OSCR

Dual mechanism of anti-seizure medications in controlling seizure activity.

Code ↔ Paper

The paper beside its authors' code: matches between them have not been computed for this paper yet.

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 · 805 lines · 27 KB · no license

  1. # Auxiliary code for __
  2. # This code generats results figures for the study mentioned above and some
  3. # supplementary material
  5. # This code has been developed on Python 3.12.5 with Spyder 6
  6. # The code has been designed to be abel to run by sections on Spyder, other
  7. # interpreters may not identify the same section berakpoints.
  9. # Further details on readme document
  11. #%% Initialize workspace
  12. # Import required libraries
  13. import os
  14. import numpy as np
  15. import pandas as pd
  16. import matplotlib.pyplot as plt
  17. import seaborn as sns
  19. from scipy import stats
  21. import statsmodels.formula.api as smf
  23. # Figure generation workspace
  24. # Uncomment the following 6 lines to obtain better quality figure generation
  25. import matplotlib.font_manager as ft
  26. import matplotlib
  27. matplotlib.use('cairo')
  28. plt.rc('font', family='arial')
  29. plt. rcParams.update({'font.size':9})
  30. font = ft.FontProperties(family = 'arial')
  32. # Define the Origin path for all data
  33. path_Org = os.getcwd()
  35. # Clustering data importins
  36. sz_Clust = pd.read_csv(path_Org + '/data/rednmf_l10.csv')
  37. Stt_Freq_asm = pd.read_csv(path_Org + '/data/rednmf_l10_szFrq.csv')
  39. # Figure saving flag: True = Save figures
  40. save_fig = True
  42. # Figure saving output folders
  43. path_save_fig = path_Org + '/ASM_Result_figs/'
  45. # Number of decimas on statistical analysis show
  46. Stt_Dgt = 4
  48. # SVG/PDF figure folders
  49. if not os.path.exists(path_save_fig + '/PDF'):
  50. os.makedirs(path_save_fig + '/PDF')
  52. #%% Results: Seizure dutation and Frecuancy with ASM levels
  54. # Duration for all seizures available on the dataset
  55. rsp = "Duration_10"
  57. # Figure Aperance variables
  58. Y_Lim1 = [0,3]
  59. Y_Lim2 = [0,3]
  60. Y_Label= "Duration (log10)"
  62. X_Lim_asm = [-10,1]
  63. X_Lim_tod = [0,24]
  65. palt = "mako"
  66. Hue_Patt = ["Normo","Under"]
  67. Axs_Patt = ["Under","Normo"]
  68. Cater = "ASM_stt"
  70. # MELM formula for duation against ASM levels and TofD
  71. # sTOD+cTOD -> Incorporation ciradian rythm (24h) as a circular variable
  72. # sTOD_h+cTOD_h -> Incorporation ultradian 12h rythm as a circular variable
  73. prd = "ASM_lvl + sTOD+cTOD + sTOD_h+cTOD_h"
  75. # Clear dataset from non-relevant data
  76. rm_dx = np.logical_or(np.isnan(sz_Clust[rsp]),np.isinf(sz_Clust[rsp]))
  77. stData = sz_Clust.drop(np.where(rm_dx)[0]).reset_index(drop=True)
  79. # Fit the mixed effect model with patietn identifier group-level effect
  80. mdl_Chr_Ams = smf.mixedlm(rsp + " ~ " + prd , stData, groups=stData.patient_id).fit()
  81. mdl_Chr_Ams.summary()
  83. # Relevent effect sizes and p-values for figure generation
  84. rnd_eff = mdl_Chr_Ams.random_effects
  85. int_eff = mdl_Chr_Ams.params.Intercept
  87. asm_eff = mdl_Chr_Ams.params.ASM_lvl
  88. asm_pvl = mdl_Chr_Ams.pvalues.ASM_lvl
  90. plt.figure(figsize=[15,5])
  92. # Plot raw data for 3 individuals: ASM level vs Seizure duration --------------
  93. # Patient Ids to show
  94. id_rep = [78,46,48]
  96. # Color code
  97. id_col = [[59/255, 117/255, 175/255],
  98. [235/255,125/255, 48/255],
  99. [80/255, 156/255, 61/255]]
  101. # Plot data poitns and Fixed effect individually
  102. for i in range(0,len(id_rep)):
  103. plt.subplot(1,3,2)
  105. # Identify subject data
  106. idx = np.where(sz_Clust.patient_id == id_rep[i])[0]
  108. # Data poitns scatterplot
  109. sns.scatterplot(x=sz_Clust.ASM_lvl[idx], y=sz_Clust[rsp][idx],
  110. label="id" + str(id_rep[i]), color = id_col[i])
  112. # Generate regresion with MELM data
  113. grp_eff = rnd_eff.get(id_rep[i]).Group # Group effect
  114. X = np.linspace(min(sz_Clust.ASM_lvl[idx]),# ASM level range
  115. max(sz_Clust.ASM_lvl[idx]))
  116. Y = X * asm_eff + grp_eff + int_eff # Regresion
  117. sns.lineplot(x=X, y=Y, color = id_col[i]) # Line with matching color code
  119. # Mark the ASM levels correponding to NOrmal-dose in grey
  120. plt.fill_between(range(X_Lim_asm[0],X_Lim_asm[1]+1), Y_Lim1[0], Y_Lim1[1],
  121. np.array(range(X_Lim_asm[0],X_Lim_asm[1]+1))>=-1,
  122. alpha=0.1, color='k')
  124. plt.subplot(1,3,1)
  125. plt.xlim(X_Lim_asm)
  126. plt.ylim(Y_Lim1)
  128. plt.legend()
  129. plt.ylabel(Y_Label)
  130. plt.xlabel("normalized ASM plasma concentration")
  132. plt.title("Three Example Patients")
  134. # Plot all individuals: ASM level vs Seizure duration with MELM ---------------
  135. plt.subplot(1,3,3)
  137. # Remove the group effect to all subjects
  138. FL = np.repeat(np.nan, stData.shape[0])
  139. for i in rnd_eff.keys():
  140. dt_dx = stData.patient_id == i
  141. if any(dt_dx):
  142. FL[dt_dx] = stData[rsp][dt_dx] - rnd_eff.get(i).Group
  143. stData[rsp] = FL
  145. # Plot the datapoints color codding the ASM_level [Normal-dose | Under-dose]
  146. sns.scatterplot(data = stData, x="ASM_lvl", y=rsp,
  147. hue = Cater, palette = palt, hue_order = Hue_Patt)
  149. # Generate regresion for all datapoints
  150. X = np.linspace(X_Lim_asm[0], X_Lim_asm[1])
  151. Y = int_eff
  152. Y += X * asm_eff * (asm_pvl < 0.05)
  153. sns.lineplot(x=X, y=Y, color = "red", label = "ASM lvl. - Fixed eff.")
  155. # Mark the normal-dose period
  156. plt.fill_between(range(X_Lim_asm[0],X_Lim_asm[1]+1), Y_Lim2[0], Y_Lim2[1],
  157. np.array(range(X_Lim_asm[0],X_Lim_asm[1]+1))>=-1,
  158. alpha=0.1, color='k')
  160. plt.xlim(X_Lim_asm)
  161. plt.ylim(Y_Lim2)
  162. plt.ylabel(Y_Label)
  163. plt.xlabel("normalized ASM plasma concentration")
  165. # Include sample size and ASM level fixed effect statistics
  166. plt.title("All Seizures (" +
  167. str(stData.shape[0])+ " Szs. | "+
  168. str(len(np.unique(stData.patient_id)))+ " Pts.)" +
  169. "\n MELM - ASM lvl. eff=" + str(round(asm_eff,Stt_Dgt)) +
  170. ", p=" + str(round(asm_pvl,Stt_Dgt)))
  172. # Plot seizure frecuancy changes between ASM level ranges on all subjects
  173. axs = plt.subplot(1,3,1)
  175. col_thr = 0
  176. for i in range(0,Stt_Freq_asm.shape[0]):
  177. # Colorcode the data based on difference in seizure frecuancy
  178. if np.diff(Stt_Freq_asm.values[i])>col_thr: # Increase -> in blue
  179. COL = 'blue'
  180. STL = ':'
  181. MRK = 'o'
  182. ALP = 1
  183. elif np.diff(Stt_Freq_asm.values[i])<-col_thr: # Decrease -> in red
  184. COL = 'red'
  185. STL = '-.'
  186. MRK = '^'
  187. ALP = 1
  188. else: # No Change -> in black
  189. COL = 'black'
  190. STL = '--'
  191. MRK = 'X'
  192. ALP = 1
  194. # Plot a line per subject
  195. # Square root of data for better visualization with Y^2 axis
  196. sns.lineplot(y=np.sqrt(Stt_Freq_asm.values[i][::-1]),x=["Under","Normo"],
  197. color=COL, alpha = ALP, legend = False, linestyle = STL,
  198. marker = MRK)
  200. xl = np.array([-0.5,0,0.5,1,1.5])
  201. yl = [0,7.5]
  203. # Indicate the Normal-dose data with grey background
  204. plt.fill_between(xl, yl[0], yl[1], xl>=0.5, alpha=0.1, color='k')
  205. plt.xlim([xl[0],xl[-1]])
  207. # Square Axis
  208. y_tick = np.linspace(yl[0],round(yl[1]),round(yl[1])+1)
  209. plt.yticks(y_tick, np.square(y_tick).astype(str))
  211. plt.ylabel("sz/day")
  212. plt.ylim(yl)
  214. # Statistical comparaison Comparaison
  215. Freq_SignRank = stats.wilcoxon(Stt_Freq_asm.Under,Stt_Freq_asm.Normo, method = 'asymptotic',
  216. alternative = 'greater')
  218. R_stat = Freq_SignRank.zstatistic/np.sqrt(Stt_Freq_asm.shape[0])
  220. # Include sample size and statistics
  221. plt.title("Sz. Frequency[" + str(Stt_Freq_asm.shape[0]) +
  222. "] \nSinged-Rank R=" + str(round(R_stat,Stt_Dgt)) +
  223. ", p="+ str(round(Freq_SignRank.pvalue,Stt_Dgt)))
  225. if save_fig:
  226. plt.savefig(path_save_fig + "/RLT_fDS_Durt_Freq.svg", format='svg',
  227. dpi=3000, bbox_inches='tight')
  228. plt.savefig(path_save_fig + "/PDF/RLT_fDS_Durt_Freq.pdf", format='pdf',
  229. dpi=3000, bbox_inches='tight')
  231. #%% Results: Seizure dutation, Seizure Staes and ASM levels
  233. # Full log10 duration for seizures within the Seizure Sates analysis
  234. rsp = "DSS_Full"
  236. # Figure Aperance variables
  237. Y_Lim1 = [0,3]
  238. Y_Lim2 = [0,3]
  239. Y_Label= "Duration (log10)"
  241. X_Lim_asm = [-10,1]
  242. X_Lim_tod = [0,24]
  244. palt = "mako"
  245. Hue_Patt = ["CMN_Normo","CMN_Under","UND_Under"]
  246. Axs_Patt = ["UND_Under","CMN_Under","CMN_Normo"]
  247. Cater = "Sz_Class"
  249. # MELM for for duation against ASM levels and TofD
  250. prd = "ASM_lvl + sTOD+cTOD + sTOD_h+cTOD_h"
  252. # Clear dataset from non-relevant data
  253. rm_dx = np.logical_or(np.isnan(sz_Clust[rsp]),np.isinf(sz_Clust[rsp]))
  254. stData = sz_Clust.drop(np.where(rm_dx)[0]).reset_index(drop=True)
  256. # Fit the mixed effect model with patietn identifier group-level effect
  257. mdl_Chr_SSS = smf.mixedlm(rsp + " ~ " + prd , stData, groups=stData.patient_id).fit()
  258. mdl_Chr_SSS.summary()
  260. # Relevent effect sizes and p-values for figure generation
  261. rnd_eff = mdl_Chr_SSS.random_effects
  262. int_eff = mdl_Chr_SSS.params.Intercept
  264. asm_eff = mdl_Chr_SSS.params.ASM_lvl
  265. asm_pvl = mdl_Chr_SSS.pvalues.ASM_lvl
  267. # Remove the group effect
  268. FL = np.repeat(np.nan, stData.shape[0])
  269. for i in rnd_eff.keys():
  270. dt_dx = stData.patient_id == i
  271. if any(dt_dx):
  272. FL[dt_dx] = stData[rsp][dt_dx] - rnd_eff.get(i).Group
  273. Data = stData.copy()
  274. stData[rsp] = FL
  276. plt.figure(figsize=[10,5])
  278. # Plot all individuals: ASM level vs Seizure duration with MELM ---------------
  279. plt.subplot (1,2,1)
  281. # Plot the datapoints color codding seizure tyeps and ASM state they appear in
  282. sns.scatterplot(data = stData, x="ASM_lvl", y=rsp,
  283. hue = Cater, palette = palt, hue_order = Hue_Patt)
  285. # Generate regresion for all datapoints
  286. X = np.linspace(X_Lim_asm[0], X_Lim_asm[1])
  287. Y = int_eff
  288. Y += X * asm_eff * (asm_pvl < 0.05)
  289. sns.lineplot(x=X, y=Y, color = "red", label = "ASM lvl. - Fixed eff.")
  291. # Mark the normal-dose period
  292. plt.fill_between(range(X_Lim_asm[0],X_Lim_asm[1]+1), Y_Lim2[0], Y_Lim2[1],
  293. np.array(range(X_Lim_asm[0],X_Lim_asm[1]+1))>=-1,
  294. alpha=0.1, color='k')
  296. plt.xlim(X_Lim_asm)
  297. plt.ylim(Y_Lim2)
  298. plt.ylabel(Y_Label)
  299. plt.xlabel("normalized ASM plasma concentration")
  301. # Include sample size and ASM level fixed effect statistics
  302. plt.title("All Seizures (" +
  303. str(stData.shape[0])+ " Szs. | "+
  304. str(len(np.unique(stData.patient_id)))+ " Pts.)" +
  305. "\n MELM - ASM lvl. eff=" + str(round(asm_eff,Stt_Dgt)) +
  306. ", p=" + str(round(asm_pvl,Stt_Dgt)))
  308. # Boxplot of seizure duration in patients with UND seizures -------------------
  309. plt.subplot (1,2,2)
  311. # Isolate individuals with any UND seizure
  312. rm_dx = np.zeros(stData.shape[0])
  313. for i in np.unique(stData.patient_id):
  314. idx = stData.patient_id == i
  315. sz_st = stData.SzSt[idx]
  316. if not any(sz_st == "UND"):
  317. rm_dx[idx] += 1
  318. data = stData.drop(np.where(rm_dx)[0]).reset_index(drop=True)
  320. # Alternative study: MELM to study Types and
  321. data_2 = Data.drop(np.where(rm_dx)[0]).reset_index(drop=True)
  322. data_2.Sz_Class[data_2.Sz_Class == "UND_Under"] = "0_UND_Under"
  323. data_2.Sz_Class[data_2.Sz_Class == "CMN_Under"] = "1_CMN_Under"
  324. data_2.Sz_Class[data_2.Sz_Class == "CMN_Normo"] = "2_CMN_Normo"
  325. mdl_Durt_TypAsm = smf.mixedlm(rsp + " ~ Sz_Class", data_2, groups=data_2.patient_id).fit()
  327. # Boxplot of distribution and data poitns segregatin by seizure type and ASM perion
  328. # Same color code and in previous plot
  329. sns.boxplot(data = data, x=Cater, y=rsp, order = Axs_Patt,
  330. hue = Cater, palette = palt, hue_order = Hue_Patt,
  331. fill=False, flierprops={"marker": ""})
  333. sns.stripplot(data = data, x=Cater, y=rsp, order = Axs_Patt,
  334. hue = Cater, palette = palt, hue_order = Hue_Patt,
  335. alpha = 0.7,jitter = 0.3)
  337. xl = np.linspace(-0.5,2.5,7)
  338. plt.fill_between(xl, Y_Lim2[0], Y_Lim2[1], xl>=1.5, alpha=0.1, color='k')
  339. plt.xlim([xl[0],xl[-1]])
  341. plt.ylim(Y_Lim2)
  343. plt.ylabel("Duration(log10)")
  344. plt.xlabel("")
  345. plt.xticks([])
  346. plt.title("Seizure Duration")
  348. # Compare the difference between CMN and UND seizure duration
  349. SSS_RankSum_UN = stats.ranksums (data[rsp][data.Sz_Class == "UND_Under"],
  350. data[rsp][data.Sz_Class == "CMN_Under"])
  351. N_UN = len(data[rsp][data.Sz_Class == "UND_Under"]) + len(data[rsp][data.Sz_Class == "CMN_Under"])
  352. R_stat_UN = SSS_RankSum_UN.statistic/np.sqrt(N_UN)
  354. # Compare the difference between CMN and UND seizure duration
  355. SSS_RankSum_CM = stats.ranksums (data[rsp][data.Sz_Class == "UND_Under"],
  356. data[rsp][data.Sz_Class == "CMN_Normo"])
  357. N_CM = len(data[rsp][data.Sz_Class == "UND_Under"]) + len(data[rsp][data.Sz_Class == "CMN_Normo"])
  358. R_stat_CM = SSS_RankSum_CM.statistic/np.sqrt(N_CM)
  360. # Statistics of sieuzre type comparaison
  361. plt.xlabel("UND vs CMN_Under - Rank Sum R=" + str(round(R_stat_UN,Stt_Dgt)) + ", p="+ str(round(SSS_RankSum_UN.pvalue,Stt_Dgt)) + '\n' +
  362. "UND vs CMN_Normo - Rank Sum R=" + str(round(R_stat_CM,Stt_Dgt)) + ", p="+ str(round(SSS_RankSum_CM.pvalue,Stt_Dgt)))
  364. # Include sample size
  365. plt.title("Individuals with UND seizures (" +
  366. str(data.shape[0])+ " Szs. | "+
  367. str(len(np.unique(data.patient_id)))+ " Pts.)")
  369. if save_fig:
  370. plt.savefig(path_save_fig + "/RLT_ssDS_Durt_SzSt.svg", format='svg',
  371. dpi=3000, bbox_inches='tight')
  372. plt.savefig(path_save_fig + "/PDF/RLT_ssDS_Durt_SzSt.pdf", format='pdf',
  373. dpi=3000, bbox_inches='tight')
  375. #%% Results: Seizures without Under-Dose Specific States
  377. # Full duration of patietns within Seizure States analysis
  378. rsp = "DSS_Full"
  380. Y_Lim1 = [0,3]
  381. Y_Lim2 = [0,3]
  382. Y_Label= "Duration (log10)"
  384. X_Lim_asm = [-10,1]
  385. X_Lim_tod = [0,24]
  387. palt = "mako"
  388. Hue_Patt = ["CMN_Normo","CMN_Under"]
  389. Axs_Patt = ["CMN_Under","CMN_Normo"]
  390. Cater = "Sz_Class"
  392. # MELM for for duation against ASM levels and TofD
  393. prd = "ASM_lvl + sTOD+cTOD + sTOD_h+cTOD_h"
  395. # Remove all UND sieuzres
  396. rm_dx = sz_Clust.SzSt == "UND"
  397. # Clear dataset from non-relevant data
  398. rm_dx_2 = np.logical_or(np.isnan(sz_Clust[rsp]),np.isinf(sz_Clust[rsp]))
  399. rm_dx = np.logical_or(rm_dx,rm_dx_2)
  401. stData = sz_Clust.drop(np.where(rm_dx)[0]).reset_index(drop=True)
  403. # Fit the mixed effect model with patietn identifier group-level effect
  404. mdl_Chr_CMN = smf.mixedlm(rsp + " ~ " + prd , stData, groups=stData.patient_id).fit()
  405. mdl_Chr_CMN.summary()
  407. # Relevent effect sizes and p-values for figure generation
  408. rnd_eff = mdl_Chr_CMN.random_effects
  409. int_eff = mdl_Chr_CMN.params.Intercept
  411. asm_eff = mdl_Chr_CMN.params.ASM_lvl
  412. asm_pvl = mdl_Chr_CMN.pvalues.ASM_lvl
  414. # Remove the group effect
  415. FL = np.repeat(np.nan, stData.shape[0])
  416. for i in rnd_eff.keys():
  417. dt_dx = stData.patient_id == i
  418. if any(dt_dx):
  419. FL[dt_dx] = stData[rsp][dt_dx] - rnd_eff.get(i).Group
  420. stData[rsp] = FL
  422. plt.figure(figsize=[10,5])
  424. # Plot CMN seizure duration vs ASM levels with MELM fixed effect --------------
  425. plt.subplot(1,2,1)
  427. # All data poitns collor codign the ASM period
  428. sns.scatterplot(data = stData, x="ASM_lvl", y=rsp,
  429. hue = Cater, palette = palt, hue_order = Hue_Patt)
  430. # Regresion form MELM
  431. X = np.linspace(X_Lim_asm[0], X_Lim_asm[1])
  432. Y = int_eff
  433. Y += X * asm_eff * (asm_pvl < 0.05)
  434. sns.lineplot(x=X, y=Y, color = "red", label = "ASM lvl. - Fixed eff.")
  436. # MArk the Mornal dose period
  437. plt.fill_between(range(X_Lim_asm[0],X_Lim_asm[1]+1), Y_Lim2[0], Y_Lim2[1],
  438. np.array(range(X_Lim_asm[0],X_Lim_asm[1]+1))>=-1,
  439. alpha=0.1, color='k')
  441. plt.xlim(X_Lim_asm)
  442. plt.ylim(Y_Lim2)
  443. plt.ylabel(Y_Label)
  444. plt.xlabel("normalized ASM plasma concentration")
  446. # OInclude sample size and MELM ASM level fixed-effect
  447. plt.title("CMN Seizures (" +
  448. str(stData.shape[0])+ " Szs. | "+
  449. str(len(np.unique(stData.patient_id)))+ " Pts.)" +
  450. "\n MELM - ASM lvl. eff=" + str(round(asm_eff,Stt_Dgt)) +
  451. ", p=" + str(round(asm_pvl,Stt_Dgt)))
  453. # Boxplot of the difference between CMN seizures during Normal and Under dose -
  454. plt.subplot(1,2,2)
  456. # Boxplot of distribution and data poitns segregatin by seizure type and ASM perion
  457. # Same color code and in previous plot
  458. sns.stripplot(data = stData, x=Cater, y=rsp, order = Axs_Patt,
  459. hue = Cater, palette = palt, hue_order = Hue_Patt,
  460. alpha = 0.7, jitter = 0.3)
  462. sns.boxplot(data = stData, x=Cater, y=rsp, order = Axs_Patt,
  463. hue = Cater, palette = palt, hue_order = Hue_Patt,
  464. fill=False, flierprops={"marker": ""})
  466. xl = np.array([-0.5,0,0.5,1,1.5])
  468. # Mark dats corresponding to Normal-dose ASM period
  469. plt.fill_between(xl, Y_Lim2[0], Y_Lim2[1], xl>=0.5, alpha=0.1, color='k')
  470. plt.xlim([xl[0],xl[-1]])
  472. plt.xlabel("")
  473. plt.xticks([])
  475. plt.ylim(Y_Lim2)
  476. plt.ylabel(Y_Label)
  478. # Statistical analysis of Normal-dose vs Under-dose CMN seizure duration
  479. CMN_RankSum = stats.ranksums (stData[rsp][stData.ASM_stt == "Under"],
  480. stData[rsp][stData.ASM_stt == "Normo"])
  481. N = len(stData[rsp][stData.ASM_stt == "Under"]) + len(stData[rsp][stData.ASM_stt == "Normo"])
  482. R_stat = CMN_RankSum.statistic/np.sqrt(N)
  484. # Include statistical results
  485. plt.xlabel("Under vs Normal - Rank Sum R=" + str(round(R_stat,Stt_Dgt)) +
  486. ", p="+ str(round(CMN_RankSum.pvalue,Stt_Dgt)))
  488. # Include sample size
  489. plt.title("CMN Seizures (" +
  490. str(stData.shape[0])+ " Szs. | "+
  491. str(len(np.unique(stData.patient_id)))+ " Pts.)")
  493. if save_fig:
  494. plt.savefig(path_save_fig + "/RLT_cmDS_Durt.svg", format='svg',
  495. dpi=3000, bbox_inches='tight')
  496. plt.savefig(path_save_fig + "/PDF/RLT_cmDS_Durt.pdf", format='pdf',
  497. dpi=3000, bbox_inches='tight')
  499. #%% Supplementary: Under-Dose States Characterization
  500. plt.figure(figsize=[10,5])
  501. # Identify UND seizures
  502. idx = sz_Clust.SzSt == "UND"
  504. # Part 1: Sequence and relative duration of Under-dose Staes ------------------
  505. plt.subplot(1,2,1)
  506. ax = plt.gca()
  507. ax.grid(axis='x',color='k', linestyle=':', linewidth=0.5)
  508. xl = np.array([-20,0,20])
  509. yl = [-0.5,2.5]
  511. # Obtian distribution of Underdose/Common Sates on seizures
  512. # 0.8 corresponds to non present state (n.p.)
  513. pop_hist_seq = pd.DataFrame({"Seq": ["n.p.", "1st", "2nd"],
  514. "Under-dose Stt.": [-sum(sz_Clust.SqSS_Uniq[idx] == 0.8),
  515. -sum(sz_Clust.SqSS_Uniq[idx] == 1),
  516. -sum(sz_Clust.SqSS_Uniq[idx] == 2)],
  517. "Common Stt.": [sum(sz_Clust.SqSS_Comn[idx] == 0.8),
  518. sum(sz_Clust.SqSS_Comn[idx] == 1),
  519. sum(sz_Clust.SqSS_Comn[idx] == 2)]})
  521. # Barplots for Sequence distribution on each ASM periods
  522. sns.barplot(x='Under-dose Stt.', y='Seq', data=pop_hist_seq, order=["2nd", "1st", "n.p."],
  523. orient='h', color="#EB6F3E", lw=0)#, label = 'Under-dose Stt.')
  525. sns.barplot(x='Common Stt.', y='Seq', data=pop_hist_seq, order=["2nd", "1st", "n.p."],
  526. orient='h', color="#531E77", lw=0)#, label = 'Common Stt.')
  528. #plt.legend()
  529. plt.ylabel("sequence(index)")
  530. plt.xlabel("counts(szs.)")
  532. # Mark the Normal-dose related data
  533. plt.fill_between(xl, yl[0], yl[1], xl>=0, alpha=0.1, color='k')
  535. plt.xlim([xl[0],xl[-1]])
  536. plt.ylim([yl[-1],yl[0]])
  538. x_ticks = np.linspace(xl[0], xl[-1],int((xl[-1]-xl[0])/5+1))
  539. plt.xticks(x_ticks,np.abs(x_ticks.astype(int)).astype(str))
  541. plt.title("Seizure States First Appearance")
  543. # Part 2: Relative duration of Under-dose Staes -------------------------------
  544. plt.subplot(1,2,2)
  546. # Obtain the relative duration of Under-dose states
  547. Data = 10**sz_Clust.DSS_Uniq / 10**sz_Clust.DSS_Full * 100
  549. # Boxplot of distribution and data p[oints]
  550. sns.boxplot(y=Data[idx],color = "#EB6F3E",fill=False, flierprops={"marker": ""})
  551. sns.stripplot(y=Data[idx],alpha = 0.7,jitter = 0.3,color = "#EB6F3E", edgecolor='k')
  553. plt.ylim([0,100])
  554. plt.ylabel("Duration (%)")
  556. # Median of the proporstion
  557. plt.xlabel("median proportion = " +str(round(Data[idx].median(),3))+ "%")
  558. plt.title("Tapered-emergent states relative duration" )
  560. if save_fig:
  561. plt.savefig(path_save_fig + "/SUP_UND_Seq_DurCont.svg", format='svg',
  562. dpi=3000, bbox_inches='tight')
  563. plt.savefig(path_save_fig + "/PDF/SUP_UND_Seq_DurCont.pdf", format='pdf',
  564. dpi=3000, bbox_inches='tight')
  566. #%% Supplementary: Common States Characterization
  567. plt.figure(figsize=[12,5])
  569. # Part 1: Number of states within of Common States ----------------------------
  570. plt.subplot(1,2,1)
  571. ax = plt.gca()
  572. ax.grid(axis='x',color='k', linestyle=':', linewidth=0.5)
  574. # Identify CMN seizures with relevant data
  575. rm_dx = sz_Clust.SzSt == "UND"
  576. rm_dx_2 = np.logical_or(np.isnan(sz_Clust.DSS_Full),np.isinf(sz_Clust.DSS_Full))
  577. rm_dx = np.logical_or(rm_dx,rm_dx_2)
  578. Data = sz_Clust.drop(np.where(rm_dx)[0]).reset_index(drop = True)
  580. # Obatin the distribution of the number of states of each seizure
  581. nss = np.unique(Data.NSS_Comn)[::-1]
  582. n_nss = np.zeros(nss.shape)
  583. u_nss = np.zeros(nss.shape)
  584. for i in range(0,len(nss)):
  585. n_nss[int(i)] = sum(Data.NSS_Comn[Data.ASM_stt == "Normo"] == nss[i])
  586. u_nss[int(i)] = sum(Data.NSS_Comn[Data.ASM_stt == "Under"] == nss[i])
  587. pop_hist_nss = pd.DataFrame({"N. of States": nss,
  588. "Under-dose N. Stt.":-u_nss,
  589. "Normal-dose N. Stt.":n_nss})
  591. # Barplots for Number of States distribution on each ASM periods
  592. sns.barplot(x='Under-dose N. Stt.', y='N. of States', data=pop_hist_nss,
  593. orient='h', color="#1B86AA", lw=0)#, label = 'Under-dose N. Stt.')
  595. sns.barplot(x='Normal-dose N. Stt.', y='N. of States', data=pop_hist_nss,
  596. orient='h', color="#483B73", lw=0)#, label = 'Normal-dose N. Stt.')
  598. plt.ylabel("N. of States")
  599. plt.xlabel("counts(szs.)")
  601. xl = np.array([-45,0,45])
  602. yl = [-0.5,len(nss)-0.5]
  603. plt.fill_between(xl, yl[0], yl[1], xl>=0, alpha=0.1, color='k')
  605. plt.xlim([xl[0],xl[-1]])
  606. plt.ylim([yl[0],yl[-1]])
  608. x_ticks = np.linspace(-40, 40,9)
  609. plt.xticks(x_ticks,np.abs(x_ticks.astype(int)).astype(str))
  611. plt.title("N. of States in seizures(" +
  612. str(Data.shape[0])+ " Szs. | "+
  613. str(len(np.unique(Data.patient_id)))+ " Pts.)")
  615. # Part 2: Contribution to length of states composing Common States ------------
  616. plt.subplot(1,2,2)
  618. # Identifu the contibution of each state to full duration
  619. Full_Seg_L_CrCf = []
  620. Full_Seg_L_CrpV = []
  622. I = np.unique(Data.patient_id)
  623. for Show_id in I:
  624. # Isolate patient data
  625. idx = np.where(Data.patient_id == Show_id)[0]
  626. SSS = [np.array(Data.sss_dx[i].split()).astype(int) for i in idx]
  628. # All states
  629. st_dx = np.unique(np.hstack(SSS))
  631. # Full duration of each seizure
  632. f_Lgt = np.array([len(s) for s in SSS])
  634. # Repository for Speramn Correlation
  635. cr_cf = np.zeros(len(st_dx)) #coefficient
  636. cr_pv = np.zeros(len(st_dx)) #pvalues
  637. cnt = 0
  638. for dx in st_dx:
  640. # Duration of the state on eahc seizure
  641. p_Lgt = np.array([sum(s==dx) for s in SSS])
  643. # Relationship between Full and Patial diration
  644. spearman = stats.spearmanr(f_Lgt,p_Lgt)
  645. cr_cf[cnt] = spearman.statistic
  646. # p-value as significance level [0:3]
  647. cr_pv[cnt] = sum([spearman.pvalue<0.05,
  648. spearman.pvalue<0.01,
  649. spearman.pvalue<0.001])
  650. cnt += 1
  652. Full_Seg_L_CrCf.append(cr_cf)
  653. Full_Seg_L_CrpV.append(cr_pv)
  655. # Scatterplot of Coefficients for each patietns color coding the signifiance
  656. X = np.hstack([np.repeat(I[i], len(Full_Seg_L_CrCf[i])) for i in range(0,len(I))])
  657. Y = np.hstack(Full_Seg_L_CrCf)
  658. H = np.hstack(Full_Seg_L_CrpV)
  659. asdf = sns.scatterplot(x = X.astype(str), y = Y, hue = H,
  660. hue_norm = (0,4), palette="viridis_r")
  662. plt.axhline(0,0,1,color='k', linestyle=":")
  664. plt.title("Full Duration vs Sates Duration")
  665. plt.ylabel("Spearman's Correlation Coefficient")
  667. plt.legend(asdf,title='p-value', loc='lower left', labels=['','n.s.', '<0.05','<0.01','<0.001'])
  669. x_ticks = ["id"+str(i) for i in I]
  670. plt.xticks(np.linspace(0,len(I)-1,len(I)),x_ticks,rotation=30)
  672. if save_fig:
  673. plt.savefig(path_save_fig + "/SUP_CMN_nSST_Corr.svg", format='svg', dpi=3000, bbox_inches='tight')
  674. plt.savefig(path_save_fig + "/PDF/SUP_CMN_nSST_Corr.pdf", format='pdf', dpi=3000, bbox_inches='tight')
  676. #%% Supplementary: Effect of ASM types
  678. # Get the full dataset seizure duration
  679. rsp = "Duration_10"
  680. rm_dx = np.logical_or(np.isnan(sz_Clust[rsp]),np.isinf(sz_Clust[rsp]))
  681. stData = sz_Clust.drop(np.where(rm_dx)[0]).reset_index(drop=True)
  683. plt.figure(figsize=[10,5])
  685. palt = "mako"
  686. Hue_Patt = ["CMN_Normo","CMN_Under","UND_Under"]
  688. # ASM types on the dataset
  689. # Type of ASM related to the target
  690. # GB -> Gabba receptor
  691. # SC -> Sodium Channels
  692. # SV -> SVA1 receptor
  693. # ML -> Multitarget
  694. # OT -> Other medication (low representation)
  695. drgt = np.array(['GB', 'ML', 'OT', 'SC', 'SV'])
  697. # Usage of each ASM type across the dataset -----------------------------------
  698. plt.subplot(1,2,1)
  700. # Number of patietns with ASM type tapered
  701. sbj_n = np.zeros(len(drgt))
  702. for i in range(0,len(drgt)):
  703. idx = stData[drgt[i]] == 1
  704. sbj_n[i] = len(np.unique(stData.patient_id[idx]))
  705. sns.barplot(x=drgt,y=sbj_n)
  706. plt.title("ASM type used on tapering")
  708. plt.ylim([0,len(np.unique(stData.patient_id))])
  709. plt.ylabel("N. of Individuals")
  710. plt.yticks(range(0, 29, 2))
  711. plt.grid(visible=True,axis="y")
  713. # Seizure duration segregates by ASM type -------------------------------------
  714. plt.subplot(1,2,2)
  716. for i in range(0,len(drgt)):
  717. # Identify patients with ASM type tapred
  718. idx = stData[drgt[i]] == 1
  720. # Distribution(boxplot) and datapoints fo all seiuzres/patietns identified
  721. # Color code seizure type
  722. sns.stripplot(x=i, y=stData[rsp][idx],legend=False, hue=stData.Sz_Class[idx],
  723. palette=palt, hue_order=Hue_Patt)
  724. sns.boxplot(x=i, y=stData[rsp][idx], fill=False, flierprops={"marker": ""}, color='k')
  726. plt.xticks(range(0,len(drgt)),drgt)
  727. plt.ylim([0,3.5])
  728. plt.ylabel("Duration (log10)")
  729. plt.title("Seizure duration by tapered ASM type")
  731. if save_fig:
  732. plt.savefig(path_save_fig + "/SUP_ASM_Type.svg", format='svg', dpi=3000, bbox_inches='tight')
  733. plt.savefig(path_save_fig + "/PDF/SUP_ASM_Type.pdf", format='pdf', dpi=3000, bbox_inches='tight')
  735. #%% Supplementary: Seizure dutation withouth Cyrc / Ultra dian effect
  736. # Repica of 2nd resutls with changed MELM formula
  737. # Fixed effects only related to ASM levels (No ToD variables included)
  738. prd ="ASM_lvl"
  740. rsp = "DSS_Full"
  742. Y_Lim1 = [0,3]
  743. Y_Lim2 = [0,3]
  744. Y_Label= "Duration (log10)"
  746. X_Lim_asm = [-10,1]
  747. X_Lim_tod = [0,24]
  749. palt = "mako"
  750. Hue_Patt = ["CMN_Normo","CMN_Under","UND_Under"]
  751. Axs_Patt = ["UND_Under","CMN_Under","CMN_Normo"]
  752. Cater = "Sz_Class"
  754. # Data of relevance
  755. rm_dx = np.logical_or(np.isnan(sz_Clust[rsp]),np.isinf(sz_Clust[rsp]))
  756. stData = sz_Clust.drop(np.where(rm_dx)[0]).reset_index(drop=True)
  758. # Fit the MELM
  759. mdl_Chr_Ams_nCrn = smf.mixedlm(rsp + " ~ " + prd , stData, groups=stData.patient_id).fit()
  760. mdl_Chr_Ams_nCrn.summary()
  762. rnd_eff = mdl_Chr_Ams_nCrn.random_effects
  763. int_eff = mdl_Chr_Ams_nCrn.params.Intercept
  765. asm_eff = mdl_Chr_Ams_nCrn.params.ASM_lvl
  766. asm_pvl = mdl_Chr_Ams_nCrn.pvalues.ASM_lvl
  768. # Remove Patient-level group effect
  769. FL = np.repeat(np.nan, stData.shape[0])
  770. for i in rnd_eff.keys():
  771. dt_dx = stData.patient_id == i
  772. if any(dt_dx):
  773. FL[dt_dx] = stData[rsp][dt_dx] - rnd_eff.get(i).Group
  774. Data = stData.copy()
  775. stData[rsp] = FL
  777. # Plt seizure duration vs ASM level with MELM ASM level fixed effect
  778. plt.figure(figsize=[5,5])
  780. sns.scatterplot(data = stData, x="ASM_lvl", y=rsp,
  781. hue = Cater, palette = palt, hue_order = Hue_Patt, legend = False)
  782. X = np.linspace(X_Lim_asm[0], X_Lim_asm[1])
  783. Y = int_eff
  784. Y += X * asm_eff * (asm_pvl < 0.05)
  785. sns.lineplot(x=X, y=Y, color = "red", legend = False)#, label = "ASM lvl. - Fixed eff.")
  786. plt.fill_between(range(X_Lim_asm[0],X_Lim_asm[1]+1), Y_Lim2[0], Y_Lim2[1],
  787. np.array(range(X_Lim_asm[0],X_Lim_asm[1]+1))>=-1,
  788. alpha=0.1, color='k')
  789. plt.xlim(X_Lim_asm)
  790. plt.ylim(Y_Lim2)
  791. plt.ylabel(Y_Label)
  792. plt.xlabel("normalized ASM plasma concentration")
  793. plt.title("All Seizures (" +
  794. str(stData.shape[0])+ " Szs. | "+
  795. str(len(np.unique(stData.patient_id)))+ " Pts.)" +
  796. "\n MELM - ASM lvl. eff=" + str(round(asm_eff,3)) + ", p=" + str(round(asm_pvl,3)))
  798. if save_fig:
  799. plt.savefig(path_save_fig + "/SUP_Durt_SzSt_nCrn.svg", format='svg',
  800. dpi=3000, bbox_inches='tight')
  801. plt.savefig(path_save_fig + "/PDF/SUP_Durt_SzSt_nCrn.pdf", format='pdf',
  802. dpi=3000, bbox_inches='tight')

gmb_SNS_ASM_Dur_V8.py at commit 05e64c0, no license · at the source

Overview

Authors: Guillermo M Besné1, Emmanuel Molefi1, Billy Smith1, Nathan Evans1, Sarah J Gascoigne1, Chris Thornton1, Fahmida A Chowdhury2, Beate Diehl2, John S Duncan2, Andrew W McEvoy2, Anna Miserocchi2, Jane de Tisi2, Matthew C Walker2, Peter N Taylor1,2,3, Yujiang Wang1,2,3
ORCID iDs: Emmanuel Molefi, Nathan Evans, Chris Thornton, John S Duncan, Jane de Tisi, Matthew C Walker, Peter N Taylor, Yujiang Wang
  1. CNNP Lab (www.cnnp-lab.com), School of Computing, Newcastle University, Newcastle upon Tyne NE4 5TG, United Kingdom
  2. Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology, University College London, London WC1N 3BG, United Kingdom
  3. Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne NE1 7RU, United Kingdom
Institutions: Newcastle University (United Kingdom); UCL Queen Square Institute of Neurology (United Kingdom); University College London (United Kingdom)
Journal: Brain communications, volume 8, issue 2, article fcag088
Dates: received 17 June 2025; accepted 12 March 2026; published online 16 March 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1093/braincomms/fcag088 · PMID 41907313 · PMCID PMC13023363 · OpenAlex W7138212801
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: human (organism), epilepsy (population)
Methods: Spectral & time-frequency, Preprocessing, Connectivity, Statistics, Machine learning, fMRI & imaging, Smoothing, state filtering, decompositions
Keywords: neurophysiology, dynamic states, seizure states, seizure propagation, bifurcation
Topic: Epilepsy research and treatment (Psychiatry and Mental health, Medicine), according to OpenAlex
Funding: UK Research and Innovation Future (MR/T04294X/1, MR/V026569/1); National Institute for Health; Care Research University College London Hospitals; University College London Biomedical Research Centre; Engineering and Physical Sciences Research Council
Citations: not cited yet (Europe PMC); 38 references in the paper

Abstract

Anti-seizure medications (ASMs) can reduce seizure duration, but their precise modes of action are unclear. Specifically, it is unknown whether ASMs shorten seizures by curtailing existing seizure activity early or by selectively suppressing certain seizure activity patterns from emerging. We retrospectively analysed intracranial EEG (iEEG) recordings of 457 seizures from 28 people with epilepsy undergoing ASM tapering. Beyond measuring seizure occurrence and duration, we categorized distinct seizure propagation activity patterns (states) based on spatial and frequency power characteristics and related these to different ASM levels. We found that reducing ASM levels led to increased seizure frequency (r = 0.87, P < 0.001) and longer seizure duration (β = −0.033, P < 0.001), consistent with prior research. Further analysis revealed two distinct mechanisms in which seizures became prolonged: Emergence of new seizure propagation patterns—In ∼40% of patients, ASM tapering unmasked additional seizure activity states, and seizures containing these ‘taper-emergent states’ were substantially longer (r = 0.49, P < 0.001). Prolongation of existing seizure patterns—Even in seizures without taper-emergent states, lower ASM levels still resulted in ∼12–224% longer durations depending on the ASM dosage and tapering (β = −0.049, P < 0.001). ASMs influence seizures through two mechanisms: they (i) suppress specific seizure propagation patterns (states) in an all-or-nothing fashion and (ii) curtail the duration of other seizure patterns. These findings highlight the complex role of ASMs in seizure modulation and could inform personalized dosing strategies for epilepsy management. These findings may also have implications in understanding the effects of ASMs on cognition and mood.

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

Repository

Its files are read in the Code ↔ Paper reader above.

cnnp-lab/2025_ASM-SzState_GMB

License: none: the authors keep all their rights
State: the link answers, verified on 30 September 2026
Evidence: files inventoried
Commit: 05e64c0ab2822c31b81574213c44c6d4d45c04d7, 10 March 2025
Languages: Python (1)
Size: 23 files, 1 script
Software Heritage: not archived
Found in: “Data availability”
Holds: README
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: Matplotlib (1 file), NumPy (1 file), pandas (1 file), SciPy (1 file), seaborn (1 file), statsmodels (1 file)
Availability: 1 check, the latest on 30 September 2026: the link answers
  • 30 September 2026: the link answers
2 files
At the source: github.com/cnnp-lab/2025_ASM-SzState_GMB

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 1 script, each with its path and the digest of its content;
  • no match between paragraphs and code yet;
  • 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

Anonymized seizure state data and ASM intake schedule, along with analysis code, is available on GitHub: https://github.com/cnnp-lab/2025_ASM-SzState_GMB.

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, 15 authors, 5 keywords, 5 funders, 35 references.

Cite

This paper

Besné, G. M., Molefi, E., Smith, B., Evans, N., Gascoigne, S. J., Thornton, C., Chowdhury, F. A., Diehl, B., Duncan, J. S., McEvoy, A. W., Miserocchi, A., de Tisi, J., Walker, M. C., Taylor, P. N., & Wang, Y. (2026). Dual mechanism of anti-seizure medications in controlling seizure activity. Brain communications, 8(2), fcag088. https://doi.org/10.1093/braincomms/fcag088

BibTeX

@article{besne2026dual,
author = {Besné, Guillermo M and Molefi, Emmanuel and Smith, Billy and Evans, Nathan and Gascoigne, Sarah J and Thornton, Chris and Chowdhury, Fahmida A and Diehl, Beate and Duncan, John S and McEvoy, Andrew W and Miserocchi, Anna and de Tisi, Jane and Walker, Matthew C and Taylor, Peter N and Wang, Yujiang},
title = {{Dual mechanism of anti-seizure medications in controlling seizure activity}},
journal = {Brain communications},
year = {2026},
month = mar,
volume = {8},
number = {2},
pages = {fcag088},
publisher = {Oxford University Press},
issn = {2632-1297},
doi = {10.1093/braincomms/fcag088},
url = {https://doi.org/10.1093/braincomms/fcag088},
pmid = {41907313},
pmcid = {PMC13023363}
}

RIS

TY - JOUR
AU - Besné, Guillermo M
AU - Molefi, Emmanuel
AU - Smith, Billy
AU - Evans, Nathan
AU - Gascoigne, Sarah J
AU - Thornton, Chris
AU - Chowdhury, Fahmida A
AU - Diehl, Beate
AU - Duncan, John S
AU - McEvoy, Andrew W
AU - Miserocchi, Anna
AU - de Tisi, Jane
AU - Walker, Matthew C
AU - Taylor, Peter N
AU - Wang, Yujiang
TI - Dual mechanism of anti-seizure medications in controlling seizure activity
T2 - Brain communications
J2 - Brain Commun
PY - 2026
DA - 2026/03/16
VL - 8
IS - 2
SP - fcag088
SN - 2632-1297
PB - Oxford University Press
DO - 10.1093/braincomms/fcag088
UR - https://doi.org/10.1093/braincomms/fcag088
LA - en
ER -

CSL-JSON

{
"id": "10.1093/braincomms/fcag088",
"type": "article-journal",
"title": "Dual mechanism of anti-seizure medications in controlling seizure activity",
"container-title": "Brain communications",
"author": [
{
"family": "Besné",
"given": "Guillermo M"
},
{
"family": "Molefi",
"given": "Emmanuel"
},
{
"family": "Smith",
"given": "Billy"
},
{
"family": "Evans",
"given": "Nathan"
},
{
"family": "Gascoigne",
"given": "Sarah J"
},
{
"family": "Thornton",
"given": "Chris"
},
{
"family": "Chowdhury",
"given": "Fahmida A"
},
{
"family": "Diehl",
"given": "Beate"
},
{
"family": "Duncan",
"given": "John S"
},
{
"family": "McEvoy",
"given": "Andrew W"
},
{
"family": "Miserocchi",
"given": "Anna"
},
{
"family": "de Tisi",
"given": "Jane"
},
{
"family": "Walker",
"given": "Matthew C"
},
{
"family": "Taylor",
"given": "Peter N"
},
{
"family": "Wang",
"given": "Yujiang"
}
],
"container-title-short": "Brain Commun",
"volume": "8",
"issue": "2",
"page": "fcag088",
"DOI": "10.1093/braincomms/fcag088",
"PMID": "41907313",
"PMCID": "PMC13023363",
"ISSN": "2632-1297",
"publisher": "Oxford University Press",
"URL": "https://doi.org/10.1093/braincomms/fcag088",
"language": "en",
"issued": {
"date-parts": [
[
2026,
3,
16
]
]
}
}

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:10.64898/2026.08.18.26360725 [code]
Temporal pole blurring in hippocampal sclerosis reflects seizure-disrupted myelination
Journal: medRxiv (preprint)
In common: statsmodels, seaborn, pandas, 3 other tools, epilepsy, author John S. Duncan
[2] doi:10.1093/braincomms/fcag120 [code]
The heartbeat evoked potential and the prediction of functional seizure semiology.
Journal: Brain communications
In common: pandas, SciPy, Matplotlib, 1 other tool, epilepsy, author Matthew C Walker
[3] doi:10.1038/s41598-026-44063-w [code]
Neuronal silence as a predictive biomarker and target for epileptic seizures suppression.
Journal: Scientific reports
In common: statsmodels, seaborn, pandas, 3 other tools, epilepsy, 1 reference
[4] doi:10.1038/s41593-026-02258-4 [code]
Laminar organization of cellular microcircuits modulating human interictal epileptiform discharges.
Journal: Nature neuroscience
In common: statsmodels, seaborn, pandas, 3 other tools, epilepsy, 1 reference
[5] doi:10.1038/s41598-026-43151-1 [code]
Stabilizing fractional dynamical networks suppresses epileptic seizures.
Journal: Scientific reports
In common: statsmodels, seaborn, pandas, 3 other tools, epilepsy, 1 reference
[6] doi:10.1016/j.ebiom.2026.106375 [code]
Brainwaves under medication: revealing class-specific neural signatures of psychotropic medication from 24,000 EEGs.
Journal: EBioMedicine
In common: statsmodels, seaborn, pandas, 3 other tools, 1 reference
[7] doi:10.1002/epi4.70336 [code]
Quantifying extra-lesional interhemispheric cortical asymmetry in focal cortical dysplasia type II.
Journal: Epilepsia open
In common: statsmodels, seaborn, pandas, 3 other tools, epilepsy, 1 reference
[8] doi:10.1007/s00234-026-04103-8 [code]
Enhanced detection of subtle cortical abnormalities in focal epilepsy using 7 T MRI surface-based models and graph neural networks.
Journal: Neuroradiology
In common: statsmodels, seaborn, pandas, 3 other tools, epilepsy, 1 reference
[9] doi:10.1038/s41467-026-75959-w [code]
Charting higher-order models of brain function beyond pairwise interactions.
Journal: Nature communications
In common: statsmodels, seaborn, pandas, 3 other tools, 1 reference
[10] doi:10.1093/brain/awaf317 [code]
High-resolution electrophysiological mapping of effective connectivity of lateral prefrontal cortex.
Journal: Brain : a journal of neurology
In common: statsmodels, pandas, SciPy, 2 other tools, epilepsy, 1 reference

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.