OSCR

Planar, spiral, and concentric traveling waves distinguish behavioral states in human memory.

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] § Methods › Identification of traveling waves ↔ TW_Code/par_funcs_fine.py, lines 163–268 · score 0.79 · circular linear regression, vector length, circular correlation, cos, sin, predicted
  2. [2] § Methods › Identification of traveling waves ↔ TW_Code/par_funcs_fine.py, lines 163–268 · score 0.71 · phase offset, predicted phase, Circular linear, angle, regression
  3. [3] § Methods › Validating CICA using simulations ↔ TW_Code/wave_spatial_autocorrelation_ica.ipynb, lines 89–132 · score 0.64 · circular shift, spatial autocorrelation, permutation, rotational wave, noise
  4. [4] § Methods › Electrophysiological recordings and preprocessing ↔ TW_Code/RAM_helpers.py, lines 291–436 · score 0.63 · electrode location, clinical, configuration, strips, depth, contacts
  5. [5] § Methods › Identification of oscillations ↔ TW_Code/Anup-th1-grids-2D-fine-CPCAM.ipynb, lines 324–400 · score 0.54 · normalized power spectrum, peak frequencies, algorithm, cluster, electrode
  6. [6] § Methods › Identification of oscillations ↔ TW_Code/par_funcs_fine.py, lines 22–49 · score 0.51 · standard deviation, power spectrum, fit, peaks, electrode

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 · 464 lines · 16 KB · no license · 3 matches

  1. """
  2. A set of functions that you
  3. """
  4. import numpy as np
  5. import statsmodels.api as sm
  6. import pdb
  7. import numexpr
  8. from scipy.signal import argrelmax
  9. from scipy.stats import ttest_ind
  10. from xarray import concat
  11. from ptsa.data.timeseries import TimeSeries
  12. from ptsa.data.filters import MorletWaveletFilter
  13. from tqdm import tqdm
  14. import pycircstat
  15. from numpy import exp
  16. from scipy.optimize import curve_fit
  17. import matplotlib.pyplot as plt
  18. from math import cos,sin
  22. def par_find_peaks_by_chan(p_spect_array, frequencies, std_thresh=1.):
  23. """
  24. Parameters
  25. ----------
  26. p_spect_array: numpy.ndarray
  27. An array with dimensions frequencies x channels
  28. frequencies: numpy.ndarray
  29. An array of the frequencies used
  30. std_thresh: float
  31. Threshold in number of standard deviations above the corrected power spectra to be counted as a peak
  33. Returns
  34. -------
  35. peaks_all_chans: numpy.ndarray with type bool
  36. An array of booleans the same shape as p_spect_array, specifying if there is a peak at a given frequency
  37. and electrode
  38. """
  40. peaks_all_chans = np.zeros(p_spect_array.shape).astype(bool)
  41. for i, chan_data in enumerate(p_spect_array.T):
  42. x = sm.tools.tools.add_constant(np.log10(frequencies))
  43. model_res = sm.RLM(chan_data, x).fit()
  44. peak_inds = argrelmax(model_res.resid)
  45. peaks = np.zeros(x.shape[0], dtype=bool)
  46. peaks[peak_inds] = True
  47. above_thresh = model_res.resid > (np.std(model_res.resid) * std_thresh)
  48. peaks_all_chans[:,i] = peaks & above_thresh
  49. return peaks_all_chans
  52. def par_robust_reg(info):
  53. """
  54. Parallelizable robust regression function
  56. info: two element list. first element, power spectra: # freqs x # elecs. Second element: log transformed freqs
  58. returns intercepts, slopes, resids
  59. """
  61. p_spects = info[0]
  62. x = sm.tools.tools.add_constant(info[1])
  64. # holds slope of fit line
  65. slopes = np.empty((p_spects.shape[1]))
  66. slopes[:] = np.nan
  68. # holds residuals
  69. resids = np.empty((p_spects.shape[0], p_spects.shape[1]))
  70. resids[:] = np.nan
  72. # holds intercepts
  73. intercepts = np.empty((p_spects.shape[1]))
  74. intercepts[:] = np.nan
  76. # holds mean height of fit line
  77. bband_power = np.empty((p_spects.shape[1]))
  78. bband_power[:] = np.nan
  80. # loop over every electrode
  81. for i, y in enumerate(p_spects.T):
  82. model_res = sm.RLM(y, x).fit()
  83. intercepts[i] = model_res.params[0]
  84. slopes[i] = model_res.params[1]
  85. bband_power[i] = model_res.fittedvalues.mean()
  86. resids[:, i] = model_res.resid
  88. return intercepts, slopes, resids, bband_power
  91. def par_robust_reg_no_low_freqs(info):
  92. """
  93. Parallelizable robust regression function
  95. info: two element list. first element, power spectra: # freqs x # elecs. Second element: log transformed freqs
  97. returns intercepts, slopes, resids
  98. """
  100. p_spects = info[0]
  101. x = sm.tools.tools.add_constant(info[1])
  102. freq_inds = info[2]
  104. # holds slope of fit line
  105. slopes = np.empty((p_spects.shape[1]))
  106. slopes[:] = np.nan
  108. # holds residuals
  109. resids = np.empty((p_spects.shape[0], p_spects.shape[1]))
  110. resids[:] = np.nan
  112. # holds intercepts
  113. intercepts = np.empty((p_spects.shape[1]))
  114. intercepts[:] = np.nan
  116. # holds mean height of fit line
  117. bband_power = np.empty((p_spects.shape[1]))
  118. bband_power[:] = np.nan
  120. # loop over every electrode
  121. for i, y in enumerate(p_spects.T):
  122. model_res = sm.RLM(y[freq_inds], x[freq_inds]).fit()
  123. intercepts[i] = model_res.params[0]
  124. slopes[i] = model_res.params[1]
  125. bband_power[i] = model_res.fittedvalues.mean()
  126. resids[:, i] = y - ((x[:, 1]*model_res.params[1]) + model_res.params[0])
  128. return intercepts, slopes, resids, bband_power
  131. def my_local_max(arr):
  132. """
  133. Returns indices of local maxima in a 1D array. Unlike scipy.signal.argrelmax, this does not ignore consecutive
  134. values that are peaks. It finds the last repetition.
  136. """
  137. b1 = arr[:-1] <= arr[1:]
  138. b2 = arr[:-1] > arr[1:]
  139. k = np.where(b1[:-1] & b2[1:])[0] + 1
  140. if arr[0] > arr[1]:
  141. k = np.append(k, 0)
  142. if arr[-1] > arr[-2]:
  143. k = np.append(k, len(arr) - 1)
  144. return k
  147. def par_find_peaks(info):
  148. """
  149. Parallelizable peak picking function, uses robust reg but returns
  151. """
  153. p_spect = info[0]
  154. x = sm.tools.tools.add_constant(info[1])
  155. model_res = sm.RLM(p_spect, x).fit()
  156. peak_inds = my_local_max(model_res.resid)
  157. peaks = np.zeros(x.shape[0], dtype=bool)
  158. peaks[peak_inds] = True
  159. above_thresh = model_res.resid > np.std(model_res.resid)
  160. peaks = peaks & above_thresh
  161. return peaks
  163. def circ_lin_regress(phases, coords, theta_r, params):
  164. """
  165. Performs 2D circular linear regression.
  167. This is ported from Honghui's matlab code.
  169. :param phases:
  170. :param coords:
  171. :return:
  172. """
  174. n = phases.shape[1]
  175. pos_x = np.expand_dims(coords[:, 0], 1)
  176. pos_y = np.expand_dims(coords[:, 1], 1)
  178. # compute predicted phases for angle and phase offset
  179. x = np.expand_dims(phases, 2) - params[:, 0] * pos_x - params[:, 1] * pos_y
  181. # Compute resultant vector length. This is faster than calling pycircstat.resultant_vector_length
  182. # now = time.time()
  183. x1 = numexpr.evaluate('sum(cos(x) / n, axis=1)')
  184. x1 = numexpr.evaluate('x1 ** 2')
  185. x2 = numexpr.evaluate('sum(sin(x) / n, axis=1)')
  186. x2 = numexpr.evaluate('x2 ** 2')
  187. Rs = numexpr.evaluate('-sqrt(x1 + x2)')
  190. # for each time and event, find the parameters with the smallest -R
  191. min_vals = theta_r[np.argmin(Rs, axis=1)]
  193. sl = min_vals[:, 1] * np.array([np.cos(min_vals[:, 0]), np.sin((min_vals[:, 0]))])
  194. offs = np.arctan2(np.sum(np.sin(phases.T - sl[0, :] * pos_x - sl[1, :] * pos_y), axis=0),
  195. np.sum(np.cos(phases.T - sl[0, :] * pos_x - sl[1, :] * pos_y), axis=0))
  196. pos_circ = np.mod(sl[0, :] * pos_x + sl[1, :] * pos_y + offs, 2 * np.pi)
  198. # compute circular correlation coefficient between actual phases and predicited phases
  199. circ_corr_coef = pycircstat.corrcc(phases.T, pos_circ, axis=0)
  201. # compute adjusted r square
  202. # pdb.set_trace()
  203. r2_adj = circ_corr_coef ** 2
  204. # r2_adj = 1 - ((1 - circ_corr_coef ** 2) * (n - 1)) / (n - 4)
  206. wave_ang = min_vals[:, 0]
  207. wave_freq = min_vals[:, 1]
  214. #re-estimate finer parameters
  215. wave_ang_fine = []
  216. wave_freq_fine = []
  217. r2_adj_fine = []
  218. offs_fine = []
  220. for i_trial in range(0, np.shape(wave_ang)[0], 1):
  222. wave_ang_deg = np.degrees(wave_ang[i_trial])
  223. wave_freq_deg = np.degrees(wave_freq[i_trial])
  225. thetas_tmp = np.radians(np.arange(np.max([0, wave_ang_deg-2.5]), np.min([wave_ang_deg+2.5, 359.99]), 0.05))
  226. rs_tmp = np.radians(np.arange(np.max([0, wave_freq_deg-0.5]), np.min([wave_freq_deg+0.5, 17.99]), 0.05))
  228. theta_r_tmp = np.stack([(x, y) for x in thetas_tmp for y in rs_tmp])
  229. params_tmp = np.stack([theta_r_tmp[:, 1] * np.cos(theta_r_tmp[:, 0]), theta_r_tmp[:, 1] * np.sin(theta_r_tmp[:, 0])], -1)
  231. # compute predicted phases for angle and phase offset
  232. x = np.expand_dims(np.expand_dims(phases[i_trial, :], 1), 0) - params_tmp[:, 0] * pos_x - params_tmp[:, 1] * pos_y
  234. # Compute resultant vector length. This is faster than calling pycircstat.resultant_vector_length
  235. # now = time.time()
  236. x1 = numexpr.evaluate('sum(cos(x) / n, axis=1)')
  237. x1 = numexpr.evaluate('x1 ** 2')
  238. x2 = numexpr.evaluate('sum(sin(x) / n, axis=1)')
  239. x2 = numexpr.evaluate('x2 ** 2')
  240. Rs = numexpr.evaluate('-sqrt(x1 + x2)')
  243. # for each time and event, find the parameters with the smallest -R
  244. min_vals = theta_r_tmp[np.argmin(Rs, axis=1)]
  246. sl = min_vals[:, 1] * np.array([np.cos(min_vals[:, 0]), np.sin((min_vals[:, 0]))])
  247. offs_tmp = np.arctan2(np.sum(np.sin(np.expand_dims(phases[i_trial, :], 1) - sl[0, :] * pos_x - sl[1, :] * pos_y), axis=0),
  248. np.sum(np.cos(np.expand_dims(phases[i_trial, :], 1) - sl[0, :] * pos_x - sl[1, :] * pos_y), axis=0))
  249. offs_fine.append(offs_tmp)
  250. pos_circ = np.mod(sl[0, :] * pos_x + sl[1, :] * pos_y + offs_tmp, 2 * np.pi)
  252. # compute circular correlation coefficient between actual phases and predicited phases
  253. circ_corr_coef = pycircstat.corrcc(np.expand_dims(phases[i_trial, :], 1), pos_circ, axis=0)
  255. # compute adjusted r square
  256. # pdb.set_trace()
  257. r2_adj_fine.append(circ_corr_coef ** 2)
  258. # r2_adj = 1 - ((1 - circ_corr_coef ** 2) * (n - 1)) / (n - 4)
  260. wave_ang_fine.append(min_vals[:, 0])
  261. wave_freq_fine.append(min_vals[:, 1])
  263. wave_ang_fine = np.squeeze(np.array(wave_ang_fine))
  264. wave_freq_fine = np.squeeze(np.array(wave_freq_fine))
  265. r2_adj_fine = np.squeeze(np.array(r2_adj_fine))
  266. offs_fine = np.squeeze(np.array(offs_fine))
  268. return wave_ang_fine, wave_freq_fine, r2_adj_fine, offs_fine
  271. def channel_peak(data):
  272. params,rs=findpeak(data)
  273. peaks = np.zeros(len(data)).astype(bool)
  274. #params=[ int(round(i)) for i in params]
  275. if len(params)==5:
  276. if 5<params[3]<195 and params[2]>.3 and params[4]<50:
  277. peaks[int(round(params[3]))]=True
  278. elif len(params)==8:
  279. if 5<params[3]<195 and params[2]>.3 and params[4]<50:
  280. peaks[int(round(params[3]))]=True
  281. if 5<params[6]<195 and params[5]>.3 and params[7]<50:
  282. peaks[int(round(params[6]))]=True
  283. elif len(params)==11:
  284. if 5<params[3]<195 and params[2]>.3 and params[4]<50:
  285. peaks[int(round(params[3]))]=True
  286. if 5<params[6]<195 and params[5]>.3 and params[7]<50:
  287. peaks[int(round(params[6]))]=True
  288. if 5<params[9]<195 and params[8]>.3 and params[10]<50:
  289. peaks[int(round(params[9]))]=True
  290. return peaks
  291. def findpeak(data):
  292. line,peaks,heights=robustfit(data)
  293. #when no peaks found from robustfit, return no peaks, extremely rare ad extremely annoying!!!
  294. if len(peaks)==0 or line[1]>0 or line[0]<0:
  295. return line,0
  296. x=np.array(list(range(len(data))))
  297. y=data
  298. # Just fit 1 peak
  299. initial=[line[1],line[0],1,peaks[0],5]
  300. upper=[0,np.inf,np.inf,199,np.inf]
  301. lower=[-np.inf,0,0,0,0]
  302. best_vals,cov = curve_fit(oneOverF1, x, y,p0=initial,bounds=(lower,upper))
  303. Rs=np.corrcoef(y,oneOverF1(x,*best_vals))[0,1]**2;
  304. Rs=Radjust(Rs,5,200)
  305. res=best_vals
  306. RS=Rs
  307. # if 1 peaks are good enough, return
  308. if RS>.999:
  309. return res,RS
  311. # fit 2 peaks
  312. upper=[0,np.inf,np.inf,199,np.inf,np.inf,199,np.inf]
  313. lower=[-np.inf,0,0,0,0,0,0,0]
  314. if len(peaks)>=2:
  315. initial=[line[1],line[0],1,peaks[0],5,1,peaks[1],5]
  316. best_vals,cov = curve_fit(oneOverF2, x, y,p0=initial,bounds=(lower,upper))
  317. Rs=np.corrcoef(y,oneOverF2(x,*best_vals))[0,1]**2;
  318. Rs=Radjust(Rs,8,200)
  319. if RS<Rs:
  320. res=best_vals
  321. RS=Rs
  323. # fit 2 peaks with interpolation
  324. # first interpolation point
  325. initial=[line[1],line[0],1,peaks[0],5,1,peaks[0]/2,5]
  326. best_vals,cov = curve_fit(oneOverF2, x, y,p0=initial,bounds=(lower,upper))
  327. Rs=np.corrcoef(y,oneOverF2(x,*best_vals))[0,1]**2;
  328. Rs=Radjust(Rs,8,200)
  329. if RS<Rs:
  330. res=best_vals
  331. RS=Rs
  333. # second interpolation point
  334. initial=[line[1],line[0],1,peaks[0],5,1,(peaks[0]+199)/2,5]
  335. best_vals,cov = curve_fit(oneOverF2, x, y,p0=initial,bounds=(lower,upper))
  336. Rs=np.corrcoef(y,oneOverF2(x,*best_vals))[0,1]**2;
  337. Rs=Radjust(Rs,8,200)
  338. if RS<Rs:
  339. res=best_vals
  340. RS=Rs
  342. # if 2 peaks are good enough, return
  343. if RS>.999 or len(peaks)<2:
  344. return res,RS
  346. # fit 3 peaks
  348. upper=[0,np.inf,np.inf,199,199,np.inf,199,199,np.inf,199,199]
  349. lower=[-np.inf,0,0,0,0,0,0,0,0,0,0]
  350. if len(peaks)>2:
  351. initial=[line[1],line[0],1,peaks[0],5,1,peaks[1],5,1,peaks[2],5]
  352. best_vals,cov = curve_fit(oneOverF3, x, y,p0=initial,bounds=(lower,upper))
  353. Rs=np.corrcoef(y,oneOverF3(x,*best_vals))[0,1]**2;
  354. Rs=Radjust(Rs,11,200)
  355. if RS<Rs:
  356. res=best_vals
  357. RS=Rs
  359. # fit 2 peaks plus different interpolation points cause the 3rd peak is not that reliable.
  360. Peak=sorted(peaks[:2])
  361. # first interpolation point
  362. initial=[line[1],line[0],1,Peak[0]/2,5,1,Peak[0],5,1,Peak[1],5]
  363. best_vals,cov = curve_fit(oneOverF3, x, y,p0=initial,bounds=(lower,upper))
  364. Rs=np.corrcoef(y,oneOverF3(x,*best_vals))[0,1]**2;
  365. Rs=Radjust(Rs,11,len(data))
  366. if RS<Rs:
  367. res=best_vals
  368. RS=Rs
  370. # second interpolation point
  371. initial=[line[1],line[0],1,Peak[0],5,1,sum(Peak)/2,5,1,Peak[1],5]
  372. best_vals,cov = curve_fit(oneOverF3, x, y,p0=initial,bounds=(lower,upper))
  373. Rs=np.corrcoef(y,oneOverF3(x,*best_vals))[0,1]**2;
  374. Rs=Radjust(Rs,11,200)
  375. if RS<Rs:
  376. res=best_vals
  377. RS=Rs
  379. # third interpolation point
  380. initial=[line[1],line[0],1,Peak[0],5,1,Peak[1],5,1,Peak[1]/2+99.5,5]
  381. best_vals,cov = curve_fit(oneOverF3, x, y,p0=initial,bounds=(lower,upper))
  382. Rs=np.corrcoef(y,oneOverF3(x,*best_vals))[0,1]**2;
  383. Rs=Radjust(Rs,11,200)
  384. if RS<Rs:
  385. res=best_vals
  386. RS=Rs
  387. return res,RS
  388. def oneOverF1(x,slope,intercept, amp1, cen1, wid1):
  389. return x*slope+intercept+amp1 * exp(-.5*(x-cen1)**2 / wid1**2)
  391. def oneOverF2(x,slope,intercept, amp1, cen1, wid1,amp2, cen2, wid2):
  392. return x*slope+intercept+amp1 * exp(-.5*(x-cen1)**2 / wid1**2)+amp2 * exp(-.5*(x-cen2)**2 / wid2**2)
  394. def oneOverF3(x,slope,intercept, amp1, cen1, wid1,amp2, cen2, wid2,amp3, cen3, wid3):
  395. return x*slope+intercept+amp1 * exp(-.5*(x-cen1)**2 / wid1**2)+amp2 * exp(-.5*(x-cen2)**2 / wid2**2)+amp3 * exp(-.5*(x-cen3)**2 / wid3**2)
  397. def Radjust(Rs,k,n):
  398. return 1-(1-Rs)*(n-1)/(n-k-1)
  400. def robustfit(data):
  401. x = sm.tools.tools.add_constant(np.array(range(200)))
  402. model_res = sm.RLM(data, x).fit()
  403. peak_inds = np.squeeze(argrelmax(model_res.resid))
  404. if peak_inds.size==1:
  405. peak_inds=np.expand_dims(peak_inds,0)
  406. peak_inds=sorted(peak_inds,key=lambda i:model_res.resid[i])[::-1]
  407. return model_res.params,peak_inds[:3],model_res.resid[peak_inds[:3]]
  409. def par_find_peaks_by_chan2(p_spect_array, frequencies, std_thresh=1.):
  410. """
  411. Parameters
  412. ----------
  413. p_spect_array: numpy.ndarray
  414. An array with dimensions frequencies x channels
  415. frequencies: numpy.ndarray
  416. An array of the frequencies used
  417. std_thresh: float
  418. Threshold in number of standard deviations above the corrected power spectra to be counted as a peak
  420. Returns
  421. -------
  422. peaks_all_chans: numpy.ndarray with type bool
  423. An array of booleans the same shape as p_spect_array, specifying if there is a peak at a given frequency
  424. and electrode
  425. """
  427. peaks_all_chans = np.zeros(p_spect_array.shape).astype(bool)
  428. for i, chan_data in enumerate(p_spect_array.T):
  429. peaks_all_chans[:,i] = channel_peak(chan_data)
  430. return peaks_all_chans
  433. def rbar(x):
  434. n=len(x)
  435. x1 = numexpr.evaluate('sum(cos(x) / n, axis=0)')
  436. x1 = numexpr.evaluate('x1 ** 2')
  437. x2 = numexpr.evaluate('sum(sin(x) / n, axis=0)')
  438. x2 = numexpr.evaluate('x2 ** 2')
  439. Rs = numexpr.evaluate('sqrt(x1 + x2)')
  440. return Rs
  441. def circ_hist2(data,n=20,x=0,y=0,radius=1,color='k'):
  442. bins=np.linspace(-np.pi,np.pi,n+1)
  443. counts,bins=np.histogram(data,bins)
  444. counts=radius*counts/np.max(counts)
  445. for i in range(len(counts)):
  446. plt.plot([x,x+counts[i]*cos(bins[i])],[y,y+counts[i]*sin(bins[i])],color=color,linewidth=.3)
  447. plt.plot([x,x+counts[i]*cos(bins[i+1])],[y,y+counts[i]*sin(bins[i+1])],color=color,linewidth=.3)
  448. arch=np.linspace(bins[i],bins[i+1],100)
  449. plt.plot(x+counts[i]*np.cos(arch),y+counts[i]*np.sin(arch),color=color,linewidth=.5)
  450. plt.axis('equal')
  453. def circ_hist(data,n=20,x=0,y=0,radius=1,color='k',alpha=.3):
  454. # circular histogram, data need to be within -pi to pi
  455. bins=np.linspace(-np.pi,np.pi,n+1)
  456. counts,bins=np.histogram(data,bins)
  457. counts=radius*counts/np.max(counts)
  458. for i in range(len(counts)):
  459. X=[x,x+counts[i]*cos(bins[i]),x+counts[i]*cos(bins[i+1]),x]
  460. Y=[y,y+counts[i]*sin(bins[i]),y+counts[i]*sin(bins[i+1]),y]
  461. plt.fill(X,Y,color=color,alpha=alpha,linewidth=0)
  462. plt.axis('equal')
  463. def adjust(rs,n,k):
  464. return 1 - ((1 - rs) * (n - 1)) / (n - 1-k)

par_funcs_fine.py at commit 68587cf, no license · at the source

Overview

Authors: Anup Das1, Erfan Zabeh2, Bard Ermentrout3, Joshua Jacobs1
ORCID iDs: Anup Das, Erfan Zabeh, Bard Ermentrout, Joshua Jacobs
  1. Department of Neurology, University of Chicago,Chicago, IL USA
  2. Department of Biomedical Engineering, Columbia University,New York, NY USA
  3. Department of Mathematics, University of Pittsburgh,Pittsburgh, PA USA
Institutions: University of Chicago (United States); Columbia University (United States); University of Pittsburgh (United States)
Journal: Nature communications, volume 17, issue 1, article 5143
Dates: received 21 July 2025; accepted 9 March 2026; published online 11 April 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1038/s41467-026-71386-z · PMID 41963323 · PMCID PMC13250086 · OpenAlex W7152990215
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: human (organism), cognitive (subfield)
Methods: Spectral & time-frequency, Connectivity, Statistics, Smoothing, state filtering, decompositions, Machine learning, Preprocessing
Keywords: Cognitive neuroscience, Neuroscience
MeSH: Brain*, Cerebral Cortex*, Memory*, Adult, Brain Mapping, Female, Humans, Male, Young Adult (* major topic)
Topic: Memory and Neural Mechanisms (Cognitive Neuroscience, Neuroscience), according to OpenAlex
Funding: National Science Foundation (CAREER)
Citations: cited by 4 papers (Europe PMC); 101 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.

anupdas777/complex_traveling_waves

License: none: the authors keep all their rights
State: the link answers, verified on 29 September 2026
Evidence: files inventoried
Commit: 68587cfd4d364699a31687eef41db080acf796e3, 20 May 2025
Languages: Jupyter (3), Python (2)
Size: 20 files, 5 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: 3 notebooks
Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: NumPy (5 files), SciPy (5 files), Matplotlib (4 files), pandas (4 files), scikit-learn (3 files), CircStat (2 files), xarray (2 files), Nilearn (1 file), Pillow (1 file), seaborn (1 file), statsmodels (1 file)
Availability: 1 check, the latest on 29 September 2026: the link answers
  • 29 September 2026: the link answers
5 files
At the source: github.com/anupdas777/complex_traveling_waves/tree/main

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/s41467-026-71386-z.

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.

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:

  • no repository, dataset or request procedure was recognized in it

Read it in the paper: doi.org/10.1038/s41467-026-71386-z.

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, 29 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 4 authors, 2 keywords, 9 MeSH terms, 1 funder, 94 references.

Cite

This paper

Das, A., Zabeh, E., Ermentrout, B., & Jacobs, J. (2026). Planar, spiral, and concentric traveling waves distinguish behavioral states in human memory. Nature communications, 17(1), 5143. https://doi.org/10.1038/s41467-026-71386-z

BibTeX

@article{das2026planar,
author = {Das, Anup and Zabeh, Erfan and Ermentrout, Bard and Jacobs, Joshua},
title = {{Planar, spiral, and concentric traveling waves distinguish behavioral states in human memory}},
journal = {Nature communications},
year = {2026},
month = apr,
volume = {17},
number = {1},
pages = {5143},
publisher = {Nature Publishing Group},
issn = {2041-1723},
doi = {10.1038/s41467-026-71386-z},
url = {https://doi.org/10.1038/s41467-026-71386-z},
pmid = {41963323},
pmcid = {PMC13250086}
}

RIS

TY - JOUR
AU - Das, Anup
AU - Zabeh, Erfan
AU - Ermentrout, Bard
AU - Jacobs, Joshua
TI - Planar, spiral, and concentric traveling waves distinguish behavioral states in human memory
T2 - Nature communications
J2 - Nat Commun
PY - 2026
DA - 2026/04/11
VL - 17
IS - 1
SP - 5143
SN - 2041-1723
PB - Nature Publishing Group
DO - 10.1038/s41467-026-71386-z
UR - https://doi.org/10.1038/s41467-026-71386-z
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41467-026-71386-z",
"type": "article-journal",
"title": "Planar, spiral, and concentric traveling waves distinguish behavioral states in human memory",
"container-title": "Nature communications",
"author": [
{
"family": "Das",
"given": "Anup"
},
{
"family": "Zabeh",
"given": "Erfan"
},
{
"family": "Ermentrout",
"given": "Bard"
},
{
"family": "Jacobs",
"given": "Joshua"
}
],
"container-title-short": "Nat Commun",
"volume": "17",
"issue": "1",
"page": "5143",
"DOI": "10.1038/s41467-026-71386-z",
"PMID": "41963323",
"PMCID": "PMC13250086",
"ISSN": "2041-1723",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41467-026-71386-z",
"language": "en",
"issued": {
"date-parts": [
[
2026,
4,
11
]
]
}
}

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:10.7554/elife.108208 [code]
Realistic coupling enables flexible macroscopic traveling waves in the mouse cortex.
Journal: eLife
In common: pandas, NumPy, 17 references
[2] doi:10.3389/fncom.2026.1844662
Quantifying cortex-wide traveling brain waves of complex patterns with a graph-based algorithm.
Journal: Frontiers in computational neuroscience
In common: 14 references
[3] doi:10.7554/elife.100674 [code]
The dominance of large-scale phase dynamics in human cortex, from delta to gamma.
Journal: eLife
In common: SciPy, Matplotlib, NumPy, 10 references
[4] doi:10.1038/s41467-026-72931-6 [code]
Three parsimonious spatiotemporal patterns in cerebellum reveal individual traits in function and behavior.
Journal: Nature communications
In common: Nilearn, statsmodels, seaborn, 5 other tools, cognitive, 5 references
[5] doi:10.7554/elife.106753 [code]
Traveling waves across scales: Different mechanisms but same canonical computation?
Journal: n/a
In common: 10 references
[6] doi:10.1073/pnas.2527296123
Traveling-wave transcranial alternating current stimulation (twtACS) causally links neural timing to cognitive function.
Journal: Proceedings of the National Academy of Sciences of the United States of America
In common: cognitive, 9 references
[7] doi:10.1371/journal.pcbi.1013488 [code]
Evaluating place cell detection methods in Rats and Humans: Implications for cross-species spatial coding.
Journal: PLoS computational biology
In common: statsmodels, seaborn, scikit-learn, 4 other tools, 1 reference, author Joshua Jacobs
[8] doi:10.1038/s41467-026-75959-w [code]
Charting higher-order models of brain function beyond pairwise interactions.
Journal: Nature communications
In common: xarray, Nilearn, statsmodels, 6 other tools, 2 references
[9] doi:10.1371/journal.pcbi.1013007 [code]
Traveling waves in the human visual cortex: An MEG-EEG model-based approach
Journal: n/a
In common: SciPy, Matplotlib, NumPy, 7 references
[10] doi:10.1016/j.isci.2026.116728 [code]
Awake cortex stabilizes traveling waves for global and reliable information routing.
Journal: iScience
In common: SciPy, NumPy, 7 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.

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.