analysis_pipeline.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "4f583e7a",
   "metadata": {
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "source": [
    "\n",
    "\n",
    "In this notebook, I will show first how to concate the TPM and counts across samples and then show how to perform DEG and generate SWAN reports.\n",
    "\n",
    "## Load library"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "446a99fb",
   "metadata": {
    "slideshow": {
     "slide_type": ""
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import anndata as ad\n",
    "from gtfparse import read_gtf\n",
    "\n",
    "from utils import *\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7ca5988-8334-45ba-9063-329533ddfd3d",
   "metadata": {},
   "source": [
    "## Get transcript information\n",
    "\n",
    "Since kallisto only gives us `trainscript_id`, we need to gather other information such as transcript name or gene id and gene name using GTF. \n",
    "\n",
    "In this part, we show how we should do this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "256bee2a-7f64-451b-b3be-afd77fc998c0",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:root:Extracted GTF attributes: ['gene_id', 'gene_type', 'gene_name', 'level', 'mgi_id', 'havana_gene', 'transcript_id', 'transcript_type', 'transcript_name', 'transcript_support_level', 'tag', 'havana_transcript', 'exon_number', 'exon_id', 'protein_id', 'ccdsid', 'ont']\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>transcript_id</th>\n",
       "      <th>transcript_type</th>\n",
       "      <th>transcript_name</th>\n",
       "      <th>gene_id</th>\n",
       "      <th>gene_type</th>\n",
       "      <th>gene_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000193812.2</th>\n",
       "      <td>ENSMUST00000193812.2</td>\n",
       "      <td>TEC</td>\n",
       "      <td>4933401J01Rik-201</td>\n",
       "      <td>ENSMUSG00000102693.2</td>\n",
       "      <td>TEC</td>\n",
       "      <td>4933401J01Rik</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000082908.3</th>\n",
       "      <td>ENSMUST00000082908.3</td>\n",
       "      <td>snRNA</td>\n",
       "      <td>Gm26206-201</td>\n",
       "      <td>ENSMUSG00000064842.3</td>\n",
       "      <td>snRNA</td>\n",
       "      <td>Gm26206</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000162897.2</th>\n",
       "      <td>ENSMUST00000162897.2</td>\n",
       "      <td>protein_coding_CDS_not_defined</td>\n",
       "      <td>Xkr4-203</td>\n",
       "      <td>ENSMUSG00000051951.6</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>Xkr4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000159265.2</th>\n",
       "      <td>ENSMUST00000159265.2</td>\n",
       "      <td>protein_coding_CDS_not_defined</td>\n",
       "      <td>Xkr4-202</td>\n",
       "      <td>ENSMUSG00000051951.6</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>Xkr4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000070533.5</th>\n",
       "      <td>ENSMUST00000070533.5</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>Xkr4-201</td>\n",
       "      <td>ENSMUSG00000051951.6</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>Xkr4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000180208.2</th>\n",
       "      <td>ENSMUST00000180208.2</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUST00000180208</td>\n",
       "      <td>ENSMUSG00000095475.2</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUSG00000095475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000178327.2</th>\n",
       "      <td>ENSMUST00000178327.2</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUST00000178327</td>\n",
       "      <td>ENSMUSG00000094855.2</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUSG00000094855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000178569.2</th>\n",
       "      <td>ENSMUST00000178569.2</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUST00000178569</td>\n",
       "      <td>ENSMUSG00000095019.2</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUSG00000095019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000179505.8</th>\n",
       "      <td>ENSMUST00000179505.8</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUST00000179505</td>\n",
       "      <td>ENSMUSG00000095041.8</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUSG00000095041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSMUST00000178343.2</th>\n",
       "      <td>ENSMUST00000178343.2</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUST00000178343</td>\n",
       "      <td>ENSMUSG00000095041.8</td>\n",
       "      <td>protein_coding</td>\n",
       "      <td>ENSMUSG00000095041</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>149443 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                             transcript_id                 transcript_type  \\\n",
       "ENSMUST00000193812.2  ENSMUST00000193812.2                             TEC   \n",
       "ENSMUST00000082908.3  ENSMUST00000082908.3                           snRNA   \n",
       "ENSMUST00000162897.2  ENSMUST00000162897.2  protein_coding_CDS_not_defined   \n",
       "ENSMUST00000159265.2  ENSMUST00000159265.2  protein_coding_CDS_not_defined   \n",
       "ENSMUST00000070533.5  ENSMUST00000070533.5                  protein_coding   \n",
       "...                                    ...                             ...   \n",
       "ENSMUST00000180208.2  ENSMUST00000180208.2                  protein_coding   \n",
       "ENSMUST00000178327.2  ENSMUST00000178327.2                  protein_coding   \n",
       "ENSMUST00000178569.2  ENSMUST00000178569.2                  protein_coding   \n",
       "ENSMUST00000179505.8  ENSMUST00000179505.8                  protein_coding   \n",
       "ENSMUST00000178343.2  ENSMUST00000178343.2                  protein_coding   \n",
       "\n",
       "                         transcript_name               gene_id  \\\n",
       "ENSMUST00000193812.2   4933401J01Rik-201  ENSMUSG00000102693.2   \n",
       "ENSMUST00000082908.3         Gm26206-201  ENSMUSG00000064842.3   \n",
       "ENSMUST00000162897.2            Xkr4-203  ENSMUSG00000051951.6   \n",
       "ENSMUST00000159265.2            Xkr4-202  ENSMUSG00000051951.6   \n",
       "ENSMUST00000070533.5            Xkr4-201  ENSMUSG00000051951.6   \n",
       "...                                  ...                   ...   \n",
       "ENSMUST00000180208.2  ENSMUST00000180208  ENSMUSG00000095475.2   \n",
       "ENSMUST00000178327.2  ENSMUST00000178327  ENSMUSG00000094855.2   \n",
       "ENSMUST00000178569.2  ENSMUST00000178569  ENSMUSG00000095019.2   \n",
       "ENSMUST00000179505.8  ENSMUST00000179505  ENSMUSG00000095041.8   \n",
       "ENSMUST00000178343.2  ENSMUST00000178343  ENSMUSG00000095041.8   \n",
       "\n",
       "                           gene_type           gene_name  \n",
       "ENSMUST00000193812.2             TEC       4933401J01Rik  \n",
       "ENSMUST00000082908.3           snRNA             Gm26206  \n",
       "ENSMUST00000162897.2  protein_coding                Xkr4  \n",
       "ENSMUST00000159265.2  protein_coding                Xkr4  \n",
       "ENSMUST00000070533.5  protein_coding                Xkr4  \n",
       "...                              ...                 ...  \n",
       "ENSMUST00000180208.2  protein_coding  ENSMUSG00000095475  \n",
       "ENSMUST00000178327.2  protein_coding  ENSMUSG00000094855  \n",
       "ENSMUST00000178569.2  protein_coding  ENSMUSG00000095019  \n",
       "ENSMUST00000179505.8  protein_coding  ENSMUSG00000095041  \n",
       "ENSMUST00000178343.2  protein_coding  ENSMUSG00000095041  \n",
       "\n",
       "[149443 rows x 6 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gencode_gtf = read_gtf(\"data/gencode.vM32.chr_patch_hapl_scaff.annotation.gtf.gz\").to_pandas()\n",
    "\n",
    "gencode_gtf_transcript = gencode_gtf[gencode_gtf[\"feature\"] == \"transcript\"]\n",
    "gencode_gtf_transcript = gencode_gtf_transcript[['transcript_id', \n",
    "                                                 'transcript_type', \n",
    "                                                 'transcript_name', \n",
    "                                                 'gene_id', \n",
    "                                                 'gene_type', \n",
    "                                                 'gene_name']]\n",
    "gencode_gtf_transcript.index = gencode_gtf_transcript.transcript_id.tolist()\n",
    "gencode_gtf_transcript"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "65acb697-5a83-43aa-add5-7c2bf1ae8fbb",
   "metadata": {},
   "source": [
    "## Concate count and tpm matrices\n",
    "\n",
    "After we grab transcript information from GTF file and clean it up, it's time to make count and TPM h5ad files in a way that `var` contains information regarding transcript and `obs` contains information about samples.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "d22b7735",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "... storing 'Genotype' as categorical\n",
      "... storing 'Sex' as categorical\n",
      "... storing 'Age' as categorical\n",
      "... storing 'Tissue' as categorical\n",
      "... storing 'Genotype description' as categorical\n",
      "... storing 'transcript_type' as categorical\n",
      "... storing 'gene_id' as categorical\n",
      "... storing 'gene_type' as categorical\n",
      "... storing 'gene_name' as categorical\n",
      "... storing 'Genotype' as categorical\n",
      "... storing 'Sex' as categorical\n",
      "... storing 'Age' as categorical\n",
      "... storing 'Tissue' as categorical\n",
      "... storing 'Genotype description' as categorical\n",
      "... storing 'transcript_type' as categorical\n",
      "... storing 'gene_id' as categorical\n",
      "... storing 'gene_type' as categorical\n",
      "... storing 'gene_name' as categorical\n"
     ]
    }
   ],
   "source": [
    "samples = pd.read_csv(\"data/sample_name.txt\", header=None, sep='   ')[0].tolist()\n",
    "\n",
    "metadata_sample_path = \"data/sample_metadata.csv\"\n",
    "\n",
    "adata_count = None\n",
    "adata_tpm = None\n",
    "\n",
    "for sample in samples:\n",
    "    transcript = pd.read_csv(f'data/output_{sample}/transcripts.txt', header=None, index_col=0)\n",
    "    transcript.index.names = ['']\n",
    "    counts = ad.read_mtx(f'data/output_{sample}/matrix.abundance.mtx', dtype='float32')[0]\n",
    "    counts.obs.index = [sample]\n",
    "    counts.var.index = transcript.index.tolist()\n",
    "    \n",
    "    tpm = ad.read_mtx(f'data/output_{sample}/matrix.abundance.tpm.mtx', dtype='float32')[0]\n",
    "    tpm.obs.index = [sample]\n",
    "    tpm.var.index = transcript.index.tolist()\n",
    "\n",
    "    if adata_count is None:\n",
    "        adata_count = counts.copy()\n",
    "        adata_tpm = tpm.copy()\n",
    "        continue\n",
    "\n",
    "    adata_count = ad.concat([adata_count, counts], axis=0)\n",
    "    adata_count.var = pd.concat([adata_count.var, gencode_gtf_transcript], axis=1).loc[adata_count.var.index, :]\n",
    "\n",
    "    adata_tpm = ad.concat([adata_tpm, tpm], axis=0)\n",
    "    adata_tpm.var = pd.concat([adata_tpm.var, gencode_gtf_transcript], axis=1).loc[adata_tpm.var.index, :]\n",
    "\n",
    "metadata_sample = pd.read_csv(metadata_sample_path)\n",
    "metadata_sample.index = metadata_sample['Sample Name'].values\n",
    "\n",
    "adata_count.obs = pd.concat([adata_count.obs, metadata_sample], axis=1)\n",
    "adata_tpm.obs = pd.concat([adata_tpm.obs, metadata_sample], axis=1)\n",
    "\n",
    "\n",
    "adata_count.write(filename=\"data/transcript_exp_count.h5ad\")\n",
    "adata_tpm.write(filename=\"data/transcript_exp_tpm.h5ad\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ffae5f0d-3866-4ba2-8c39-6195e61be95d",
   "metadata": {},
   "source": [
    "## Differential Isoform Expression (DIE) analysis\n",
    "\n",
    "In this part, you will perform the DIE analysis. \n",
    "\n",
    "Based on the mouse model and comparison they requested, you may need to adjust the filtering.\n",
    "\n",
    "Example:\n",
    "\n",
    "```\n",
    "filtering = {'Age': [\"4 months\"],\n",
    "             'Sex': [\"F\"],\n",
    "             'Genotype': ['5xCLU-h2kbKI-HO', '5xFADHEMI']}\n",
    "```\n",
    "\n",
    "Here we compare 5xCLU-h2kbKI-HO vs 5xFADHEMI using female 4 month samples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "cd69195d-d17c-4fa5-9c92-0e9fdc60714f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Fitting size factors...\n",
      "... done in 0.01 seconds.\n",
      "\n",
      "Fitting dispersions...\n",
      "... done in 1.50 seconds.\n",
      "\n",
      "Fitting dispersion trend curve...\n",
      "... done in 0.42 seconds.\n",
      "\n",
      "Fitting MAP dispersions...\n",
      "... done in 2.68 seconds.\n",
      "\n",
      "Fitting LFCs...\n",
      "... done in 1.84 seconds.\n",
      "\n",
      "Calculating cook's distance...\n",
      "... done in 0.01 seconds.\n",
      "\n",
      "Replacing 0 outlier genes.\n",
      "\n",
      "Running Wald tests...\n",
      "... done in 1.17 seconds.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Log2 fold change & Wald test p-value: Genotype 5xCLU-h2kbKI-HO vs 5xFADHEMI\n",
      "                        baseMean  log2FoldChange     lfcSE      stat  \\\n",
      "ENSMUST00000070533.5    6.793352       -0.597513  0.606963 -0.984432   \n",
      "ENSMUST00000159265.2    0.169971        1.007837  4.116927  0.244803   \n",
      "ENSMUST00000162897.2    0.496792       -0.849160  2.336193 -0.363480   \n",
      "ENSMUST00000192336.2    1.635561       -4.133322  2.169238 -1.905426   \n",
      "ENSMUST00000194099.2    0.327735       -1.813249  3.052885 -0.593946   \n",
      "...                          ...             ...       ...       ...   \n",
      "ENSMUST00000177591.2    8.748174       -1.165066  0.592498 -1.966364   \n",
      "ENSMUST00000179077.2    2.964353       -2.277283  1.185406 -1.921099   \n",
      "ENSMUST00000179483.9    7.462435       -0.949987  1.545736 -0.614585   \n",
      "ENSMUST00000238471.2  288.075700       -2.511676  0.556585 -4.512656   \n",
      "ENSMUST00000238758.2    1.983961       -0.956381  1.172401 -0.815745   \n",
      "\n",
      "                        pvalue      padj  \n",
      "ENSMUST00000070533.5  0.324903  0.999945  \n",
      "ENSMUST00000159265.2  0.806609       NaN  \n",
      "ENSMUST00000162897.2  0.716246       NaN  \n",
      "ENSMUST00000192336.2  0.056725       NaN  \n",
      "ENSMUST00000194099.2  0.552548       NaN  \n",
      "...                        ...       ...  \n",
      "ENSMUST00000177591.2  0.049257  0.999945  \n",
      "ENSMUST00000179077.2  0.054719       NaN  \n",
      "ENSMUST00000179483.9  0.538829  0.999945  \n",
      "ENSMUST00000238471.2  0.000006  0.002259  \n",
      "ENSMUST00000238758.2  0.414646       NaN  \n",
      "\n",
      "[36011 rows x 6 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:fontTools.subset:maxp pruned\n",
      "INFO:fontTools.subset:LTSH dropped\n",
      "INFO:fontTools.subset:cmap pruned\n",
      "INFO:fontTools.subset:kern dropped\n",
      "INFO:fontTools.subset:post pruned\n",
      "INFO:fontTools.subset:PCLT dropped\n",
      "INFO:fontTools.subset:JSTF dropped\n",
      "INFO:fontTools.subset:DSIG dropped\n",
      "INFO:fontTools.subset:GPOS pruned\n",
      "INFO:fontTools.subset:GSUB pruned\n",
      "INFO:fontTools.subset:glyf pruned\n",
      "INFO:fontTools.subset:Added gid0 to subset\n",
      "INFO:fontTools.subset:Added first four glyphs to subset\n",
      "INFO:fontTools.subset:Closing glyph list over 'GSUB': 62 glyphs before\n",
      "INFO:fontTools.subset:Glyph names: ['.notdef', '.null', 'A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L', 'M', 'N', 'O', 'P', 'R', 'S', 'T', 'U', 'V', 'Z', 'a', 'b', 'c', 'd', 'e', 'eight', 'equal', 'f', 'five', 'four', 'g', 'h', 'hyphen', 'i', 'j', 'k', 'l', 'm', 'minus', 'n', 'nine', 'nonmarkingreturn', 'o', 'one', 'p', 'parenleft', 'parenright', 'period', 'r', 's', 'seven', 'six', 'space', 't', 'three', 'two', 'u', 'v', 'x', 'zero']\n",
      "INFO:fontTools.subset:Glyph IDs:   [0, 1, 2, 3, 11, 12, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 51, 53, 54, 55, 56, 57, 61, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 85, 86, 87, 88, 89, 91, 237]\n",
      "INFO:fontTools.subset:Closed glyph list over 'GSUB': 62 glyphs after\n",
      "INFO:fontTools.subset:Glyph names: ['.notdef', '.null', 'A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L', 'M', 'N', 'O', 'P', 'R', 'S', 'T', 'U', 'V', 'Z', 'a', 'b', 'c', 'd', 'e', 'eight', 'equal', 'f', 'five', 'four', 'g', 'h', 'hyphen', 'i', 'j', 'k', 'l', 'm', 'minus', 'n', 'nine', 'nonmarkingreturn', 'o', 'one', 'p', 'parenleft', 'parenright', 'period', 'r', 's', 'seven', 'six', 'space', 't', 'three', 'two', 'u', 'v', 'x', 'zero']\n",
      "INFO:fontTools.subset:Glyph IDs:   [0, 1, 2, 3, 11, 12, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 51, 53, 54, 55, 56, 57, 61, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 85, 86, 87, 88, 89, 91, 237]\n",
      "INFO:fontTools.subset:Closing glyph list over 'glyf': 62 glyphs before\n",
      "INFO:fontTools.subset:Glyph names: ['.notdef', '.null', 'A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L', 'M', 'N', 'O', 'P', 'R', 'S', 'T', 'U', 'V', 'Z', 'a', 'b', 'c', 'd', 'e', 'eight', 'equal', 'f', 'five', 'four', 'g', 'h', 'hyphen', 'i', 'j', 'k', 'l', 'm', 'minus', 'n', 'nine', 'nonmarkingreturn', 'o', 'one', 'p', 'parenleft', 'parenright', 'period', 'r', 's', 'seven', 'six', 'space', 't', 'three', 'two', 'u', 'v', 'x', 'zero']\n",
      "INFO:fontTools.subset:Glyph IDs:   [0, 1, 2, 3, 11, 12, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 51, 53, 54, 55, 56, 57, 61, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 85, 86, 87, 88, 89, 91, 237]\n",
      "INFO:fontTools.subset:Closed glyph list over 'glyf': 62 glyphs after\n",
      "INFO:fontTools.subset:Glyph names: ['.notdef', '.null', 'A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L', 'M', 'N', 'O', 'P', 'R', 'S', 'T', 'U', 'V', 'Z', 'a', 'b', 'c', 'd', 'e', 'eight', 'equal', 'f', 'five', 'four', 'g', 'h', 'hyphen', 'i', 'j', 'k', 'l', 'm', 'minus', 'n', 'nine', 'nonmarkingreturn', 'o', 'one', 'p', 'parenleft', 'parenright', 'period', 'r', 's', 'seven', 'six', 'space', 't', 'three', 'two', 'u', 'v', 'x', 'zero']\n",
      "INFO:fontTools.subset:Glyph IDs:   [0, 1, 2, 3, 11, 12, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 51, 53, 54, 55, 56, 57, 61, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 85, 86, 87, 88, 89, 91, 237]\n",
      "INFO:fontTools.subset:Retaining 62 glyphs\n",
      "INFO:fontTools.subset:head subsetting not needed\n",
      "INFO:fontTools.subset:hhea subsetting not needed\n",
      "INFO:fontTools.subset:maxp subsetting not needed\n",
      "INFO:fontTools.subset:OS/2 subsetting not needed\n",
      "INFO:fontTools.subset:hmtx subsetted\n",
      "INFO:fontTools.subset:VDMX subsetting not needed\n",
      "INFO:fontTools.subset:hdmx subsetted\n",
      "INFO:fontTools.subset:cmap subsetted\n",
      "INFO:fontTools.subset:fpgm subsetting not needed\n",
      "INFO:fontTools.subset:prep subsetting not needed\n",
      "INFO:fontTools.subset:cvt  subsetting not needed\n",
      "INFO:fontTools.subset:loca subsetting not needed\n",
      "INFO:fontTools.subset:post subsetted\n",
      "INFO:fontTools.subset:gasp subsetting not needed\n",
      "INFO:fontTools.subset:GDEF subsetted\n",
      "INFO:fontTools.subset:GPOS subsetted\n",
      "INFO:fontTools.subset:GSUB subsetted\n",
      "INFO:fontTools.subset:name subsetting not needed\n",
      "INFO:fontTools.subset:glyf subsetted\n",
      "INFO:fontTools.subset:head pruned\n",
      "INFO:fontTools.subset:OS/2 Unicode ranges pruned: [0, 38]\n",
      "INFO:fontTools.subset:OS/2 CodePage ranges pruned: [0]\n",
      "INFO:fontTools.subset:glyf pruned\n",
      "INFO:fontTools.subset:GDEF pruned\n",
      "INFO:fontTools.subset:GPOS pruned\n",
      "INFO:fontTools.subset:GSUB pruned\n",
      "INFO:fontTools.subset:name pruned\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "adata_tpm = ad.read_h5ad(\"data/transcript_exp_tpm.h5ad\")\n",
    "adata_tpm.X = adata_tpm.to_df().apply(pd.to_numeric)\n",
    "\n",
    "filtering = {'Age': [\"4 months\"], #c(\"12mo\"),\n",
    "             'Sex': [\"F\"], #c(\"M\"),\n",
    "             'Genotype': ['5xCLU-h2kbKI-HO', '5xFADHEMI']}\n",
    "\n",
    "if not os.path.exists(f\"DIE/\"):\n",
    "    os.makedirs(\"DIE/\", mode=0o777)\n",
    "\n",
    "fname = f\"DIE/{'_'.join(filtering['Genotype'])}_{'_'.join(filtering['Age'])}_{'_'.join(filtering['Sex'])}\"\n",
    "\n",
    "run_deseq2(adata_tpm,\n",
    "           obs_filtering=filtering,\n",
    "           obs_condition='Genotype',\n",
    "           l2fc_thresh=0, \n",
    "           adj_p_thresh=0.05,\n",
    "           label='transcript_name',\n",
    "           ofile=fname)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "370ed3ed-2048-4770-b51c-386f473a0d9d",
   "metadata": {},
   "source": [
    "## Differential Gene Expression (DGE) analysis\n",
    "\n",
    "If they also intersted in doing DGE analysis, you first need to calculate the gene expression by summing up the transcript expression for each gene and then adjust the filtering if needed. The filtering structure is similar to DIE."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "844a91c3-f42c-4c1a-955f-49f3bf8e6e70",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Fitting size factors...\n",
      "... done in 0.00 seconds.\n",
      "\n",
      "Fitting dispersions...\n",
      "... done in 0.85 seconds.\n",
      "\n",
      "Fitting dispersion trend curve...\n",
      "... done in 0.23 seconds.\n",
      "\n",
      "Fitting MAP dispersions...\n",
      "... done in 1.76 seconds.\n",
      "\n",
      "Fitting LFCs...\n",
      "... done in 0.96 seconds.\n",
      "\n",
      "Calculating cook's distance...\n",
      "... done in 0.01 seconds.\n",
      "\n",
      "Replacing 0 outlier genes.\n",
      "\n",
      "Running Wald tests...\n",
      "... done in 0.58 seconds.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Log2 fold change & Wald test p-value: Genotype 5xCLU-h2kbKI-HO vs 5xFADHEMI\n",
      "                       baseMean  log2FoldChange     lfcSE      stat    pvalue  \\\n",
      "ENSMUSG00000102693.2  26.139014        0.110623  0.317000  0.348968  0.727113   \n",
      "ENSMUSG00000051951.6   1.166198        0.468498  1.448066  0.323534  0.746291   \n",
      "ENSMUSG00000102851.2   0.000000             NaN       NaN       NaN       NaN   \n",
      "ENSMUSG00000103377.2   0.162738       -0.906424  3.880483 -0.233585  0.815307   \n",
      "ENSMUSG00000104017.2   0.163249       -0.906424  3.876806 -0.233807  0.815135   \n",
      "...                         ...             ...       ...       ...       ...   \n",
      "ENSMUSG00000084166.2   0.822960       -1.840602  1.954414 -0.941767  0.346312   \n",
      "ENSMUSG00000084014.4   0.163249       -0.906424  3.876806 -0.233807  0.815135   \n",
      "ENSMUSG00000101407.7   0.489674       -2.387328  2.618530 -0.911705  0.361924   \n",
      "ENSMUSG00000099351.2  13.461559       -0.194702  0.436822 -0.445725  0.655796   \n",
      "ENSMUSG00000100011.2  12.472643       -0.213834  0.464047 -0.460802  0.644941   \n",
      "\n",
      "                          padj  \n",
      "ENSMUSG00000102693.2  0.999999  \n",
      "ENSMUSG00000051951.6       NaN  \n",
      "ENSMUSG00000102851.2       NaN  \n",
      "ENSMUSG00000103377.2       NaN  \n",
      "ENSMUSG00000104017.2       NaN  \n",
      "...                        ...  \n",
      "ENSMUSG00000084166.2       NaN  \n",
      "ENSMUSG00000084014.4       NaN  \n",
      "ENSMUSG00000101407.7       NaN  \n",
      "ENSMUSG00000099351.2  0.999999  \n",
      "ENSMUSG00000100011.2  0.999999  \n",
      "\n",
      "[18804 rows x 6 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:fontTools.subset:maxp pruned\n",
      "INFO:fontTools.subset:LTSH dropped\n",
      "INFO:fontTools.subset:cmap pruned\n",
      "INFO:fontTools.subset:kern dropped\n",
      "INFO:fontTools.subset:post pruned\n",
      "INFO:fontTools.subset:PCLT dropped\n",
      "INFO:fontTools.subset:JSTF dropped\n",
      "INFO:fontTools.subset:DSIG dropped\n",
      "INFO:fontTools.subset:GPOS pruned\n",
      "INFO:fontTools.subset:GSUB pruned\n",
      "INFO:fontTools.subset:glyf pruned\n",
      "INFO:fontTools.subset:Added gid0 to subset\n",
      "INFO:fontTools.subset:Added first four glyphs to subset\n",
      "INFO:fontTools.subset:Closing glyph list over 'GSUB': 59 glyphs before\n",
      "INFO:fontTools.subset:Glyph names: ['.notdef', '.null', 'A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L', 'M', 'N', 'O', 'R', 'S', 'U', 'V', 'a', 'b', 'c', 'd', 'e', 'eight', 'equal', 'f', 'five', 'four', 'g', 'h', 'hyphen', 'i', 'j', 'k', 'l', 'm', 'minus', 'n', 'nine', 'nonmarkingreturn', 'o', 'one', 'p', 'parenleft', 'parenright', 'period', 'r', 's', 'seven', 'six', 'space', 't', 'three', 'two', 'u', 'v', 'x', 'zero']\n",
      "INFO:fontTools.subset:Glyph IDs:   [0, 1, 2, 3, 11, 12, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 53, 54, 56, 57, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 85, 86, 87, 88, 89, 91, 237]\n",
      "INFO:fontTools.subset:Closed glyph list over 'GSUB': 59 glyphs after\n",
      "INFO:fontTools.subset:Glyph names: ['.notdef', '.null', 'A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L', 'M', 'N', 'O', 'R', 'S', 'U', 'V', 'a', 'b', 'c', 'd', 'e', 'eight', 'equal', 'f', 'five', 'four', 'g', 'h', 'hyphen', 'i', 'j', 'k', 'l', 'm', 'minus', 'n', 'nine', 'nonmarkingreturn', 'o', 'one', 'p', 'parenleft', 'parenright', 'period', 'r', 's', 'seven', 'six', 'space', 't', 'three', 'two', 'u', 'v', 'x', 'zero']\n",
      "INFO:fontTools.subset:Glyph IDs:   [0, 1, 2, 3, 11, 12, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 53, 54, 56, 57, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 85, 86, 87, 88, 89, 91, 237]\n",
      "INFO:fontTools.subset:Closing glyph list over 'glyf': 59 glyphs before\n",
      "INFO:fontTools.subset:Glyph names: ['.notdef', '.null', 'A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L', 'M', 'N', 'O', 'R', 'S', 'U', 'V', 'a', 'b', 'c', 'd', 'e', 'eight', 'equal', 'f', 'five', 'four', 'g', 'h', 'hyphen', 'i', 'j', 'k', 'l', 'm', 'minus', 'n', 'nine', 'nonmarkingreturn', 'o', 'one', 'p', 'parenleft', 'parenright', 'period', 'r', 's', 'seven', 'six', 'space', 't', 'three', 'two', 'u', 'v', 'x', 'zero']\n",
      "INFO:fontTools.subset:Glyph IDs:   [0, 1, 2, 3, 11, 12, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 53, 54, 56, 57, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 85, 86, 87, 88, 89, 91, 237]\n",
      "INFO:fontTools.subset:Closed glyph list over 'glyf': 59 glyphs after\n",
      "INFO:fontTools.subset:Glyph names: ['.notdef', '.null', 'A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L', 'M', 'N', 'O', 'R', 'S', 'U', 'V', 'a', 'b', 'c', 'd', 'e', 'eight', 'equal', 'f', 'five', 'four', 'g', 'h', 'hyphen', 'i', 'j', 'k', 'l', 'm', 'minus', 'n', 'nine', 'nonmarkingreturn', 'o', 'one', 'p', 'parenleft', 'parenright', 'period', 'r', 's', 'seven', 'six', 'space', 't', 'three', 'two', 'u', 'v', 'x', 'zero']\n",
      "INFO:fontTools.subset:Glyph IDs:   [0, 1, 2, 3, 11, 12, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 53, 54, 56, 57, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 85, 86, 87, 88, 89, 91, 237]\n",
      "INFO:fontTools.subset:Retaining 59 glyphs\n",
      "INFO:fontTools.subset:head subsetting not needed\n",
      "INFO:fontTools.subset:hhea subsetting not needed\n",
      "INFO:fontTools.subset:maxp subsetting not needed\n",
      "INFO:fontTools.subset:OS/2 subsetting not needed\n",
      "INFO:fontTools.subset:hmtx subsetted\n",
      "INFO:fontTools.subset:VDMX subsetting not needed\n",
      "INFO:fontTools.subset:hdmx subsetted\n",
      "INFO:fontTools.subset:cmap subsetted\n",
      "INFO:fontTools.subset:fpgm subsetting not needed\n",
      "INFO:fontTools.subset:prep subsetting not needed\n",
      "INFO:fontTools.subset:cvt  subsetting not needed\n",
      "INFO:fontTools.subset:loca subsetting not needed\n",
      "INFO:fontTools.subset:post subsetted\n",
      "INFO:fontTools.subset:gasp subsetting not needed\n",
      "INFO:fontTools.subset:GDEF subsetted\n",
      "INFO:fontTools.subset:GPOS subsetted\n",
      "INFO:fontTools.subset:GSUB subsetted\n",
      "INFO:fontTools.subset:name subsetting not needed\n",
      "INFO:fontTools.subset:glyf subsetted\n",
      "INFO:fontTools.subset:head pruned\n",
      "INFO:fontTools.subset:OS/2 Unicode ranges pruned: [0, 38]\n",
      "INFO:fontTools.subset:OS/2 CodePage ranges pruned: [0]\n",
      "INFO:fontTools.subset:glyf pruned\n",
      "INFO:fontTools.subset:GDEF pruned\n",
      "INFO:fontTools.subset:GPOS pruned\n",
      "INFO:fontTools.subset:GSUB pruned\n",
      "INFO:fontTools.subset:name pruned\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "adata_tpm = ad.read_h5ad(\"data/transcript_exp_tpm.h5ad\")\n",
    "adata_tpm.X = adata_tpm.to_df().apply(pd.to_numeric)\n",
    "\n",
    "#sum up transcript experssion per genes\n",
    "df_tpm_gene = adata_tpm.to_df().T.groupby(adata_tpm.var['gene_id']).sum().T\n",
    "gene_var = adata_tpm.var[['gene_id', 'gene_type', 'gene_name']]\n",
    "gene_var.index = gene_var.gene_id.tolist()\n",
    "gene_var.drop_duplicates(inplace=True)\n",
    "adata_tpm_gene = ad.AnnData(X=df_tpm_gene.values, var=gene_var, obs=adata_tpm.obs)\n",
    "\n",
    "filtering = {'Age': [\"4 months\"], #c(\"12mo\"),\n",
    "             'Sex': [\"F\"], #c(\"M\"),\n",
    "             'Genotype': ['5xCLU-h2kbKI-HO', '5xFADHEMI']}\n",
    "\n",
    "if not os.path.exists(f\"DGE/\"):\n",
    "    os.makedirs(\"DGE/\", mode=0o777)\n",
    "\n",
    "fname = f\"DGE/{'_'.join(filtering['Genotype'])}_{'_'.join(filtering['Age'])}_{'_'.join(filtering['Sex'])}\"\n",
    "\n",
    "run_deseq2(adata_tpm_gene,\n",
    "           obs_filtering=filtering,\n",
    "           obs_condition='Genotype',\n",
    "           l2fc_thresh=0, \n",
    "           adj_p_thresh=0.05,\n",
    "           label='gene_name',\n",
    "           ofile=fname)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3c194631-be49-4b21-87fa-045e0d1f1308",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}