{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7e302b3b",
   "metadata": {},
   "source": [
    "# Null models and permutation testing\n",
    "\n",
    "We touched on null models briefly in [Notebook 1](intro01_spatial_colocalizations.ipynb) (conceptual background) and [Notebook 2](intro02_nispace_intro.ipynb) (where we called `permute(\"maps\")` without much explanation). This notebook explains what's happening under the hood and when to use which approach.\n",
    "\n",
    "There are two fundamentally different situations that call for different null models:\n",
    "\n",
    "1. **You have a single group-level map** (e.g., the pain NeuroQuery map, a published effect size): permute the *reference maps*.\n",
    "2. **You have individual subject data and a group design** (e.g., the anorexia nervosa dataset): permute the *group labels*.\n",
    "\n",
    "For X-Set Enrichment Analysis, there's a third type — we'll cover that in [Notebook 10](intro10_xsea.ipynb)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7464c77d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-01T13:03:33.198593Z",
     "iopub.status.busy": "2026-06-01T13:03:33.198418Z",
     "iopub.status.idle": "2026-06-01T13:03:33.832936Z",
     "shell.execute_reply": "2026-06-01T13:03:33.832521Z"
    }
   },
   "outputs": [],
   "source": [
    "import tqdm.notebook\n",
    "tqdm.notebook.tqdm = tqdm.tqdm\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6aecac40",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-01T13:03:33.834846Z",
     "iopub.status.busy": "2026-06-01T13:03:33.834681Z",
     "iopub.status.idle": "2026-06-01T13:03:42.328878Z",
     "shell.execute_reply": "2026-06-01T13:03:42.328485Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.datasets: Loading pet maps.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.datasets: Loading integrated collection 'UniqueTracers' for dataset 'pet'.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.datasets: Filtering maps by collection.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.datasets: Loading data parcellated with 'Schaefer200Parcels7Networks'\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.api: *** NiSpace.fit() - Data extraction and preparation. ***\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.core.parcellation: Building multi-space Parcellation for 'Schaefer200Parcels7Networks' from library.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.core.parcellation: Available spaces: MNI152NLin2009cAsym, MNI152NLin6Asym, fsaverage, fsLR\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.core.parcellation: Parcellation 'Schaefer200Parcels7Networks': validation passed.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.core.parcellation: Lazy-loading parcellation image for space 'MNI152NLin2009cAsym'.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.core.parcellation: Parcellation 'Schaefer200Parcels7Networks': active space set to 'MNI152NLin2009cAsym'.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.api: Checking input data for 'x' (should be, e.g., PET data):\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.io: Input type: DataFrame, assuming parcellated data with shape (n_files/subjects/etc, n_parcels).\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mWARNING | 01/06/26 15:03:35 | nispace.io: Parcellated data contains nan values!\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.api: Got 'x' data for 29 x 200 parcels.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.api: Checking input data for 'y' (should be, e.g., subject data):\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:35 | nispace.io: Input type: list, assuming imaging data.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:36 | nispace.io: Background (bg) handling: ignoring bg: True (bg value: ['auto', 0.0]); dropping bg parcels: False\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:36 | nispace.io: Parcellating imaging data.\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Parcellating (4 proc):   0%|                                                                                                                                     | 0/1 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Parcellating (4 proc): 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 121.11it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:39 | nispace.api: Got 'y' data for 1 x 200 parcels.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:39 | nispace.api: Z-standardizing 'X' data.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:39 | nispace.api: *** NiSpace.colocalize() - Estimating X & Y colocalizations. ***\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:39 | nispace.api: Running 'spearman' colocalization.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:39 | nispace.api: Pre-ranking X and Y data.\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Colocalizing (spearman, 4 proc):   0%|                                                                                                                           | 0/1 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Colocalizing (spearman, 4 proc): 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 1671.70it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<nispace.api.NiSpace at 0x14ae67e80>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from nispace.datasets import fetch_reference, fetch_example\n",
    "from nispace.io import load_img\n",
    "from nispace.api import NiSpace\n",
    "\n",
    "# set up with the pain map\n",
    "pet_maps = fetch_reference(\"pet\", parcellation=\"Schaefer200\",\n",
    "                           collection=\"UniqueTracers\", print_references=False)\n",
    "pain_map = load_img(\"neuroquery/pain.nii.gz\")\n",
    "\n",
    "nsp = NiSpace(x=pet_maps, y=pain_map, y_labels=\"Pain\",\n",
    "              parcellation=\"Schaefer200\", seed=42, n_proc=4)\n",
    "nsp.fit()\n",
    "nsp.colocalize(\"spearman\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7ca07bb",
   "metadata": {},
   "source": [
    "## Part 1: Surrogate map null models\n",
    "\n",
    "When you call `permute(\"maps\")`, NiSpace generates spatially-constrained random versions of the reference maps and recomputes the correlation with your input map for each permutation. The empirical distribution of these null correlations becomes the null distribution.\n",
    "\n",
    "The null method is **auto-selected** based on the parcellation type — you rarely need to specify it explicitly:\n",
    "\n",
    "- **Cortical-only parcellations** with a surface space (fsLR, fsaverage): spin test (`\"alexander_bloch\"`) is chosen automatically\n",
    "- **Combined (cortex + subcortex) or volumetric-only parcellations**: Moran spectral randomization (`\"moran\"`) is chosen automatically\n",
    "\n",
    "Three methods are available in total (more are continuously added):\n",
    "\n",
    "| Method | Key idea | When selected |\n",
    "|--------|----------|---------------|\n",
    "| `\"alexander_bloch\"` (spin) | Random rotations on the cortical sphere, automatically retaining interhemispheric symmetry to a large extent | **Auto-default** for cortical parcellations (e.g., Schaefer200) |\n",
    "| `\"moran\"` (Moran spectral randomization) | Randomize spectral coefficients of the spatial covariance, null map symmetry will depend on distance matrix symmetry | **Auto-default** for combined/volumetric parcellations |\n",
    "| `\"burt2020\"` (variogram) | Match the spatial variogram of the original map, null map symmetry will depend on distance matrix symmetry | Explicit choice; any parcellation |\n",
    "\n",
    "Schaefer200 is a cortical-only atlas with fsLR available, so the spin test is selected automatically when we omit `null_method`.\n",
    "\n",
    "### The spin test (alexander_bloch)\n",
    "\n",
    "The spin test (Alexander-Bloch et al., 2018) works by projecting the cortical parcellation onto a sphere and applying random rotations. This preserves the spatial topology of the original map while breaking any specific spatial relationship with the input map."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a37c5e40",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-01T13:03:42.330537Z",
     "iopub.status.busy": "2026-06-01T13:03:42.330380Z",
     "iopub.status.idle": "2026-06-01T13:03:46.323537Z",
     "shell.execute_reply": "2026-06-01T13:03:46.323116Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.api: *** NiSpace.permute() - Estimate exact non-parametric p values. ***\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.api: Permutation of: X maps.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.api: Using default null method 'alexander_bloch' (parcellation null space: 'fsLR').\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.core.parcellation: Lazy-loading parcellation image for space 'fsLR'.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.api: Loading observed colocalizations (method = 'spearman').\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.core.permute: No null maps found.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.core.permute: Generating null maps (n = 1000, null_method = 'alexander_bloch').\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.core.parcellation: Lazy-loading parcellation image for space 'fsaverage'.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.nulls: Null map generation: Assuming n = 29 data vector(s) for n = 200 parcels.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.nulls: Using provided precomputed spin matrix.\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Spin null maps:   0%|                                                                                                                                           | 0/29 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Spin null maps:  79%|██████████████████████████████████████████████████████████████████████████████████████████████████████▎                          | 23/29 [00:00<00:00, 222.21it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Spin null maps: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:00<00:00, 223.39it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.nulls: Null data generation finished.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:42 | nispace.core.permute: Z-standardizing null maps.\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc):   0%|                                                                                                                        | 0/1000 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc):   1%|▉                                                                                                               | 8/1000 [00:00<01:47,  9.21it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc):  26%|████████████████████████████▎                                                                                | 260/1000 [00:00<00:02, 359.93it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc): 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:01<00:00, 967.87it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null colocalizations (spearman, 4 proc):   0%|                                                                                                                | 0/1000 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null colocalizations (spearman, 4 proc):  76%|████████████████████████████████████████████████████████████████████████████                        | 760/1000 [00:00<00:00, 6674.40it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null colocalizations (spearman, 4 proc): 100%|███████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 8715.04it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:46 | nispace.core.permute: Calculating exact p-values (tails = {'rho': 'upper'}).\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Spin test — 5 lowest p-values:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Pain</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>set</th>\n",
       "      <th>map</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Noradrenaline/Acetylcholine</th>\n",
       "      <th>target-VAChT_tracer-feobv_n-18_dx-hc_pub-aghourian2017</th>\n",
       "      <td>0.020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Opioids/Endocannabinoids</th>\n",
       "      <th>target-KOR_tracer-ly2795050_n-28_dx-hc_pub-vijay2018</th>\n",
       "      <td>0.031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Noradrenaline/Acetylcholine</th>\n",
       "      <th>target-NET_tracer-mrb_n-10_dx-hc_pub-hesse2017</th>\n",
       "      <td>0.068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Histamine</th>\n",
       "      <th>target-H3_tracer-gsk189254_n-8_dx-hc_pub-gallezot2017</th>\n",
       "      <td>0.075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Glutamate</th>\n",
       "      <th>target-mGluR5_tracer-abp688_n-73_dx-hc_pub-smart2019</th>\n",
       "      <td>0.167</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                 Pain\n",
       "set                         map                                                      \n",
       "Noradrenaline/Acetylcholine target-VAChT_tracer-feobv_n-18_dx-hc_pub-aghour...  0.020\n",
       "Opioids/Endocannabinoids    target-KOR_tracer-ly2795050_n-28_dx-hc_pub-vija...  0.031\n",
       "Noradrenaline/Acetylcholine target-NET_tracer-mrb_n-10_dx-hc_pub-hesse2017      0.068\n",
       "Histamine                   target-H3_tracer-gsk189254_n-8_dx-hc_pub-gallez...  0.075\n",
       "Glutamate                   target-mGluR5_tracer-abp688_n-73_dx-hc_pub-smar...  0.167"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# spin test — auto-selected for Schaefer200 (cortical + fsLR), so null_method can be omitted\n",
    "nsp.permute(\"maps\", n_perm=1000, p_tails=\"upper\")\n",
    "p_spin = nsp.get_p_values()\n",
    "print(\"Spin test — 5 lowest p-values:\")\n",
    "p_spin.T.sort_values(by=\"Pain\").head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb966271",
   "metadata": {},
   "source": [
    "### Moran Spectral Randomization (moran)\n",
    "\n",
    "Moran spectral randomization is automatically selected when the spin test is not applicable — specifically for **combined (cortex + subcortex)** or **volumetric-only** parcellations that have no spherical surface projection.\n",
    "\n",
    "It works by decomposing the spatial covariance structure of the parcellation into eigenvectors (Moran eigenvectors), then randomizing the spectral coefficients while preserving the overall structure. The result is a null map with the same degree of spatial autocorrelation, but with the specific spatial pattern scrambled.\n",
    "\n",
    "Two key properties:\n",
    "1. **Works with any parcellation geometry** — subcortical structures, volumetric atlases, anything.\n",
    "2. **No precomputed matrix needed** — eigenvectors are derived on the fly from the parcellation distance matrix.\n",
    "\n",
    "With Schaefer200 the spin test is the auto-selected default, so to run Moran we specify `null_method=\"moran\"` explicitly:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c45ac6d7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-01T13:03:46.325362Z",
     "iopub.status.busy": "2026-06-01T13:03:46.325138Z",
     "iopub.status.idle": "2026-06-01T13:03:47.189006Z",
     "shell.execute_reply": "2026-06-01T13:03:47.188678Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:46 | nispace.core.permute: Found existing null maps.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mWARNING | 01/06/26 15:03:46 | nispace.core.permute: Null method changed. Will re-generate.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:46 | nispace.core.permute: Generating null maps (n = 1000, null_method = 'moran').\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:46 | nispace.nulls: Null map generation: Assuming n = 29 data vector(s) for n = 200 parcels.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:46 | nispace.nulls: Using provided distance matrix/matrices.\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Moran null maps (4 proc):   0%|                                                                                                                                 | 0/29 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Moran null maps (4 proc):  83%|██████████████████████████████████████████████████████████████████████████████████████████████████▍                    | 24/29 [00:00<00:00, 150.83it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Moran null maps (4 proc): 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:00<00:00, 181.00it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:46 | nispace.nulls: Null data generation finished.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:46 | nispace.core.permute: Z-standardizing null maps.\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc):   0%|                                                                                                                        | 0/1000 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc):  50%|██████████████████████████████████████████████████████▍                                                     | 504/1000 [00:00<00:00, 4415.82it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc): 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 7475.52it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null colocalizations (spearman, 4 proc):   0%|                                                                                                                | 0/1000 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null colocalizations (spearman, 4 proc): 100%|██████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 12674.60it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:47 | nispace.core.permute: Calculating exact p-values (tails = {'rho': 'upper'}).\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Moran spectral — 5 lowest p-values:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Pain</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>set</th>\n",
       "      <th>map</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">Noradrenaline/Acetylcholine</th>\n",
       "      <th>target-VAChT_tracer-feobv_n-18_dx-hc_pub-aghourian2017</th>\n",
       "      <td>0.005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>target-NET_tracer-mrb_n-10_dx-hc_pub-hesse2017</th>\n",
       "      <td>0.021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Opioids/Endocannabinoids</th>\n",
       "      <th>target-KOR_tracer-ly2795050_n-28_dx-hc_pub-vijay2018</th>\n",
       "      <td>0.027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Glutamate</th>\n",
       "      <th>target-mGluR5_tracer-abp688_n-73_dx-hc_pub-smart2019</th>\n",
       "      <td>0.048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Histamine</th>\n",
       "      <th>target-H3_tracer-gsk189254_n-8_dx-hc_pub-gallezot2017</th>\n",
       "      <td>0.065</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                 Pain\n",
       "set                         map                                                      \n",
       "Noradrenaline/Acetylcholine target-VAChT_tracer-feobv_n-18_dx-hc_pub-aghour...  0.005\n",
       "                            target-NET_tracer-mrb_n-10_dx-hc_pub-hesse2017      0.021\n",
       "Opioids/Endocannabinoids    target-KOR_tracer-ly2795050_n-28_dx-hc_pub-vija...  0.027\n",
       "Glutamate                   target-mGluR5_tracer-abp688_n-73_dx-hc_pub-smar...  0.048\n",
       "Histamine                   target-H3_tracer-gsk189254_n-8_dx-hc_pub-gallez...  0.065"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nsp.permute(\"maps\", maps_method=\"moran\", n_perm=1000, p_tails=\"upper\")\n",
    "p_moran = nsp.get_p_values()\n",
    "print(\"Moran spectral — 5 lowest p-values:\")\n",
    "p_moran.T.sort_values(by=\"Pain\").head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "80eb0e80",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-01T13:03:47.190694Z",
     "iopub.status.busy": "2026-06-01T13:03:47.190590Z",
     "iopub.status.idle": "2026-06-01T13:03:47.196099Z",
     "shell.execute_reply": "2026-06-01T13:03:47.195811Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>rho</th>\n",
       "      <th>p_spin</th>\n",
       "      <th>p_moran</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>set</th>\n",
       "      <th>map</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Noradrenaline/Acetylcholine</th>\n",
       "      <th>target-VAChT_tracer-feobv_n-18_dx-hc_pub-aghourian2017</th>\n",
       "      <td>0.435971</td>\n",
       "      <td>0.020</td>\n",
       "      <td>0.005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Opioids/Endocannabinoids</th>\n",
       "      <th>target-KOR_tracer-ly2795050_n-28_dx-hc_pub-vijay2018</th>\n",
       "      <td>0.339014</td>\n",
       "      <td>0.031</td>\n",
       "      <td>0.027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Noradrenaline/Acetylcholine</th>\n",
       "      <th>target-NET_tracer-mrb_n-10_dx-hc_pub-hesse2017</th>\n",
       "      <td>0.309315</td>\n",
       "      <td>0.068</td>\n",
       "      <td>0.021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Histamine</th>\n",
       "      <th>target-H3_tracer-gsk189254_n-8_dx-hc_pub-gallezot2017</th>\n",
       "      <td>0.304823</td>\n",
       "      <td>0.075</td>\n",
       "      <td>0.065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Glutamate</th>\n",
       "      <th>target-mGluR5_tracer-abp688_n-73_dx-hc_pub-smart2019</th>\n",
       "      <td>0.219451</td>\n",
       "      <td>0.167</td>\n",
       "      <td>0.048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Serotonin</th>\n",
       "      <th>target-5HTT_tracer-dasb_n-18_dx-hc_pub-savli2012</th>\n",
       "      <td>0.176491</td>\n",
       "      <td>0.227</td>\n",
       "      <td>0.237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Opioids/Endocannabinoids</th>\n",
       "      <th>target-MOR_tracer-carfentanil_n-204_dx-hc_pub-kantonen2020</th>\n",
       "      <td>0.151909</td>\n",
       "      <td>0.254</td>\n",
       "      <td>0.121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dopamine</th>\n",
       "      <th>target-DAT_tracer-fpcit_n-174_dx-hc_pub-dukart2018</th>\n",
       "      <td>0.129229</td>\n",
       "      <td>0.272</td>\n",
       "      <td>0.283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Opioids/Endocannabinoids</th>\n",
       "      <th>target-CB1_tracer-omar_n-77_dx-hc_pub-normandin2015</th>\n",
       "      <td>0.102278</td>\n",
       "      <td>0.325</td>\n",
       "      <td>0.264</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GABA</th>\n",
       "      <th>target-GABAa5_tracer-ro154513_n-10_dx-hc_pub-lukow2022</th>\n",
       "      <td>0.111743</td>\n",
       "      <td>0.334</td>\n",
       "      <td>0.287</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                     rho  \\\n",
       "set                         map                                                            \n",
       "Noradrenaline/Acetylcholine target-VAChT_tracer-feobv_n-18_dx-hc_pub-aghour...  0.435971   \n",
       "Opioids/Endocannabinoids    target-KOR_tracer-ly2795050_n-28_dx-hc_pub-vija...  0.339014   \n",
       "Noradrenaline/Acetylcholine target-NET_tracer-mrb_n-10_dx-hc_pub-hesse2017      0.309315   \n",
       "Histamine                   target-H3_tracer-gsk189254_n-8_dx-hc_pub-gallez...  0.304823   \n",
       "Glutamate                   target-mGluR5_tracer-abp688_n-73_dx-hc_pub-smar...  0.219451   \n",
       "Serotonin                   target-5HTT_tracer-dasb_n-18_dx-hc_pub-savli2012    0.176491   \n",
       "Opioids/Endocannabinoids    target-MOR_tracer-carfentanil_n-204_dx-hc_pub-k...  0.151909   \n",
       "Dopamine                    target-DAT_tracer-fpcit_n-174_dx-hc_pub-dukart2018  0.129229   \n",
       "Opioids/Endocannabinoids    target-CB1_tracer-omar_n-77_dx-hc_pub-normandin...  0.102278   \n",
       "GABA                        target-GABAa5_tracer-ro154513_n-10_dx-hc_pub-lu...  0.111743   \n",
       "\n",
       "                                                                                p_spin  \\\n",
       "set                         map                                                          \n",
       "Noradrenaline/Acetylcholine target-VAChT_tracer-feobv_n-18_dx-hc_pub-aghour...   0.020   \n",
       "Opioids/Endocannabinoids    target-KOR_tracer-ly2795050_n-28_dx-hc_pub-vija...   0.031   \n",
       "Noradrenaline/Acetylcholine target-NET_tracer-mrb_n-10_dx-hc_pub-hesse2017       0.068   \n",
       "Histamine                   target-H3_tracer-gsk189254_n-8_dx-hc_pub-gallez...   0.075   \n",
       "Glutamate                   target-mGluR5_tracer-abp688_n-73_dx-hc_pub-smar...   0.167   \n",
       "Serotonin                   target-5HTT_tracer-dasb_n-18_dx-hc_pub-savli2012     0.227   \n",
       "Opioids/Endocannabinoids    target-MOR_tracer-carfentanil_n-204_dx-hc_pub-k...   0.254   \n",
       "Dopamine                    target-DAT_tracer-fpcit_n-174_dx-hc_pub-dukart2018   0.272   \n",
       "Opioids/Endocannabinoids    target-CB1_tracer-omar_n-77_dx-hc_pub-normandin...   0.325   \n",
       "GABA                        target-GABAa5_tracer-ro154513_n-10_dx-hc_pub-lu...   0.334   \n",
       "\n",
       "                                                                                p_moran  \n",
       "set                         map                                                          \n",
       "Noradrenaline/Acetylcholine target-VAChT_tracer-feobv_n-18_dx-hc_pub-aghour...    0.005  \n",
       "Opioids/Endocannabinoids    target-KOR_tracer-ly2795050_n-28_dx-hc_pub-vija...    0.027  \n",
       "Noradrenaline/Acetylcholine target-NET_tracer-mrb_n-10_dx-hc_pub-hesse2017        0.021  \n",
       "Histamine                   target-H3_tracer-gsk189254_n-8_dx-hc_pub-gallez...    0.065  \n",
       "Glutamate                   target-mGluR5_tracer-abp688_n-73_dx-hc_pub-smar...    0.048  \n",
       "Serotonin                   target-5HTT_tracer-dasb_n-18_dx-hc_pub-savli2012      0.237  \n",
       "Opioids/Endocannabinoids    target-MOR_tracer-carfentanil_n-204_dx-hc_pub-k...    0.121  \n",
       "Dopamine                    target-DAT_tracer-fpcit_n-174_dx-hc_pub-dukart2018    0.283  \n",
       "Opioids/Endocannabinoids    target-CB1_tracer-omar_n-77_dx-hc_pub-normandin...    0.264  \n",
       "GABA                        target-GABAa5_tracer-ro154513_n-10_dx-hc_pub-lu...    0.287  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# compare spin and Moran side by side\n",
    "comparison = pd.DataFrame({\n",
    "    \"rho\": nsp.get_colocalizations().T[\"Pain\"],\n",
    "    \"p_spin\": p_spin.T[\"Pain\"],\n",
    "    \"p_moran\": p_moran.T[\"Pain\"],\n",
    "}).sort_values(\"p_spin\")\n",
    "comparison.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f52b011",
   "metadata": {},
   "source": [
    "## Part 2: Group permutation\n",
    "\n",
    "When you have individual subject data and a group comparison design, permuting the reference maps isn't quite right — you should permute the group labels instead. The logic is:\n",
    "\n",
    "> If the colocalization between the patient effect size map and a reference map is real, then randomly reassigning subjects to groups should produce smaller effect sizes, and hence smaller colocalizations.\n",
    "\n",
    "This is implemented with `permute(\"groups\")`. We use the anorexia nervosa example dataset below.\n",
    "\n",
    "> **Note:** This dataset is **simulated** and not intended for scientific use."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a9610cfa",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-01T13:03:47.197568Z",
     "iopub.status.busy": "2026-06-01T13:03:47.197453Z",
     "iopub.status.idle": "2026-06-01T13:03:48.177681Z",
     "shell.execute_reply": "2026-06-01T13:03:48.177348Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:47 | nispace.datasets: Loading example dataset: 'anorexianervosa', parcellated with: Schaefer200Parcels7Networks.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:47 | nispace.core.parcellation: Building multi-space Parcellation for 'Schaefer200Parcels7Networks' from library.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:47 | nispace.core.parcellation: Available spaces: MNI152NLin2009cAsym, MNI152NLin6Asym, fsaverage, fsLR\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:47 | nispace.core.parcellation: Parcellation 'Schaefer200Parcels7Networks': validation passed.\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:47 | nispace.io: Input type: DataFrame, assuming parcellated data with shape (n_files/subjects/etc, n_parcels).\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mWARNING | 01/06/26 15:03:47 | nispace.io: Parcellated data contains nan values!\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:47 | nispace.io: Input type: DataFrame, assuming parcellated data with shape (n_files/subjects/etc, n_parcels).\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Colocalizing (spearman, 4 proc):   0%|                                                                                                                           | 0/1 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Colocalizing (spearman, 4 proc): 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 1630.76it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:47 | nispace.core.parcellation: Lazy-loading parcellation image for space 'fsLR'.\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Permuting groups (4 proc):   0%|                                                                                                                              | 0/1000 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Permuting groups (4 proc): 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 700334.61it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null transformations (spearman, 4 proc):   0%|                                                                                                                | 0/1000 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null transformations (spearman, 4 proc):   1%|▊                                                                                                       | 8/1000 [00:00<00:18, 53.67it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null transformations (spearman, 4 proc): 100%|███████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 4370.25it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc):   0%|                                                                                                                        | 0/1000 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc):   1%|▉                                                                                                               | 8/1000 [00:00<00:19, 52.11it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Processing null arrays (4 proc): 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 3951.35it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null colocalizations (spearman, 4 proc):   0%|                                                                                                                | 0/1000 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "Null colocalizations (spearman, 4 proc): 100%|██████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:00<00:00, 14063.57it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:48 | nispace.core.permute: Calculating exact p-values (tails = {'rho': 'two'}).\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Group permutation (mean) — 5 lowest p-values:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>set</th>\n",
       "      <th>map</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Serotonin</th>\n",
       "      <th>target-5HTT_tracer-dasb_n-18_dx-hc_pub-savli2012</th>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GABA</th>\n",
       "      <th>target-GABAa5_tracer-ro154513_n-10_dx-hc_pub-lukow2022</th>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dopamine</th>\n",
       "      <th>target-FDOPA_tracer-fluorodopa_n-12_dx-hc_pub-garciagomez2018</th>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Serotonin</th>\n",
       "      <th>target-5HT1a_tracer-way100635_n-35_dx-hc_pub-savli2012</th>\n",
       "      <td>0.002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>General</th>\n",
       "      <th>target-VMAT2_tracer-dtbz_n-76_dx-hc_pub-larsen2020</th>\n",
       "      <td>0.004</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                               mean\n",
       "set       map                                                      \n",
       "Serotonin target-5HTT_tracer-dasb_n-18_dx-hc_pub-savli2012    0.001\n",
       "GABA      target-GABAa5_tracer-ro154513_n-10_dx-hc_pub-lu...  0.001\n",
       "Dopamine  target-FDOPA_tracer-fluorodopa_n-12_dx-hc_pub-g...  0.001\n",
       "Serotonin target-5HT1a_tracer-way100635_n-35_dx-hc_pub-sa...  0.002\n",
       "General   target-VMAT2_tracer-dtbz_n-76_dx-hc_pub-larsen2020  0.004"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# set up with the anorexia nervosa dataset\n",
    "an_data = fetch_example(\"anorexianervosa\", parcellation=\"Schaefer200\")\n",
    "groups = an_data.index.str.extract(r'(AN|HC)$')[0].values\n",
    "\n",
    "nsp_groups = NiSpace(x=pet_maps, y=an_data, parcellation=\"Schaefer200\",\n",
    "                     seed=42, n_proc=4)\n",
    "nsp_groups.fit()\n",
    "\n",
    "# compute effect sizes first\n",
    "nsp_groups.transform_y(transform=\"hedges(a,b)\", groups=groups)\n",
    "nsp_groups.colocalize(\"spearman\")\n",
    "\n",
    "# group permutation: randomly shuffle the AN/HC labels and recompute\n",
    "nsp_groups.permute(\n",
    "    \"groups\",\n",
    "    groups=groups,\n",
    "    n_perm=1000, # increase to >=10000 for final analyses\n",
    "    p_tails=\"two\"  # two-tailed: we don't have a directional hypothesis here\n",
    ")\n",
    "p_groups = nsp_groups.get_p_values()\n",
    "y_label = p_groups.index[0]\n",
    "print(f\"Group permutation ({y_label}) — 5 lowest p-values:\")\n",
    "p_groups.T.sort_values(by=y_label).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "40e68d5a",
   "metadata": {},
   "source": [
    "### Which null model should I use?\n",
    "\n",
    "A practical guide:\n",
    "\n",
    "| Situation | Recommended null model |\n",
    "|-----------|------------------------|\n",
    "| Single cortical map (published effect size, meta-analytic map) | `permute(\"maps\")` — spin auto-selected for Schaefer200 and similar cortical atlases |\n",
    "| Single map with combined or subcortical parcellation | `permute(\"maps\")` — Moran auto-selected |\n",
    "| Individual subject data with group comparison | `permute(\"groups\")` |\n",
    "| XSEA (set enrichment) | `permute(\"maps\")` by default — see [Notebook 10](intro10_xsea.ipynb) |\n",
    "\n",
    "Group permutation is more computationally expensive (the effect size must be recomputed for each permutation) but is more appropriate when you have individual-level data.\n",
    "\n",
    "## Part 3: Visualizing the null distribution\n",
    "\n",
    "After calling `permute()`, the null distribution is stored internally and used when you call `nsp.plot()`. The grey shading in the plot represents the null distribution percentiles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f24436f2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-06-01T13:03:48.179391Z",
     "iopub.status.busy": "2026-06-01T13:03:48.179287Z",
     "iopub.status.idle": "2026-06-01T13:03:48.607941Z",
     "shell.execute_reply": "2026-06-01T13:03:48.607652Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32mINFO | 01/06/26 15:03:48 | nispace.plotting: Significance annotation: 4/29 p_uncorrected < 0.05, 0/29 p_corrected < 0.05 (no correction applied)\u001b[0m\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x730 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(<Figure size 500x730 with 1 Axes>,\n",
       " <Axes: title={'center': \"$Spearman's\\\\ Rho$ colocalization\\n(permutation of $X\\\\ maps$)\"}, xlabel='$Rho$'>,\n",
       " <seaborn._core.plot.Plotter at 0x318674610>)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plot with null distribution shading\n",
    "nsp.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "794f9353",
   "metadata": {},
   "source": [
    "## Summary\n",
    "\n",
    "- `permute(\"maps\")` without specifying `null_method` — auto-selects spin for cortical parcellations, Moran for combined/volumetric\n",
    "- `permute(\"maps\", null_method=\"alexander_bloch\")` — spin test; for cortical surface parcellations\n",
    "- `permute(\"maps\", null_method=\"moran\")` — Moran spectral; for combined or volumetric parcellations, or as explicit override\n",
    "- `permute(\"maps\", null_method=\"burt2020\")` — variogram matching; works with any parcellation\n",
    "- `permute(\"groups\", groups=...)` — group label permutation; preferred for individual subject data\n",
    "- All methods produce empirical p-values stored in the NiSpace object, retrievable with `get_p_values()`\n",
    "\n",
    "Next: [Notebook 6](intro06_multiple_comparisons.ipynb) covers multiple comparisons correction — because running 29 receptor tests at once without correction is not a great idea."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "nsp309",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.20"
  },
  "nbsphinx": {
   "execute": "never"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}