{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "2c42d2fb",
   "metadata": {},
   "source": [
    "# SPACER Predict Notebook"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ee27a86",
   "metadata": {},
   "source": [
    "## Example Data Setup\n",
    "Run the cell below to use the synthetic example dataset.\n",
    "Swap the three path variables for your own files when running on real data.\n",
    "\n",
    "| Variable | Example value | Description |\n",
    "|---|---|---|\n",
    "| `GENE_CSV` | `data/example_genes.csv` | 377-gene reference list |\n",
    "| `ADATA_H5AD` | `data/example_spatial.h5ad` | 600-cell AnnData (400 tumour + 200 stromal) |\n",
    "| `MODEL_PATH` | `sample_output/best_model.pth` | Model saved by `train_show.ipynb` |\n",
    "\n",
    "**Synthetic label layout:** Macrophage+ cells occupy the right half of the tissue (X > 190)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "bad68658",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:37.400218Z",
     "iopub.status.busy": "2026-05-22T00:54:37.399993Z",
     "iopub.status.idle": "2026-05-22T00:54:37.405989Z",
     "shell.execute_reply": "2026-05-22T00:54:37.405657Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model found: sample_output/best_model.pth\n"
     ]
    }
   ],
   "source": [
    "# ── Example data paths (swap for your own files) ──────────────────────────────\n",
    "# Run `python create_example_data.py` once to generate the files below.\n",
    "import os\n",
    "\n",
    "GENE_CSV    = 'data/example_genes.csv'      # reference gene list\n",
    "ADATA_H5AD  = 'data/example_spatial.h5ad'  # synthetic AnnData\n",
    "MODEL_PATH  = 'sample_output/best_model.pth'  # from train_show.ipynb\n",
    "IMMUNE_CELL = 'macrophage'\n",
    "\n",
    "# If train_show.ipynb has not been run yet, save an untrained placeholder\n",
    "if not os.path.exists(MODEL_PATH):\n",
    "    import torch, pandas as pd\n",
    "    from model.model import MIL as _MIL\n",
    "    _genes = pd.read_csv(GENE_CSV)['Gene'].tolist()\n",
    "    os.makedirs(os.path.dirname(MODEL_PATH), exist_ok=True)\n",
    "    torch.save(_MIL(_genes).state_dict(), MODEL_PATH)\n",
    "    print(f\"Saved untrained placeholder → {MODEL_PATH}\")\n",
    "    print(\"Run train_show.ipynb first for a properly trained model.\")\n",
    "else:\n",
    "    print(f\"Model found: {MODEL_PATH}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "24d612a1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:37.407611Z",
     "iopub.status.busy": "2026-05-22T00:54:37.407407Z",
     "iopub.status.idle": "2026-05-22T00:54:42.759257Z",
     "shell.execute_reply": "2026-05-22T00:54:42.758290Z"
    }
   },
   "outputs": [],
   "source": [
    "import csv\n",
    "import torch\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from tqdm import tqdm\n",
    "import scanpy as sc\n",
    "from torch.utils.data import DataLoader\n",
    "from model.model import MIL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "28fa2cad",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:42.763307Z",
     "iopub.status.busy": "2026-05-22T00:54:42.762604Z",
     "iopub.status.idle": "2026-05-22T00:54:42.790376Z",
     "shell.execute_reply": "2026-05-22T00:54:42.789868Z"
    }
   },
   "outputs": [],
   "source": [
    "from torch.utils.data import Dataset\n",
    "import scipy.sparse as sp\n",
    "from scipy.spatial.distance import cdist\n",
    "from tqdm import trange\n",
    "from scipy.sparse import issparse\n",
    "\n",
    "\n",
    "def map_immune_cell(tag: str) -> str:\n",
    "    mapping = {\n",
    "        \"tcell\":       \"T\",\n",
    "        \"bcell\":       \"B\",\n",
    "        \"macrophage\":  \"Macrophage\",\n",
    "        \"neutrophil\":  \"Neutrophil\",\n",
    "        \"fibroblast\":  \"Fibroblast\",\n",
    "        \"endothelial\": \"Endothelial\",\n",
    "    }\n",
    "    if tag in mapping:\n",
    "        return mapping[tag]\n",
    "    raise ValueError(f\"Invalid immune cell type: {tag}\")\n",
    "\n",
    "\n",
    "def preprocess_data(adata, immune_cell, n_genes=500, resolution=\"low\"):\n",
    "    has_label = immune_cell is not None\n",
    "    if has_label and immune_cell not in adata.obs.columns:\n",
    "        immune_cell = map_immune_cell(immune_cell)\n",
    "\n",
    "    adata = adata.copy()\n",
    "    adata.var_names_make_unique()\n",
    "\n",
    "    tumour = adata[adata.obs[\"cell_type\"].astype(int) == 1].copy()\n",
    "\n",
    "    if issparse(tumour.X):\n",
    "        mean_expr = np.asarray(tumour.X.mean(axis=0)).ravel()\n",
    "    else:\n",
    "        mean_expr = tumour.X.mean(axis=0)\n",
    "\n",
    "    gene_names = tumour.var_names\n",
    "    n_genes = min(n_genes, int(0.2 * len(gene_names))) if n_genes > len(gene_names) else n_genes\n",
    "    top_idx  = mean_expr.argsort()[-n_genes:][::-1]\n",
    "    top_genes = gene_names[top_idx]\n",
    "\n",
    "    tumour_gene_set = [\n",
    "        \"PRAME\", \"MUC1\", \"EPCAM\", \"PMEL\", \"MAGEA3\", \"WT1\",\n",
    "        \"MYC\", \"CCND1\", \"CDK4\", \"CDK6\", \"BCL2\", \"BIRC5\",\n",
    "    ]\n",
    "    hla_genes = [g for g in adata.var_names if g.startswith(\"HLA\")]\n",
    "\n",
    "    keep = list(dict.fromkeys(tumour_gene_set + hla_genes + list(top_genes)))\n",
    "    drop = {\"CD68\", \"STAT1\", \"MMP13\"}\n",
    "    keep = [g for g in keep if g in adata.var_names and g not in drop]\n",
    "\n",
    "    adata = adata[:, keep].copy()\n",
    "\n",
    "    if has_label:\n",
    "        adata.obs[immune_cell] = adata.obs[immune_cell].astype(float)\n",
    "        if resolution != \"high\":\n",
    "            if not set(adata.obs[immune_cell].unique()).issubset({0, 1}):\n",
    "                thresh = np.percentile(\n",
    "                    adata[adata.obs[\"cell_type\"] == 1].obs[immune_cell], 75\n",
    "                )\n",
    "                adata.obs[immune_cell] = (adata.obs[immune_cell] > thresh).astype(int)\n",
    "    else:\n",
    "        adata.obs[\"dummy_label\"] = -1\n",
    "        immune_cell = \"dummy_label\"\n",
    "\n",
    "    return adata\n",
    "\n",
    "\n",
    "class BagsDataset(Dataset):\n",
    "    # mode='train': returns [pos + k-1 neg] bags; mode='infer': returns one bag (label=-1)\n",
    "    def __init__(self, input_data, immune_cell=\"tcell\", max_instances=None,\n",
    "                 radius=200, resolution=\"low\", n_genes=500, k=2, mode=\"train\"):\n",
    "        self.mode = mode.lower()\n",
    "        self.k    = k\n",
    "        self.radius = radius\n",
    "        self.resolution = resolution\n",
    "        self.max_instances = max_instances\n",
    "        self.n_genes = n_genes\n",
    "        self.immune_cell = (\n",
    "            map_immune_cell(immune_cell) if (immune_cell and self.mode == \"train\") else None\n",
    "        )\n",
    "\n",
    "        if isinstance(input_data, str):\n",
    "            df = pd.read_csv(input_data)\n",
    "            adata_list = []\n",
    "            for _, row in df.iterrows():\n",
    "                ad  = sc.read_h5ad(row[\"adata\"])\n",
    "                r   = row.get(\"radius\",     self.radius)\n",
    "                res = row.get(\"resolution\", self.resolution)\n",
    "                ad_prep = preprocess_data(ad, self.immune_cell, n_genes, res)\n",
    "                adata_list.append((ad_prep, r, res))\n",
    "            self.batches = self._build_bags(adata_list)\n",
    "        elif hasattr(input_data, \"X\"):\n",
    "            ad = preprocess_data(input_data, self.immune_cell, n_genes, resolution)\n",
    "            self.batches = self._build_bags([(ad, radius, resolution)])\n",
    "        else:\n",
    "            raise ValueError(\"input_data must be an AnnData or a CSV path\")\n",
    "\n",
    "    def __len__(self):  return len(self.batches)\n",
    "    def __getitem__(self, idx): return self.batches[idx]\n",
    "\n",
    "    def _build_bags(self, adata_radius_list):\n",
    "        train_batches, infer_items = [], []\n",
    "        for adata, radius, resolution in adata_radius_list:\n",
    "            coords     = adata.obs[[\"X\", \"Y\"]].astype(float).to_numpy()\n",
    "            expr       = adata.X\n",
    "            barcodes   = adata.obs_names.to_numpy()\n",
    "            gene_names = adata.var_names.tolist()\n",
    "            if self.mode == \"train\":\n",
    "                labels = adata.obs[self.immune_cell].astype(int).to_numpy()\n",
    "            else:\n",
    "                labels = np.full(adata.n_obs, -1, dtype=int)\n",
    "            cell_types = adata.obs[\"cell_type\"].astype(int).to_numpy()\n",
    "            pos_bags, neg_bags = [], []\n",
    "\n",
    "            for i in trange(adata.n_obs, desc=f\"[{self.mode}] r={radius}\", leave=False):\n",
    "                if cell_types[i] == 0:\n",
    "                    continue\n",
    "                dist  = cdist([coords[i]], coords)[0]\n",
    "                neigh = np.where(dist <= radius)[0]\n",
    "                neigh = neigh[cell_types[neigh] == 1]\n",
    "                if resolution == \"high\":\n",
    "                    neigh = neigh[neigh != i]\n",
    "                if len(neigh) == 0:\n",
    "                    continue\n",
    "                if self.max_instances and len(neigh) > self.max_instances:\n",
    "                    continue\n",
    "                bag = {\n",
    "                    \"distances\":       dist[neigh, None].astype(np.float32),\n",
    "                    \"gene_expression\": expr[neigh],\n",
    "                    \"label\":           labels[i],\n",
    "                    \"core_idx\":        i,\n",
    "                    \"gene_names\":      gene_names,\n",
    "                    \"cell_id\":         barcodes[i],\n",
    "                }\n",
    "                if self.mode == \"train\":\n",
    "                    (pos_bags if labels[i] == 1 else neg_bags).append(bag)\n",
    "                else:\n",
    "                    infer_items.append(bag)\n",
    "\n",
    "            if self.mode == \"train\":\n",
    "                k_neg   = self.k - 1\n",
    "                n_batch = min(len(pos_bags), len(neg_bags) // k_neg) if k_neg else len(pos_bags)\n",
    "                if n_batch == 0:\n",
    "                    continue\n",
    "                np.random.shuffle(neg_bags)\n",
    "                pos_bags = pos_bags[:n_batch]\n",
    "                neg_bags = neg_bags[:n_batch * k_neg]\n",
    "                for b in range(n_batch):\n",
    "                    train_batches.append(\n",
    "                        [pos_bags[b]] + neg_bags[b * k_neg : (b + 1) * k_neg]\n",
    "                    )\n",
    "        return train_batches if self.mode == \"train\" else infer_items\n",
    "\n",
    "\n",
    "def custom_collate_fn(batch):\n",
    "    bags = batch[0] if isinstance(batch[0], list) else batch\n",
    "    dists, exprs, labels, cores, gnames, cids = [], [], [], [], [], []\n",
    "    for bag in bags:\n",
    "        dists.append(torch.tensor(bag[\"distances\"], dtype=torch.float32))\n",
    "        ge = bag[\"gene_expression\"]\n",
    "        if sp.issparse(ge):\n",
    "            ge = ge.todense()\n",
    "        exprs.append(torch.tensor(ge, dtype=torch.float32))\n",
    "        labels.append(torch.tensor(bag[\"label\"], dtype=torch.float32))\n",
    "        cores.append(bag[\"core_idx\"])\n",
    "        gnames.append(bag[\"gene_names\"])\n",
    "        cids.append(bag[\"cell_id\"])\n",
    "    return dists, exprs, labels, cores, gnames, cids"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c143b616",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:42.792630Z",
     "iopub.status.busy": "2026-05-22T00:54:42.792295Z",
     "iopub.status.idle": "2026-05-22T00:54:42.806798Z",
     "shell.execute_reply": "2026-05-22T00:54:42.806379Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using device: cpu\n"
     ]
    }
   ],
   "source": [
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(f\"Using device: {device}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "1a56c31b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:42.808663Z",
     "iopub.status.busy": "2026-05-22T00:54:42.808393Z",
     "iopub.status.idle": "2026-05-22T00:54:42.812141Z",
     "shell.execute_reply": "2026-05-22T00:54:42.811798Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 377 reference genes from data/example_genes.csv\n"
     ]
    }
   ],
   "source": [
    "def load_all_genes(reference_gene_file):\n",
    "    all_genes = []\n",
    "    with open(reference_gene_file, 'r') as csvfile:\n",
    "        reader = csv.DictReader(csvfile)\n",
    "        for row in reader:\n",
    "            all_genes.append(row['Gene'])\n",
    "    return all_genes\n",
    "\n",
    "all_genes = load_all_genes(GENE_CSV)\n",
    "print(f\"Loaded {len(all_genes)} reference genes from {GENE_CSV}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "77a30741",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:42.813726Z",
     "iopub.status.busy": "2026-05-22T00:54:42.813468Z",
     "iopub.status.idle": "2026-05-22T00:54:42.829668Z",
     "shell.execute_reply": "2026-05-22T00:54:42.829335Z"
    }
   },
   "outputs": [],
   "source": [
    "model = MIL(all_genes).to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "8389834b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:42.831394Z",
     "iopub.status.busy": "2026-05-22T00:54:42.831063Z",
     "iopub.status.idle": "2026-05-22T00:54:42.834977Z",
     "shell.execute_reply": "2026-05-22T00:54:42.834643Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded weights from sample_output/best_model.pth\n"
     ]
    }
   ],
   "source": [
    "model.load_state_dict(torch.load(MODEL_PATH, map_location=device))\n",
    "model.eval()\n",
    "print(f\"Loaded weights from {MODEL_PATH}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "91ce4622",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:42.836590Z",
     "iopub.status.busy": "2026-05-22T00:54:42.836339Z",
     "iopub.status.idle": "2026-05-22T00:54:42.897910Z",
     "shell.execute_reply": "2026-05-22T00:54:42.896942Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                     "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction bags: 400\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r"
     ]
    }
   ],
   "source": [
    "adata = sc.read_h5ad(ADATA_H5AD)\n",
    "\n",
    "prediction_dataset = BagsDataset(\n",
    "    adata,\n",
    "    immune_cell=IMMUNE_CELL,\n",
    "    radius=50,\n",
    "    n_genes=500,\n",
    "    resolution='low',\n",
    "    mode='infer',\n",
    ")\n",
    "prediction_dataloader = DataLoader(prediction_dataset, batch_size=1, collate_fn=custom_collate_fn)\n",
    "adata.obs['Macrophage_pred'] = np.nan\n",
    "print(f\"Prediction bags: {len(prediction_dataset)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "93e82674",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:42.900623Z",
     "iopub.status.busy": "2026-05-22T00:54:42.900458Z",
     "iopub.status.idle": "2026-05-22T00:54:42.904252Z",
     "shell.execute_reply": "2026-05-22T00:54:42.903845Z"
    }
   },
   "outputs": [],
   "source": [
    "def predict(model, prediction_dataloader, adata, device=None):\n",
    "    if device is None:\n",
    "        device = next(model.parameters()).device\n",
    "    model.eval()\n",
    "\n",
    "    with torch.no_grad():\n",
    "        for batch_data in tqdm(prediction_dataloader, desc=\"Predicting\"):\n",
    "            distances_list, gene_expressions_list, _, _, gene_names_list, cell_ids_list = batch_data\n",
    "            distances_list        = [d.to(device) for d in distances_list]\n",
    "            gene_expressions_list = [g.to(device) for g in gene_expressions_list]\n",
    "\n",
    "            outputs = model(distances_list, gene_expressions_list, gene_names_list)\n",
    "            if outputs is None:\n",
    "                continue\n",
    "\n",
    "            for output, cell_id in zip(outputs, cell_ids_list):\n",
    "                adata.obs.at[cell_id, 'Macrophage_pred'] = output.cpu().item()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "d94c4dd9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:42.906748Z",
     "iopub.status.busy": "2026-05-22T00:54:42.906474Z",
     "iopub.status.idle": "2026-05-22T00:54:43.033648Z",
     "shell.execute_reply": "2026-05-22T00:54:43.033157Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Predicting: 100%|██████████| 400/400 [00:00<00:00, 3273.39it/s]\n"
     ]
    }
   ],
   "source": [
    "predict(model, prediction_dataloader, adata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "83886b4f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.035643Z",
     "iopub.status.busy": "2026-05-22T00:54:43.035487Z",
     "iopub.status.idle": "2026-05-22T00:54:43.051185Z",
     "shell.execute_reply": "2026-05-22T00:54:43.050733Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total cells: 600\n",
      "Cells with prediction: 400\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>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>cell_type</th>\n",
       "      <th>Macrophage</th>\n",
       "      <th>Macrophage_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>CELL_00000-1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.225890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00001-1</th>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.188221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00002-1</th>\n",
       "      <td>40.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.139861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00003-1</th>\n",
       "      <td>60.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.118036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00004-1</th>\n",
       "      <td>80.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.113145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00005-1</th>\n",
       "      <td>100.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.123089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00006-1</th>\n",
       "      <td>120.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.167911</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00007-1</th>\n",
       "      <td>140.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.174974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00008-1</th>\n",
       "      <td>160.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.158479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00009-1</th>\n",
       "      <td>180.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.120093</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  X    Y  cell_type  Macrophage  Macrophage_pred\n",
       "CELL_00000-1    0.0  0.0          1           0         0.225890\n",
       "CELL_00001-1   20.0  0.0          1           0         0.188221\n",
       "CELL_00002-1   40.0  0.0          1           0         0.139861\n",
       "CELL_00003-1   60.0  0.0          1           0         0.118036\n",
       "CELL_00004-1   80.0  0.0          1           0         0.113145\n",
       "CELL_00005-1  100.0  0.0          1           0         0.123089\n",
       "CELL_00006-1  120.0  0.0          1           0         0.167911\n",
       "CELL_00007-1  140.0  0.0          1           0         0.174974\n",
       "CELL_00008-1  160.0  0.0          1           0         0.158479\n",
       "CELL_00009-1  180.0  0.0          1           0         0.120093"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Total cells: {len(adata.obs)}\")\n",
    "print(f\"Cells with prediction: {adata.obs['Macrophage_pred'].notna().sum()}\")\n",
    "adata.obs[['X','Y','cell_type','Macrophage','Macrophage_pred']].dropna().head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "590a84d9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.053177Z",
     "iopub.status.busy": "2026-05-22T00:54:43.052840Z",
     "iopub.status.idle": "2026-05-22T00:54:43.057625Z",
     "shell.execute_reply": "2026-05-22T00:54:43.057258Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cells after filtering to predicted: 400\n"
     ]
    }
   ],
   "source": [
    "mask = ~adata.obs['Macrophage_pred'].isna()\n",
    "adata = adata[mask].copy()\n",
    "print(f\"Cells after filtering to predicted: {len(adata.obs)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "30ce7c0e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.059329Z",
     "iopub.status.busy": "2026-05-22T00:54:43.058977Z",
     "iopub.status.idle": "2026-05-22T00:54:43.062808Z",
     "shell.execute_reply": "2026-05-22T00:54:43.062460Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Macrophage\n",
       "0    220\n",
       "1    180\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata.obs['Macrophage'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "6f45f5e2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.064386Z",
     "iopub.status.busy": "2026-05-22T00:54:43.064214Z",
     "iopub.status.idle": "2026-05-22T00:54:43.066642Z",
     "shell.execute_reply": "2026-05-22T00:54:43.066313Z"
    }
   },
   "outputs": [],
   "source": [
    "adata.obs['Macrophage_pred'] = adata.obs['Macrophage_pred'].fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "a94f26d6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.068363Z",
     "iopub.status.busy": "2026-05-22T00:54:43.068012Z",
     "iopub.status.idle": "2026-05-22T00:54:43.126070Z",
     "shell.execute_reply": "2026-05-22T00:54:43.125607Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AUROC: 0.9826\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "auroc = roc_auc_score(adata.obs['Macrophage'], adata.obs['Macrophage_pred'])\n",
    "print(f\"AUROC: {auroc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "12a2c61f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.128421Z",
     "iopub.status.busy": "2026-05-22T00:54:43.128037Z",
     "iopub.status.idle": "2026-05-22T00:54:43.707334Z",
     "shell.execute_reply": "2026-05-22T00:54:43.706388Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "df = adata.obs.dropna(subset=['Macrophage_pred', 'Macrophage']).copy()\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.histplot(df.loc[df['Macrophage'] == 0, 'Macrophage_pred'],\n",
    "             color='steelblue', alpha=0.5, label='Macrophage=0')\n",
    "sns.histplot(df.loc[df['Macrophage'] == 1, 'Macrophage_pred'],\n",
    "             color='tomato',    alpha=0.5, label='Macrophage=1')\n",
    "plt.legend()\n",
    "plt.xlabel('Macrophage_pred score')\n",
    "plt.ylabel('Frequency')\n",
    "plt.title('Prediction Score Distribution by Ground Truth Label')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "83807b03",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.709997Z",
     "iopub.status.busy": "2026-05-22T00:54:43.709653Z",
     "iopub.status.idle": "2026-05-22T00:54:43.850084Z",
     "shell.execute_reply": "2026-05-22T00:54:43.849680Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best threshold (Youden's J): 0.539  J=0.860\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, auc\n",
    "\n",
    "df_roc = adata.obs.dropna(subset=['Macrophage_pred', 'Macrophage']).copy()\n",
    "df_roc['Macrophage'] = df_roc['Macrophage'].astype(int)\n",
    "\n",
    "fpr, tpr, thresholds = roc_curve(df_roc['Macrophage'], df_roc['Macrophage_pred'])\n",
    "roc_auc = auc(fpr, tpr)\n",
    "\n",
    "youdenJ     = tpr - fpr\n",
    "best_thresh = thresholds[np.argmax(youdenJ)]\n",
    "print(f\"Best threshold (Youden\\'s J): {best_thresh:.3f}  J={youdenJ.max():.3f}\")\n",
    "\n",
    "plt.figure(figsize=(6, 6))\n",
    "plt.plot(fpr, tpr, label=f'ROC (AUC = {roc_auc:.3f})', color='steelblue')\n",
    "plt.scatter(fpr[np.argmax(youdenJ)], tpr[np.argmax(youdenJ)],\n",
    "            marker='o', color='red', label=f'Best threshold ({best_thresh:.2f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', alpha=0.5)\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic')\n",
    "plt.legend(loc='lower right')\n",
    "plt.grid(True, alpha=0.3)\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "74f2cb30",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.852504Z",
     "iopub.status.busy": "2026-05-22T00:54:43.852309Z",
     "iopub.status.idle": "2026-05-22T00:54:43.977555Z",
     "shell.execute_reply": "2026-05-22T00:54:43.977221Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_recall_curve, average_precision_score\n",
    "\n",
    "df_pr = adata.obs.dropna(subset=['Macrophage_pred', 'Macrophage']).copy()\n",
    "df_pr['Macrophage'] = df_pr['Macrophage'].astype(int)\n",
    "precision, recall, _ = precision_recall_curve(df_pr['Macrophage'], df_pr['Macrophage_pred'])\n",
    "ap       = average_precision_score(df_pr['Macrophage'], df_pr['Macrophage_pred'])\n",
    "baseline = df_pr['Macrophage'].mean()\n",
    "\n",
    "plt.figure(figsize=(6, 6))\n",
    "plt.plot(recall, precision, color='darkorange', label=f'PR curve (AP = {ap:.3f})')\n",
    "plt.axhline(y=baseline, color='gray', linestyle='--', label=f'No-skill ({baseline:.3f})')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.title('Precision-Recall Curve')\n",
    "plt.legend(loc='upper right')\n",
    "plt.grid(True, alpha=0.3)\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "3280ae30",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.979545Z",
     "iopub.status.busy": "2026-05-22T00:54:43.979224Z",
     "iopub.status.idle": "2026-05-22T00:54:43.984235Z",
     "shell.execute_reply": "2026-05-22T00:54:43.983914Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Median prediction score (threshold): 0.4995\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Macrophage_pred_binary\n",
       "1    221\n",
       "0    179\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "threshold_val = np.percentile(adata.obs['Macrophage_pred'], 50)\n",
    "print(f\"Median prediction score (threshold): {threshold_val:.4f}\")\n",
    "adata.obs['Macrophage_pred_binary'] = (adata.obs['Macrophage_pred'] >= 0.4).astype(int)\n",
    "adata.obs['Macrophage_pred_binary'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "9d1ddbdd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:43.985921Z",
     "iopub.status.busy": "2026-05-22T00:54:43.985635Z",
     "iopub.status.idle": "2026-05-22T00:54:44.055373Z",
     "shell.execute_reply": "2026-05-22T00:54:44.055038Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
    "\n",
    "df_cm = adata.obs.dropna(subset=['Macrophage_pred_binary', 'Macrophage']).copy()\n",
    "cm   = confusion_matrix(df_cm['Macrophage'].astype(int),\n",
    "                        df_cm['Macrophage_pred_binary'].astype(int))\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm,\n",
    "                               display_labels=['No Macrophage', 'Macrophage'])\n",
    "fig, ax = plt.subplots(figsize=(5, 4))\n",
    "disp.plot(ax=ax, colorbar=False, cmap='Blues')\n",
    "ax.set_title('Confusion Matrix (threshold = 0.4)')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "2f57b6b6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:44.057206Z",
     "iopub.status.busy": "2026-05-22T00:54:44.056922Z",
     "iopub.status.idle": "2026-05-22T00:54:44.769876Z",
     "shell.execute_reply": "2026-05-22T00:54:44.768730Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x600 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_cells = len(adata.obs)\n",
    "marker_size = max(2, min(20, 2000 / n_cells))  # scale dot size to dataset\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(16, 6))\n",
    "\n",
    "sc0 = axes[0].scatter(\n",
    "    adata.obs['X'], adata.obs['Y'],\n",
    "    c=adata.obs['Macrophage'].astype(float),\n",
    "    cmap='RdBu_r', s=marker_size, alpha=0.7\n",
    ")\n",
    "plt.colorbar(sc0, ax=axes[0], label='Ground Truth (Macrophage)')\n",
    "axes[0].set_title('Ground Truth: Macrophage Label')\n",
    "axes[0].set_xlabel('X'); axes[0].set_ylabel('Y')\n",
    "axes[0].invert_yaxis(); axes[0].set_aspect('equal')\n",
    "\n",
    "vmin = adata.obs['Macrophage_pred'].quantile(0.01)\n",
    "vmax = adata.obs['Macrophage_pred'].quantile(0.99)\n",
    "sc1 = axes[1].scatter(\n",
    "    adata.obs['X'], adata.obs['Y'],\n",
    "    c=adata.obs['Macrophage_pred'],\n",
    "    cmap='plasma', s=marker_size, alpha=0.7, vmin=vmin, vmax=vmax\n",
    ")\n",
    "plt.colorbar(sc1, ax=axes[1], label='Predicted Score')\n",
    "axes[1].set_title('SPACER Predicted Macrophage Score')\n",
    "axes[1].set_xlabel('X'); axes[1].set_ylabel('Y')\n",
    "axes[1].invert_yaxis(); axes[1].set_aspect('equal')\n",
    "\n",
    "plt.suptitle('Spatial Distribution of Macrophage Predictions', fontsize=14)\n",
    "plt.tight_layout()\n",
    "plt.savefig('spatial_macrophage_predictions.png', dpi=150, bbox_inches='tight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "2bc0dd49",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-05-22T00:54:44.772148Z",
     "iopub.status.busy": "2026-05-22T00:54:44.771944Z",
     "iopub.status.idle": "2026-05-22T00:54:44.781103Z",
     "shell.execute_reply": "2026-05-22T00:54:44.780284Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True positives: 179 / 180 ground-truth positive cells\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>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>Macrophage</th>\n",
       "      <th>Macrophage_pred</th>\n",
       "      <th>Macrophage_pred_binary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>CELL_00011-1</th>\n",
       "      <td>220.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.426641</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00012-1</th>\n",
       "      <td>240.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.695324</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00013-1</th>\n",
       "      <td>260.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.858642</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00014-1</th>\n",
       "      <td>280.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.699150</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00015-1</th>\n",
       "      <td>300.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.555146</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00016-1</th>\n",
       "      <td>320.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.600728</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00017-1</th>\n",
       "      <td>340.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.821454</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00018-1</th>\n",
       "      <td>360.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.956521</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00019-1</th>\n",
       "      <td>380.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.969552</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELL_00031-1</th>\n",
       "      <td>220.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.463269</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  X     Y  Macrophage  Macrophage_pred  Macrophage_pred_binary\n",
       "CELL_00011-1  220.0   0.0           1         0.426641                       1\n",
       "CELL_00012-1  240.0   0.0           1         0.695324                       1\n",
       "CELL_00013-1  260.0   0.0           1         0.858642                       1\n",
       "CELL_00014-1  280.0   0.0           1         0.699150                       1\n",
       "CELL_00015-1  300.0   0.0           1         0.555146                       1\n",
       "CELL_00016-1  320.0   0.0           1         0.600728                       1\n",
       "CELL_00017-1  340.0   0.0           1         0.821454                       1\n",
       "CELL_00018-1  360.0   0.0           1         0.956521                       1\n",
       "CELL_00019-1  380.0   0.0           1         0.969552                       1\n",
       "CELL_00031-1  220.0  20.0           1         0.463269                       1"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "matched = adata.obs[(adata.obs['Macrophage'] == 1) & (adata.obs['Macrophage_pred_binary'] == 1)]\n",
    "print(f\"True positives: {len(matched)} / {(adata.obs['Macrophage']==1).sum()} ground-truth positive cells\")\n",
    "matched[['X','Y','Macrophage','Macrophage_pred','Macrophage_pred_binary']].head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7b64fd87",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.9.19"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}