OSCR

Lifespan normative modeling of brain microstructure.

Code ↔ Paper

4 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 4 matches
  1. [1] § Results › Lifespan trajectories of brain microstructural metrics ↔ code/nm_hbr_controls10_rob_spline_age_sexbatch_v29.py, lines 50–94 · score 0.68 · hbn si, aomic id1000, ping philips, adni3 s31, abcd siemens, site CAMCAN
  2. [2] § Results › Lifespan trajectories of brain microstructural metrics ↔ code/nm_hbr_controls1_rob_spline_age_sexbatch_v29.py, lines 50–94 · score 0.68 · hbn si, aomic id1000, ping philips, adni3 s31, abcd siemens, site CAMCAN
  3. [3] § Methods › White matter regions assessed ↔ code/nm_hbr_NIMHANS_spline_age_sexbatch_transfer.py, lines 44–46 · score 0.57 · SFO, SLF, ACR, ALIC, CGC, EC
  4. [4] § Methods › White matter regions assessed ↔ code/nm_hbr_UCLA_spline_age_sexbatch_transfer.py, lines 44–46 · score 0.57 · SFO, SLF, ACR, ALIC, CGC, EC

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 · 326 lines · 18 KB · GPL-3.0 · 1 match

  1. # -*- coding: utf-8 -*-
  2. """
  3. Created in August 2024
  5. # @author: Julio Villalón
  6. """
  8. import os
  9. import pandas as pd
  10. import numpy as np
  11. from sklearn.model_selection import train_test_split
  12. from pcntoolkit.normative import estimate, evaluate
  13. from scipy.stats import norm
  14. import argparse
  15. import pickle
  16. __author__ = 'Julio Villalón'
  18. parser = argparse.ArgumentParser(description='This runs the HBR normative modeling with HBR.')
  19. parser.add_argument('-controls','--controls_csv', help='Input table with controls, including the site ID.',required=True)
  20. parser.add_argument('-dirO','--dirOutput', help='Ouput directory. Put slash at the end.',required=True)
  21. parser.add_argument('-age_column','--age_column', help='Name of the age column header.',required=True)
  22. parser.add_argument('-site_column','--site_column', help='Name of the site column header.',required=True)
  23. parser.add_argument('-sex_column','--sex_column', help='Name of the sex column header.',required=True)
  24. parser.add_argument('-outscaler','--outscaler', help='Scaling approach for output responses,\
  25. could be None (Default), standardize, minmax, or robminmax.',required=True)
  26. args = parser.parse_args()
  28. ## show the inputs ##
  29. print("Input controls file: %s" % args.controls_csv)
  30. print("Output directory: %s" % args.dirOutput)
  31. print("Name of the site column: %s" % args.site_column)
  32. print("Name of the age column: %s" % args.age_column)
  33. print("Name of the sex column: %s" % args.sex_column)
  34. print("Scaling approach for output response, outscaler: %s" % args.outscaler)
  36. data_dir = args.dirOutput
  37. if not os.path.exists(data_dir):
  38. os.makedirs(data_dir)
  40. # 21 ROIS + Average WM = 22 total rois. Bilateral, not left and right.
  41. rois =['ACR','ALIC','Average','BCC','CGC','CGH','CST','EC','FX','FXST','GCC',
  42. 'UNC','PCR','PLIC','PTR','RLIC','SCC','SCR','SFO','SLF','SS','TAP']
  44. #Reading in the controls table. Making sure
  45. dmri = args.controls_csv
  46. train_all = pd.read_csv(dmri, dtype={'subjectID': str, 'SID': str, 'Protocol_No': int, 'Protocol': str, 'Study': str})
  47. train_all["site"] = train_all[args.site_column] # Creating a new column copied from the original site column.
  48. # This makes the new columns called site_HCP, site_ABCD, site_XXX
  50. train_all_site = pd.get_dummies(train_all, columns=['site'])
  51. train_all_site['site_ID'] = train_all['site'] #adding the columns with site names to the new frame
  52. train_all_site['site_ID_bin'] = train_all['Protocol_No'] # this is the numeric version of Protocol
  54. controls_big = train_all_site.copy()
  55. train_site = controls_big[['site_ID_bin', args.sex_column]]
  58. controls_cov_big = controls_big[['subjectID',
  59. args.age_column,
  60. args.sex_column,
  61. 'site_ID',
  62. 'site_ID_bin',
  63. 'site_ABCD_SIEMENS', 'site_ABCD_GE',
  64. 'site_ABCD_PHILIPS', 'site_ADNI3_GE36',
  65. 'site_ADNI3_GE54', 'site_ADNI3_P33',
  66. 'site_ADNI3_P36', 'site_ADNI3_S127',
  67. 'site_ADNI3_S31', 'site_ADNI3_S55',
  68. 'site_AOMIC_ID1000', 'site_AOMIC_PIOP1',
  69. 'site_AOMIC_PIOP2', 'site_CAMCAN', 'site_CHBMP',
  70. 'site_CHCP', 'site_HBN_CBIC', 'site_HBN_CUNY',
  71. 'site_HBN_RUBIC', 'site_HBN_SI', 'site_HCP_A',
  72. 'site_HCP_D', 'site_HCP_YA',
  73. 'site_NIH_Peds_dti04_SIEMENS', 'site_NIH_Peds_dti04_GE',
  74. 'site_NIH_Peds_edti02_SIEMENS', 'site_NIH_Peds_edti02_GE',
  75. 'site_OASIS3', 'site_PING_GE', 'site_PING_SIEMENS',
  76. 'site_PING_PHILIPS', 'site_PNC', 'site_PPMI',
  77. 'site_QTAB', 'site_QTIM', 'site_SLIM', 'site_UKBB']]
  79. controls_features_big = controls_big[rois]
  82. X_train8020_f1, X_test8020_f1, y_train8020_f1, y_test8020_f1 = train_test_split(controls_cov_big,
  83. controls_features_big, stratify=train_site,
  84. test_size=0.2, random_state=90057)
  86. X_train_f1 = X_train8020_f1.copy()
  87. X_test_f1 = X_test8020_f1.copy()
  88. y_train_f1 = y_train8020_f1.copy()
  89. y_test_f1 = y_test8020_f1.copy()
  91. X_train_f1.reset_index(drop=True, inplace=True)
  92. X_test_f1.reset_index(drop=True, inplace=True)
  93. y_train_f1.reset_index(drop=True, inplace=True)
  94. y_test_f1.reset_index(drop=True, inplace=True)
  96. ABCD_SIEMENS_te= X_test_f1.index[X_test_f1['site_ABCD_SIEMENS'] == 1].to_list()
  97. ABCD_GE_te= X_test_f1.index[X_test_f1['site_ABCD_GE'] == 1].to_list()
  98. ABCD_PHILIPS_te= X_test_f1.index[X_test_f1['site_ABCD_PHILIPS'] == 1].to_list()
  99. ADNI3_GE36_te= X_test_f1.index[X_test_f1['site_ADNI3_GE36'] == 1].to_list()
  100. ADNI3_GE54_te= X_test_f1.index[X_test_f1['site_ADNI3_GE54'] == 1].to_list()
  101. ADNI3_P33_te= X_test_f1.index[X_test_f1['site_ADNI3_P33'] == 1].to_list()
  102. ADNI3_P36_te= X_test_f1.index[X_test_f1['site_ADNI3_P36'] == 1].to_list()
  103. ADNI3_S127_te= X_test_f1.index[X_test_f1['site_ADNI3_S127'] == 1].to_list()
  104. ADNI3_S31_te= X_test_f1.index[X_test_f1['site_ADNI3_S31'] == 1].to_list()
  105. ADNI3_S55_te= X_test_f1.index[X_test_f1['site_ADNI3_S55'] == 1].to_list()
  106. AOMIC_ID1000_te= X_test_f1.index[X_test_f1['site_AOMIC_ID1000'] == 1].to_list()
  107. AOMIC_PIOP1_te= X_test_f1.index[X_test_f1['site_AOMIC_PIOP1'] == 1].to_list()
  108. AOMIC_PIOP2_te= X_test_f1.index[X_test_f1['site_AOMIC_PIOP2'] == 1].to_list()
  109. CAMCAN_te= X_test_f1.index[X_test_f1['site_CAMCAN'] == 1].to_list()
  110. CHBMP_te= X_test_f1.index[X_test_f1['site_CHBMP'] == 1].to_list()
  111. CHCP_te= X_test_f1.index[X_test_f1['site_CHCP'] == 1].to_list()
  112. HBN_CBIC_te= X_test_f1.index[X_test_f1['site_HBN_CBIC'] == 1].to_list()
  113. HBN_CUNY_te= X_test_f1.index[X_test_f1['site_HBN_CUNY'] == 1].to_list()
  114. HBN_RUBIC_te= X_test_f1.index[X_test_f1['site_HBN_RUBIC'] == 1].to_list()
  115. HBN_SI_te= X_test_f1.index[X_test_f1['site_HBN_SI'] == 1].to_list()
  116. HCP_A_te= X_test_f1.index[X_test_f1['site_HCP_A'] == 1].to_list()
  117. HCP_D_te= X_test_f1.index[X_test_f1['site_HCP_D'] == 1].to_list()
  118. HCP_YA_te= X_test_f1.index[X_test_f1['site_HCP_YA'] == 1].to_list()
  119. NIH_Peds_dti04_SIEMENS_te= X_test_f1.index[X_test_f1['site_NIH_Peds_dti04_SIEMENS'] == 1].to_list()
  120. NIH_Peds_dti04_GE_te= X_test_f1.index[X_test_f1['site_NIH_Peds_dti04_GE'] == 1].to_list()
  121. NIH_Peds_edti02_SIEMENS_te= X_test_f1.index[X_test_f1['site_NIH_Peds_edti02_SIEMENS'] == 1].to_list()
  122. NIH_Peds_edti02_GE_te= X_test_f1.index[X_test_f1['site_NIH_Peds_edti02_GE'] == 1].to_list()
  123. OASIS3_te= X_test_f1.index[X_test_f1['site_OASIS3'] == 1].to_list()
  124. PING_GE_te= X_test_f1.index[X_test_f1['site_PING_GE'] == 1].to_list()
  125. PING_SIEMENS_te= X_test_f1.index[X_test_f1['site_PING_SIEMENS'] == 1].to_list()
  126. PING_PHILIPS_te= X_test_f1.index[X_test_f1['site_PING_PHILIPS'] == 1].to_list()
  127. PNC_te= X_test_f1.index[X_test_f1['site_PNC'] == 1].to_list()
  128. PPMI_te= X_test_f1.index[X_test_f1['site_PPMI'] == 1].to_list()
  129. QTAB_te= X_test_f1.index[X_test_f1['site_QTAB'] == 1].to_list()
  130. QTIM_te= X_test_f1.index[X_test_f1['site_QTIM'] == 1].to_list()
  131. SLIM_te= X_test_f1.index[X_test_f1['site_SLIM'] == 1].to_list()
  132. UKBB_te= X_test_f1.index[X_test_f1['site_UKBB'] == 1].to_list()
  135. dfbatch_tr = X_train_f1[['site_ID_bin', args.sex_column]]
  136. dfbatch_te = X_test_f1[['site_ID_bin', args.sex_column]]
  138. dfbatch_tr.to_csv(os.path.join(data_dir, 'batch_tr.txt'),
  139. na_rep='NaN', index=False, header=False, sep=' ')
  140. with open(data_dir + '/batch_tr.pkl', 'wb') as file:
  141. pickle.dump(dfbatch_tr, file)
  143. dfbatch_te.to_csv(os.path.join(data_dir, 'batch_te.txt'),
  144. na_rep='NaN', index=False, header=False, sep=' ')
  145. with open(data_dir + '/batch_te.pkl', 'wb') as file:
  146. pickle.dump(dfbatch_te, file)
  148. dfsubject_tr = X_train_f1[['subjectID']]
  149. dfsubject_te = X_test_f1[['subjectID']]
  150. dfsubject_tr.to_csv(os.path.join(data_dir, 'subjects_tr.txt'),
  151. na_rep='NaN', index=False, header=False, sep=' ')
  152. dfsubject_te.to_csv(os.path.join(data_dir, 'subjects_te.txt'),
  153. na_rep='NaN', index=False, header=False, sep=' ')
  156. X_train_f1 = X_train_f1.drop(['subjectID', args.sex_column, 'site_ID', 'site_ID_bin',
  157. 'site_ABCD_SIEMENS', 'site_ABCD_GE',
  158. 'site_ABCD_PHILIPS', 'site_ADNI3_GE36',
  159. 'site_ADNI3_GE54', 'site_ADNI3_P33',
  160. 'site_ADNI3_P36', 'site_ADNI3_S127',
  161. 'site_ADNI3_S31', 'site_ADNI3_S55',
  162. 'site_AOMIC_ID1000', 'site_AOMIC_PIOP1',
  163. 'site_AOMIC_PIOP2', 'site_CAMCAN', 'site_CHBMP',
  164. 'site_CHCP', 'site_HBN_CBIC', 'site_HBN_CUNY',
  165. 'site_HBN_RUBIC', 'site_HBN_SI', 'site_HCP_A',
  166. 'site_HCP_D', 'site_HCP_YA',
  167. 'site_NIH_Peds_dti04_SIEMENS', 'site_NIH_Peds_dti04_GE',
  168. 'site_NIH_Peds_edti02_SIEMENS', 'site_NIH_Peds_edti02_GE',
  169. 'site_OASIS3', 'site_PING_GE', 'site_PING_SIEMENS',
  170. 'site_PING_PHILIPS', 'site_PNC', 'site_PPMI',
  171. 'site_QTAB', 'site_QTIM', 'site_SLIM', 'site_UKBB'], axis=1)
  173. X_test_f1 = X_test_f1.drop(['subjectID', args.sex_column, 'site_ID', 'site_ID_bin',
  174. 'site_ABCD_SIEMENS', 'site_ABCD_GE',
  175. 'site_ABCD_PHILIPS', 'site_ADNI3_GE36',
  176. 'site_ADNI3_GE54', 'site_ADNI3_P33',
  177. 'site_ADNI3_P36', 'site_ADNI3_S127',
  178. 'site_ADNI3_S31', 'site_ADNI3_S55',
  179. 'site_AOMIC_ID1000', 'site_AOMIC_PIOP1',
  180. 'site_AOMIC_PIOP2', 'site_CAMCAN', 'site_CHBMP',
  181. 'site_CHCP', 'site_HBN_CBIC', 'site_HBN_CUNY',
  182. 'site_HBN_RUBIC', 'site_HBN_SI', 'site_HCP_A',
  183. 'site_HCP_D', 'site_HCP_YA',
  184. 'site_NIH_Peds_dti04_SIEMENS', 'site_NIH_Peds_dti04_GE',
  185. 'site_NIH_Peds_edti02_SIEMENS', 'site_NIH_Peds_edti02_GE',
  186. 'site_OASIS3', 'site_PING_GE', 'site_PING_SIEMENS',
  187. 'site_PING_PHILIPS', 'site_PNC', 'site_PPMI',
  188. 'site_QTAB', 'site_QTIM', 'site_SLIM', 'site_UKBB'], axis=1)
  190. X_train_f1.loc[:, args.age_column] = X_train_f1[args.age_column] / 100
  191. X_test_f1.loc[:, args.age_column] = X_test_f1[args.age_column] / 100
  193. roi_ids = ['ACR','ALIC','Average','BCC','CGC','CGH','CST','EC','FX','FXST',
  194. 'GCC','UNC','PCR','PLIC','PTR','RLIC','SCC','SCR','SFO','SLF','SS',
  195. 'TAP']
  197. # Run normative model for controls train/test split
  198. sites = [ABCD_SIEMENS_te, ABCD_GE_te, ABCD_PHILIPS_te, ADNI3_GE36_te,
  199. ADNI3_GE54_te, ADNI3_P33_te, ADNI3_P36_te, ADNI3_S127_te,
  200. ADNI3_S31_te, ADNI3_S55_te, AOMIC_ID1000_te, AOMIC_PIOP1_te,
  201. AOMIC_PIOP2_te, CAMCAN_te, CHBMP_te, CHCP_te, HBN_CBIC_te,
  202. HBN_CUNY_te, HBN_RUBIC_te, HBN_SI_te, HCP_A_te, HCP_D_te,
  203. HCP_YA_te, NIH_Peds_dti04_SIEMENS_te, NIH_Peds_dti04_GE_te,
  204. NIH_Peds_edti02_SIEMENS_te, NIH_Peds_edti02_GE_te, OASIS3_te,
  205. PING_GE_te, PING_SIEMENS_te, PING_PHILIPS_te, PNC_te, PPMI_te, QTAB_te,
  206. QTIM_te, SLIM_te, UKBB_te]
  208. site_names = ['ABCD_SIEMENS_te', 'ABCD_GE_te', 'ABCD_PHILIPS_te',
  209. 'ADNI3_GE36_te', 'ADNI3_GE54_te', 'ADNI3_P33_te', 'ADNI3_P36_te',
  210. 'ADNI3_S127_te', 'ADNI3_S31_te', 'ADNI3_S55_te', 'AOMIC_ID1000_te',
  211. 'AOMIC_PIOP1_te', 'AOMIC_PIOP2_te', 'CAMCAN_te', 'CHBMP_te',
  212. 'CHCP_te', 'HBN_CBIC_te', 'HBN_CUNY_te', 'HBN_RUBIC_te',
  213. 'HBN_SI_te', 'HCP_A_te', 'HCP_D_te', 'HCP_YA_te',
  214. 'NIH_Peds_dti04_SIEMENS_te', 'NIH_Peds_dti04_GE_te',
  215. 'NIH_Peds_edti02_SIEMENS_te', 'NIH_Peds_edti02_GE_te', 'OASIS3_te',
  216. 'PING_GE_te', 'PING_SIEMENS_te', 'PING_PHILIPS_te', 'PNC_te',
  217. 'PPMI_te', 'QTAB_te', 'QTIM_te', 'SLIM_te', 'UKBB_te']
  220. for roi in roi_ids:
  221. print('Saving the tables for ROI:', roi)
  222. roi_dir = os.path.join(data_dir, roi)
  224. if not os.path.exists(roi_dir):
  225. os.makedirs(roi_dir)
  227. np.savetxt(os.path.join(roi_dir, 'cov_int_controls_tr.txt'), X_train_f1)
  228. np.savetxt(os.path.join(roi_dir, 'cov_int_controls_te.txt'), X_test_f1)
  230. # Saving the Y for each roi
  231. np.savetxt(os.path.join(roi_dir, 'resp_controls_tr.txt'), y_train_f1[roi])
  232. np.savetxt(os.path.join(roi_dir, 'resp_controls_te.txt'), y_test_f1[roi])
  235. # Create pandas dataframes with header names to save out the overall and per-site model evaluation metrics
  236. hbr_metrics = pd.DataFrame(columns = ['ROI', 'MSLL', 'EV', 'SMSE', 'RMSE', 'Rho', 'NLL'])
  237. hbr_site_metrics = pd.DataFrame(columns = ['ROI', 'site', 'y_mean', 'y_var', 'yhat_mean', 'yhat_var', 'MSLL', 'EV', 'SMSE', 'RMSE', 'Rho'])
  240. # Loop through ROIs for controls 80/20 train/test split
  241. for roi in roi_ids:
  242. print('Running ROI:', roi)
  243. roi_dir = os.path.join(data_dir, roi)
  244. os.chdir(roi_dir)
  246. # configure the covariates to use. Change *_bspline_* to *_int_* to
  247. cov_file_tr = os.path.join(roi_dir, 'cov_int_controls_tr.txt')
  248. cov_file_te = os.path.join(roi_dir, 'cov_int_controls_te.txt')
  250. # load train & test response files
  251. resp_file_tr = os.path.join(roi_dir, 'resp_controls_tr.txt')
  252. resp_file_te = os.path.join(roi_dir, 'resp_controls_te.txt')
  254. batch_tr = os.path.join(data_dir, 'batch_tr.pkl')
  255. batch_te = os.path.join(data_dir, 'batch_te.pkl')
  257. # run a basic model
  258. yhat_te, s2_te, nm, Z, metrics_te = estimate(cov_file_tr,
  259. resp_file_tr,
  260. testresp=resp_file_te,
  261. testcov=cov_file_te,
  262. alg = 'hbr',
  263. trbefile=batch_tr,
  264. tsbefile=batch_te,
  265. model_type='bspline',
  266. savemodel = True,
  267. saveoutput = False,
  268. linear_mu ='True',
  269. linear_sigma='True',
  270. random_intercept_mu='True',
  271. random_intercept_sigma='True',
  272. random_slope_mu='False',
  273. random_slope_sigma='False',
  274. outscaler=args.outscaler
  275. )
  277. p_mosi = 1 - norm.sf(np.abs(Z)) * 2
  278. Z_p = np.concatenate((Z, p_mosi), axis=1)
  279. np.savetxt(os.path.join(roi_dir, 'Z_p_estimate.txt'), Z_p)
  280. np.savetxt(os.path.join(roi_dir, 'Yhat_estimate.txt'), yhat_te)
  281. np.savetxt(os.path.join(roi_dir, 'Ys2_estimate.txt'), s2_te)
  283. # display and save metrics
  284. keys = list(metrics_te.keys())
  285. print(keys)
  286. print(metrics_te.items())
  287. print('EV=', metrics_te['EXPV'][0])
  288. print('RHO=', metrics_te['Rho'][0])
  289. print('MSLL=', metrics_te['MSLL'][0])
  290. print('SMSE=', metrics_te['SMSE'][0])
  292. hbr_metrics.loc[len(hbr_metrics)] = [roi, metrics_te['MSLL'][0], metrics_te['EXPV'][0], metrics_te['SMSE'][0],
  293. metrics_te['RMSE'][0], metrics_te['Rho'][0], metrics_te['NLL'][0]]
  295. # Compute metrics per site in test set, save to pandas df
  296. # load true test data
  297. X_te = np.loadtxt(cov_file_te)
  298. y_te = np.loadtxt(resp_file_te)
  299. y_te = y_te[:, np.newaxis] # make sure it is a 2-d array
  301. for num, site in enumerate(sites):
  302. y_mean_te_site = np.array([[np.mean(y_te[site])]])
  303. y_var_te_site = np.array([[np.var(y_te[site])]])
  304. yhat_mean_te_site = np.array([[np.mean(yhat_te[site])]])
  305. yhat_var_te_site = np.array([[np.var(yhat_te[site])]])
  307. metrics_te_site = evaluate(y_te[site], yhat_te[site], s2_te[site], y_mean_te_site, y_var_te_site)
  309. site_name = site_names[num]
  310. hbr_site_metrics.loc[len(hbr_site_metrics)] = [roi, site_names[num],
  311. y_mean_te_site[0],
  312. y_var_te_site[0],
  313. yhat_mean_te_site[0],
  314. yhat_var_te_site[0],
  315. metrics_te_site['MSLL'][0],
  316. metrics_te_site['EXPV'][0],
  317. metrics_te_site['SMSE'][0],
  318. metrics_te_site['RMSE'][0],
  319. metrics_te_site['Rho'][0]]
  321. os.chdir(data_dir)
  322. # Save per site test set metrics variable to CSV file
  323. hbr_site_metrics.to_csv(os.path.join(data_dir, 'hbr_controls_site_metrics_f1.csv'), index=False, index_label=None)
  325. # Save overall test set metrics to CSV file
  326. hbr_metrics.to_csv(os.path.join(data_dir, 'hbr_controls_metrics_f1.csv'), index=False, index_label=None)

nm_hbr_controls10_rob_spline_age_sexbatch_v29.py at commit 6d28185, under GPL-3.0 · at the source

Overview

Authors: Julio E Villalón-Reina1, Alyssa H Zhu2, Leila Nabulsi1, Sophia I Thomopoulos1, Clara A Moreau3, Yixue Feng1, Tamoghna Chattopadhyay1, Sebastian M Benavidez1, Leila Kushan4, John P John5, Himanshu Joshi5, Iyad Ba Gari2, Katherine E Lawrence1, Talia M Nir2, Neda Jahanshad2, Carrie E Bearden4, Seyed Mostafa Kia6,7, Andre F Marquand6, the Alzheimer’s Disease Neuroimaging Initiative, Paul M Thompson1
ORCID iDs: Julio E Villalón-Reina, Sophia I Thomopoulos, Clara A Moreau, Leila Kushan, Himanshu Joshi, Iyad Ba Gari, Carrie E Bearden, Seyed Mostafa Kia, Andre F Marquand, Paul M Thompson
  1. Imaging Genetics Center, Mark & Mary Stevens Institute for Neuroimaging & Informatics, Keck School of Medicine, University of Southern California, Los Angeles, CA USA
  2. Laboratory of Brain eScience, Mark & Mary Stevens Institute for Neuroimaging & Informatics, Keck School of Medicine, University of Southern California, Los Angeles, USA
  3. Centre de recherche Azrieli, CHU Sainte-Justine, Department of Psychiatry and Addictology, University of Montreal, Montreal, QC Canada
  4. Semel Institute for Neuroscience and Human Behavior, Departments of Psychiatry and Biobehavioral Sciences and Psychology, University of California, Los Angeles, CA USA
  5. Multimodal Brain Image Analysis Laboratory, Department of Psychiatry, National Institute of Mental Health and Neuro Sciences, Bengaluru, India
  6. Donders Centre for Cognitive Neuroimaging, Donders Institute for Brain, Cognition and Behaviour. Radboud University, Nijmegen, The Netherlands
  7. Center of Cognitive Science and Artificial Intelligence, Tilburg School of Humanities and Digital Sciences, Tilburg University, Tilburg, The Netherlands
Institutions: University of Southern California (United States); Centre Hospitalier Universitaire Sainte-Justine (Canada); Université de Montréal (Canada); University of California, Los Angeles (United States); National Institute of Mental Health and Neurosciences (India); Radboud University Nijmegen (Netherlands); Tilburg University (Netherlands); Donders Centre for Cognitive Neuroimaging (Netherlands); Donders Institute for Brain, Cognition and Behaviour (Netherlands)
Journal: Nature communications, volume 17, issue 1, article 4693
Dates: received 17 December 2024; accepted 21 April 2026; published online 27 May 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1038/s41467-026-72875-x · PMID 42204143 · PMCID PMC13216618 · OpenAlex W7162560706
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: structural MRI / diffusion (modality), human (organism), Alzheimer's / dementia (population), schizophrenia / psychosis (population)
Methods: Statistics, Preprocessing, Connectivity, fMRI & imaging, Physiology & signal measures
Keywords: Predictive markers, Development of the nervous system
MeSH: Brain*, Longevity*, White Matter*, Adolescent, Adult, Aged, Aged, 80 and over, Alzheimer Disease, Bayes Theorem, Child, Child, Preschool, Cognitive Dysfunction, Diffusion Tensor Imaging, Female, Humans, Male, Middle Aged, Schizophrenia, Young Adult (* major topic)
Topic: Advanced Neuroimaging Techniques and Applications (Radiology, Nuclear Medicine and Imaging, Medicine), according to OpenAlex
Funding: European Research Council (101001118, 215698); Biotechnology and Biological Sciences Research Council (BB/H008217/1); NIMH NIH HHS (R01 MH085953); U.S. Department of Health & Human Services | NIH | Fogarty International Center (RF1AG057892); Wellcome Trust (215698/Z/19/Z); NIA NIH HHS (R01 AG060610); U.S. Department of Health & Human Services | NIH | National Institute of Mental Health (NIMH) (K01MH135160); U.S. Department of Health & Human Services | NIH | National Institute of Mental Health (R01 MH085953, K01MH135160); U.S. Department of Health & Human Services | NIH | National Institute on Aging (R01 AG060610); U.S. Department of Health & Human Services | NIH | Fogarty International Center (FIC) (RF1AG057892)
Citations: cited by 7 papers (Europe PMC); 85 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.

Repositories

Its files are read in the Code ↔ Paper reader above, with 4 matches between paragraphs and lines of code.

villalonreina/ENIGMA-DTI-Normative-Modeling

License: GPL-3.0
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: 6d281856f057c637ff665adfcf9a79af5580a9f5, 24 March 2026
Languages: Python (15)
Size: 23 files, 15 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README, license file, environment (binder/environment.yml, binder/postBuild, binder/runtime.txt)
Not found: CITATION.cff, tests, continuous integration, documentation
Tools: NumPy (14 files), pandas (14 files), SciPy (13 files), scikit-learn (12 files), Matplotlib (1 file), seaborn (1 file)
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
17 files
At the source: github.com/villalonreina/ENIGMA-DTI-Normative-Modeling

ENIGMA-git/ENIGMA-DTI-Normative-Modeling

License: GPL-3.0
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: 6d281856f057c637ff665adfcf9a79af5580a9f5, 24 March 2026
Languages: Python (15)
Size: 23 files, 15 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README, license file, environment (binder/environment.yml, binder/postBuild, binder/runtime.txt)
Not found: CITATION.cff, tests, continuous integration, documentation
Tools: NumPy (14 files), pandas (14 files), SciPy (13 files), scikit-learn (12 files), Matplotlib (1 file), seaborn (1 file)
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
17 files
At the source: github.com/ENIGMA-git/ENIGMA-DTI-Normative-Modeling

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:

Read it in the paper: doi.org/10.1038/s41467-026-72875-x.

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;
  • 30 scripts, each with its path and the digest of its content;
  • 4 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

Datasets cited

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/10.1038/s41467-026-72875-x.

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, 28 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 20 authors, 2 keywords, 19 MeSH terms, 10 funders, 79 references.

Cite

This paper

Villalón-Reina, J. E., Zhu, A. H., Nabulsi, L., Thomopoulos, S. I., Moreau, C. A., Feng, Y., Chattopadhyay, T., Benavidez, S. M., Kushan, L., John, J. P., Joshi, H., Ba Gari, I., Lawrence, K. E., Nir, T. M., Jahanshad, N., Bearden, C. E., Kia, S. M., Marquand, A. F., the Alzheimer’s Disease Neuroimaging Initiative, & Thompson, P. M. (2026). Lifespan normative modeling of brain microstructure. Nature communications, 17(1), 4693. https://doi.org/10.1038/s41467-026-72875-x

BibTeX

@article{villalonreina2026lifespan,
author = {Villalón-Reina, Julio E and Zhu, Alyssa H and Nabulsi, Leila and Thomopoulos, Sophia I and Moreau, Clara A and Feng, Yixue and Chattopadhyay, Tamoghna and Benavidez, Sebastian M and Kushan, Leila and John, John P and Joshi, Himanshu and Ba Gari, Iyad and Lawrence, Katherine E and Nir, Talia M and Jahanshad, Neda and Bearden, Carrie E and Kia, Seyed Mostafa and Marquand, Andre F and {the Alzheimer’s Disease Neuroimaging Initiative} and Thompson, Paul M},
title = {{Lifespan normative modeling of brain microstructure}},
journal = {Nature communications},
year = {2026},
month = may,
volume = {17},
number = {1},
pages = {4693},
publisher = {Nature Publishing Group},
issn = {2041-1723},
doi = {10.1038/s41467-026-72875-x},
url = {https://doi.org/10.1038/s41467-026-72875-x},
pmid = {42204143},
pmcid = {PMC13216618}
}

RIS

TY - JOUR
AU - Villalón-Reina, Julio E
AU - Zhu, Alyssa H
AU - Nabulsi, Leila
AU - Thomopoulos, Sophia I
AU - Moreau, Clara A
AU - Feng, Yixue
AU - Chattopadhyay, Tamoghna
AU - Benavidez, Sebastian M
AU - Kushan, Leila
AU - John, John P
AU - Joshi, Himanshu
AU - Ba Gari, Iyad
AU - Lawrence, Katherine E
AU - Nir, Talia M
AU - Jahanshad, Neda
AU - Bearden, Carrie E
AU - Kia, Seyed Mostafa
AU - Marquand, Andre F
AU - the Alzheimer’s Disease Neuroimaging Initiative
AU - Thompson, Paul M
TI - Lifespan normative modeling of brain microstructure
T2 - Nature communications
J2 - Nat Commun
PY - 2026
DA - 2026/05/27
VL - 17
IS - 1
SP - 4693
SN - 2041-1723
PB - Nature Publishing Group
DO - 10.1038/s41467-026-72875-x
UR - https://doi.org/10.1038/s41467-026-72875-x
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41467-026-72875-x",
"type": "article-journal",
"title": "Lifespan normative modeling of brain microstructure",
"container-title": "Nature communications",
"author": [
{
"family": "Villalón-Reina",
"given": "Julio E"
},
{
"family": "Zhu",
"given": "Alyssa H"
},
{
"family": "Nabulsi",
"given": "Leila"
},
{
"family": "Thomopoulos",
"given": "Sophia I"
},
{
"family": "Moreau",
"given": "Clara A"
},
{
"family": "Feng",
"given": "Yixue"
},
{
"family": "Chattopadhyay",
"given": "Tamoghna"
},
{
"family": "Benavidez",
"given": "Sebastian M"
},
{
"family": "Kushan",
"given": "Leila"
},
{
"family": "John",
"given": "John P"
},
{
"family": "Joshi",
"given": "Himanshu"
},
{
"family": "Ba Gari",
"given": "Iyad"
},
{
"family": "Lawrence",
"given": "Katherine E"
},
{
"family": "Nir",
"given": "Talia M"
},
{
"family": "Jahanshad",
"given": "Neda"
},
{
"family": "Bearden",
"given": "Carrie E"
},
{
"family": "Kia",
"given": "Seyed Mostafa"
},
{
"family": "Marquand",
"given": "Andre F"
},
{
"literal": "the Alzheimer’s Disease Neuroimaging Initiative"
},
{
"family": "Thompson",
"given": "Paul M"
}
],
"container-title-short": "Nat Commun",
"volume": "17",
"issue": "1",
"page": "4693",
"DOI": "10.1038/s41467-026-72875-x",
"PMID": "42204143",
"PMCID": "PMC13216618",
"ISSN": "2041-1723",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41467-026-72875-x",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
27
]
]
}
}

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/s41586-026-10454-2 [code]
White matter micro- and macrostructure brain charts for the human lifespan.
Journal: Nature
In common: pandas, SciPy, NumPy, OpenNeuro ds002785, OpenNeuro ds002790, 1 other dataset, 9 references
[2] doi:10.1002/hbm.70519 [code]
Lifespan Trajectories of Asymmetry in White Matter Tracts.
Journal: Human brain mapping
In common: OpenNeuro ds002785, OpenNeuro ds002790, 1 other dataset, structural MRI / diffusion, 5 references
[3] doi:10.1162/imag.a.1269 [code]
From early to contemporary normative modeling: Mapping individual differences in neurophysiological signals.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: seaborn, scikit-learn, pandas, 3 other tools, 12 references
[4] doi:10.64898/2026.08.13.26360304 [code]
Lifespan brain structural variation reveals shared organization across mental health conditions
Journal: medRxiv (preprint)
In common: seaborn, scikit-learn, pandas, 3 other tools, 6 references, 2 authors
[5] doi:10.21203/rs.3.rs-9246968/v1 [code]
Copy number variants reveal divergent genetic and diagnostic cortical signatures across psychiatric disorders
Journal: Research Square (preprint)
In common: pandas, NumPy, 1 reference, 4 authors
[6] doi:10.1038/s41398-026-04078-3 [code]
Brain age prediction in generalized anxiety disorder using a convolutional neural network.
Journal: Translational psychiatry
In common: structural MRI / diffusion, 7 references, 2 authors
[7] doi:10.64898/2026.03.31.26349848 [code]
Mapping Individual Neuroanatomical Alterations to Schizophrenia Psychopathology with Normative Modeling
Journal: medRxiv (preprint)
In common: seaborn, pandas, Matplotlib, 1 other tool, schizophrenia / psychosis, structural MRI / diffusion, 9 references
[8] doi:10.1038/s41380-026-03641-0
Cerebral cortical alterations in adolescent early-onset psychosis: a surface-based morphometry mega-analysis.
Journal: Molecular psychiatry
In common: schizophrenia / psychosis, structural MRI / diffusion, 2 references, 3 authors
[9] doi:10.1038/s41398-026-04426-3 [code]
Resting-state brain age is associated with cognitive and sensorimotor abnormalities in schizophrenia spectrum disorders: a validation & longitudinal study.
Journal: Translational psychiatry
In common: pandas, NumPy, OpenNeuro ds002785, OpenNeuro ds002790, schizophrenia / psychosis, structural MRI / diffusion, 3 references
[10] doi:10.21203/rs.3.rs-9914920/v1 [code]
Prediction of cognitive performance by demographics, sleep, and brain morphometry: machine learning findings from ENIGMA-Sleep Working Group
Journal: Research Square (preprint)
In common: seaborn, scikit-learn, pandas, 3 other tools, structural MRI / diffusion, 1 reference, 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.

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.