pandas, NumPy, Alzheimer's / dementia, structural MRI / diffusion, 9 references, 3 authors
- [5] doi:10.1126/sciadv.adt7176
- Proteomic signatures of corona and herpes viral antibodies identify IGDCC4 as a mediator of neurodegeneration.Journal: Science advancesIn common: Alzheimer's / dementia, 3 references, 5 authors
- [6] doi:10.1038/s41591-025-03999-8 [code]
- MRI-based multi-organ clocks for healthy aging and disease assessment.Journal: Nature medicineIn common: statsmodels, scikit-learn, pandas, 1 other tool, structural MRI / diffusion, 1 reference, 4 authors
- [7] doi:10.1038/s43587-024-00682-4 [code]
- Proteomics identifies potential immunological drivers of postinfection brain atrophy and cognitive decline.Journal: Nature agingIn common: Alzheimer's / dementia, structural MRI / diffusion, 1 reference, 4 authors
- [8] doi:10.1038/s41467-024-46796-6 [code]
- The genetic architecture of multimodal human brain age.Journal: Nature communicationsIn common: structural MRI / diffusion, 4 references, 3 authors
- [9] doi:10.1038/s43856-025-01219-5 [code]
- Neuroanatomical dimensions in major depression linked to cognition, adverse life events, self-harm, metabolomics and genetics.Journal: Communications medicineIn common: 3 references, 3 authors
- [10] doi:10.1038/s41551-025-01506-5 [code]
- Brain-heart-eye axis revealed by multi-organ imaging genetics and proteomics.Journal: Nature biomedical engineeringIn common: scikit-learn, pandas, NumPy, 4 references, 2 authors
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.
Claim this paper
Correct its record
Say what each link of this record is, remove the ones that are not the paper's, add the ones that are missing. The correction becomes a new version of the record, in its Versions section.
Validate its tracing map
You validate the map as this page shows it: 2 repositories of the authors' code, each at its verified commit and with its license, 11 scripts, and 1 match between paragraphs and code (see the Code and Map sections). It then receives a DOI on Zenodo, with you (your ORCID iD) and OSCR as its creators; the code itself is not deposited.
The map's fingerprint: sha256:5fcb4823055e3f41…
Add the badge to the code's README
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 -->
[](https://openscicode.org/badge/doi/10.1038/s41467-026-72091-7)
<!-- oscr-badge:end -->reStructuredText
.. oscr-badge:start
.. image:: https://openscicode.org/badge/doi/10.1038/s41467-026-72091-7.svg
:target: https://openscicode.org/badge/doi/10.1038/s41467-026-72091-7
:alt: OSCR
.. oscr-badge:endHTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.1038/s41467-026-72091-7"><img src="https://openscicode.org/badge/doi/10.1038/s41467-026-72091-7.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.
