{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "bfb56630-274e-471e-8d73-6e819e3a2aa7",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import celltypist\n",
    "from celltypist import models\n",
    "import scanpy as sc\n",
    "import pandas as pd \n",
    "import numpy as np\n",
    "import anndata\n",
    "import re\n",
    "import h5py\n",
    "import scipy.sparse as scs\n",
    "import concurrent.futures\n",
    "import scanpy.external as sce\n",
    "import gc\n",
    "from concurrent.futures import ProcessPoolExecutor\n",
    "import copy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5219cccf-6813-47d7-a30f-cc059789743e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "adata=sc.read_h5ad('/home/jupyter/reference_generating_new/NK/NKcells_processed_2023-11-22.h5ad')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "50571812-fb58-470c-aebc-237d2b02396f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(159245, 57)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.shape(adata.obs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d5d070d7-bebb-4a16-a83a-ac28ee3fa249",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "NK_Cells_Annotation=pd.read_csv('NK_Cells_Res1.5_Annotation.csv')\n",
    "NK_Cells_Annotation['leiden_resolution_1.5']=[str (x) for x in NK_Cells_Annotation['leiden_resolution_1.5']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e402c976-5bc1-49e0-9ede-ac3a17beeca3",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "NK_Cells_Annotation=NK_Cells_Annotation.query(\"AIFI_L2_Final=='Yes' & AIFI_L3_Final=='Yes'\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f89b4982-0282-43a5-bc0d-93a102adf023",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "adata.obs=adata.obs.merge(NK_Cells_Annotation, on=[\"leiden_resolution_1.5\"], how=\"left\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bd7302cd-5994-4113-8ef7-006642570e3f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_Part_1=adata.obs[adata.obs['leiden_resolution_1.5'].isin(NK_Cells_Annotation['leiden_resolution_1.5'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "6e0f3ba5-a9a0-47b3-8752-644bc74dd92b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df_Part_1.to_parquet(\"Annotation_Part_1.parquet\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "58e23d7d-6b59-442b-9e81-2b033a55f0cf",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:root] *",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}