20220524_Figure2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "46ae1326",
   "metadata": {},
   "source": [
    "# Load library"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ead044ac",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "if not os.getcwd().endswith('m6A'):\n",
    "    os.chdir('m6A/')\n",
    "%load_ext rpy2.ipython\n",
    "from scripts.common import *\n",
    "\n",
    "data_folder='/home/dengw1/workspace/mouse_lowinput/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b582adad",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%R\n",
    "source('scripts/common.r')\n",
    "data_folder<-'/home/dengw1/workspace/mouse_lowinput/'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5721232a",
   "metadata": {},
   "source": [
    "# Figure 2A"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bf4f5095",
   "metadata": {},
   "outputs": [],
   "source": [
    "peak_intensity=pd.read_csv('m6A_data/peak_intensity_m_1_renamed_gmean.csv',sep='\\t',index_col=0)\n",
    "valid_idx=[]\n",
    "[valid_idx.append(x) for x in peak_intensity.columns if 'Fetal' not in x]\n",
    "def tau_index(site_in):\n",
    "    max_=max(site_in)\n",
    "    tau=sum([1-(x/max_) for x in site_in])/len(site_in)\n",
    "    return tau\n",
    "tau={}\n",
    "for site in peak_intensity.index:\n",
    "    tau[site]=tau_index(list(peak_intensity.loc[site,valid_idx]))\n",
    "pd.Series(tau).to_frame('tau_index').to_csv('ts_data/tau_index_dir.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d2981385",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R\n",
    "tau_idx<-read.csv('ts_data/tau_index_dir.txt',header=T,stringsAsFactors=F)\n",
    "cut<-quantile(tau_idx$tau_index,probs=c(0.1,0.9),na.rm=T)\n",
    "\n",
    "df_org<-read.csv('ts_data/df_with_sig_renamed.csv',sep='\\t',header=T,row.names=1)\n",
    "pdf('figures/Figure2A_density_tau.pdf')\n",
    "scale_ <- function(x) sprintf(\"%.2f\", x)\n",
    "\n",
    "p1<-ggplot() + \n",
    "    stat_bin(data=df_org,\n",
    "             aes(x=Sum, y = ..density..), geom=\"area\",fill=\"#69b3a2\",color='#69b3a2',alpha=1,binwidth=1)+\n",
    "    annotate('text',x=23, y=0.45, label=\"All peaks\", color=\"#69b3a2\",size=7)  +\n",
    "    ylab('Density')+scale_colour_Publication()+ theme_Publication()+ \n",
    "    scale_y_continuous(labels=scale_,limits=c(0,0.6))+xlab('')+\n",
    "    theme(axis.title.x=element_blank(),\n",
    "        axis.text.x=element_blank(),\n",
    "        axis.ticks.x=element_blank())\n",
    "p2<-ggplot() + \n",
    "    stat_bin(data=df_org[tau_idx[tau_idx$tau_index>=cut['90%'],]$X,],\n",
    "             aes(x=Sum, y = ..density..), geom=\"area\",fill=\"#404080\",color='#404080',alpha=1,binwidth=1)+\n",
    "    annotate('text',x=30, y=0.45, label=\"Tau >= 90% quanftile\", color=\"#404080\",size=7)  +\n",
    "    ## force x axis to end at 66\n",
    "    annotate('text',x=66, y=0, label=\"\", color=\"#404080\",size=7)  +\n",
    "    ylab('Density')+scale_colour_Publication()+ theme_Publication()+ \n",
    "    scale_y_continuous(labels=scale_,limits=c(0,0.6))+\n",
    "    theme(axis.title.x=element_blank(),\n",
    "        axis.text.x=element_blank(),\n",
    "        axis.ticks.x=element_blank())\n",
    "     \n",
    "\n",
    "p3<-ggplot() + \n",
    "    stat_bin(data=df_org[tau_idx[tau_idx$tau_index<=cut['10%'],]$X,],\n",
    "             aes(x=Sum, y = ..density..), geom=\"area\",fill=\"#fb9a99\",color='#fb9a99',alpha=1,binwidth=1)+\n",
    "    annotate('text',x=30, y=0.45, label=\"Tau <= 10% quantile\", color=\"#fb9a99\",size=7) +\n",
    "    ylab('Density')+scale_colour_Publication()+ theme_Publication()+ \n",
    "    scale_y_continuous(labels=scale_,limits=c(0,0.6))+\n",
    "    theme(axis.title.x=element_blank())\n",
    "print(ggarrange(p1,p2,p3,nrow=3))\n",
    "dev.off()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "668c3cfc",
   "metadata": {},
   "source": [
    "# Figure 2B"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "b3d18642",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "all_peak,high_tau,low_tau,"
     ]
    }
   ],
   "source": [
    "fo=open('ts_data/tau_dist.txt','w')\n",
    "def one(tissue):\n",
    "    print(tissue,end=',')\n",
    "    ret=subprocess.check_output('python3 scripts/Peak_distribution_on_utr_cds.py ts_data/%s.bed'%tissue,\n",
    "                                shell=True).decode('utf-8')\n",
    "    return [tissue,ret]\n",
    "\n",
    "\n",
    "pool=Pool(processes=25)\n",
    "all_=pool.map(one,['high_tau','low_tau','all_peak'])\n",
    "pool.close()\n",
    "pool.join()\n",
    "\n",
    "for item in all_:\n",
    "    info=item[1].strip().split('\\n')\n",
    "    for i in range(len(info)):\n",
    "        fo.write('%s\\t%s\\t%s\\n'%(item[0],i+1,info[i].split('\\t')[0]))\n",
    "fo.close()\n",
    "\n",
    "df=pd.read_csv('ts_data/tau_dist.txt',sep='\\t',header=None,names=['Group','X','Y'])\n",
    "\n",
    "df.loc[df['Group']=='high_tau','Y']=list(df[df['Group']=='high_tau']['Y']/sum(df[df['Group']=='high_tau']['Y']))\n",
    "df.loc[df['Group']=='low_tau','Y']=list(df[df['Group']=='low_tau']['Y']/sum(df[df['Group']=='low_tau']['Y']))\n",
    "df.loc[df['Group']=='all_peak','Y']=list(df[df['Group']=='all_peak']['Y']/sum(df[df['Group']=='all_peak']['Y']))\n",
    "\n",
    "df['Accumulate']=np.zeros(450)\n",
    "df.iloc[0:150,3]=[sum(df.iloc[0:x+1,2]) for x in range(150)]\n",
    "df.iloc[150:300,3]=[sum(df.iloc[150:x+151,2]) for x in range(150)]\n",
    "df.iloc[300:450,3]=[sum(df.iloc[300:x+301,2]) for x in range(150)]\n",
    "df.loc[df['Group']=='high_tau','Group']=['High Tau' for x in range(150)]\n",
    "df.loc[df['Group']=='low_tau','Group']=['Low Tau' for x in range(150)]\n",
    "df.loc[df['Group']=='all_peak','Group']=['All' for x in range(150)]\n",
    "df.to_csv('ts_data/tau_dist.txt',sep='\\t',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "b8942ff8",
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "\tPosition file = ts_data/high_tau.bed\n",
      "\tGenome = hg19\n",
      "\tOutput Directory = ts_data/tau/homer_high_tau\n",
      "\tOperating in RNA mode\n",
      "\tMotif length set at 5,6,7,\n",
      "\tUsing 10 CPUs\n",
      "\tFragment size set to 100\n",
      "\tWill optimize 10 putative motifs\n",
      "\tFound mset for \"human\", will check against vertebrates motifs\n",
      "sh: 1: bed2pos.pl: not found\n",
      "sh: 1: checkPeakFile.pl: not found\n",
      "sh: 1: cleanUpPeakFile.pl: not found\n",
      "ls: cannot access '/root/homer/.//data/genomes/hg19//preparsed//hg19.*.cgbins': No such file or directory\n",
      "\tCould not find background files for 100 bp fragments\n",
      "\tBelow are the sizes that are already available prepared.\n",
      "\tHOMER will now create background files for 100 bp fragments\n",
      "\tTo CANCEL and rerun with a differet \"-size <#>\", hit <CTRL+C> now!\n",
      "\t\t5\n",
      "\t\t4\n",
      "\t\t3\n",
      "\t\t2\n",
      "\t\t1\n",
      "\tPreparsing genome for 100 bp fragments...(will probably take 1-5 min)\n",
      "sh: 1: preparseGenome.pl: not found\n",
      "wc: /root/homer/.//data/genomes/hg19//preparsed//hg19.100.gcbins: No such file or directory\n",
      "Use of uninitialized value $lineCount in numeric lt (<) at /root/homer/bin/findMotifsGenome.pl line 648.\n",
      "!!!! Might have something wrong with preparsed files\n",
      "!!!! Rerun and add \"-preparse\" to the command line to force recreation of the files\n",
      "sh: 1: resizePosFile.pl: not found\n",
      "sh: 1: homerTools: not found\n",
      "sh: 1: cleanUpSequences.pl: not found\n",
      "sh: 1: removePoorSeq.pl: not found\n",
      "\n",
      "\tNot removing redundant sequences\n",
      "\n",
      "sh: 1: homerTools: not found\n",
      "sh: 1: freq2group.pl: not found\n",
      "\n",
      "\tTotal sequences set to 50000\n",
      "\n",
      "\tChoosing background that matches in CpG/GC content...\n",
      "wc: /root/homer/.//data/genomes/hg19//preparsed//hg19.100.gcbins: No such file or directory\n",
      "cat: /root/homer/.//data/genomes/hg19//preparsed//hg19.100.gcbins: No such file or directory\n",
      "sh: 1: makeBinaryFile.pl: not found\n",
      "sh: 1: randRemoveBackground.pl: not found\n",
      "sh: 1: assignGeneWeights.pl: not found\n",
      "\tAssembling sequence file...\n",
      "sh: 1: filterListBy.pl: not found\n",
      "\tNormalizing lower order oligos using homer2\n",
      "sh: 1: homer2: not found\n",
      "\tFinished preparing sequence/group files\n",
      "\n",
      "\t----------------------------------------------------------\n",
      "\tKnown motif enrichment\n",
      "\t----------------------------------------------------------\n",
      "\tDe novo motif finding (HOMER)\n",
      "sh: 1: homer2: not found\n",
      "sh: 1: homer2: not found\n",
      "sh: 1: homer2: not found\n",
      "cat: 'ts_data/tau/homer_high_tau/homerMotifs.motifs*': No such file or directory\n",
      "sh: 1: compareMotifs.pl: not found\n",
      "\tJob finished - if results look good, please send beer to ..\n",
      "\n",
      "\tCleaning up tmp files...\n",
      "\n",
      "\n",
      "\tPosition file = ts_data/low_tau.bed\n",
      "\tGenome = hg19\n",
      "\tOutput Directory = ts_data/tau/homer_low_tau\n",
      "\tOperating in RNA mode\n",
      "\tMotif length set at 5,6,7,\n",
      "\tUsing 10 CPUs\n",
      "\tFragment size set to 100\n",
      "\tWill optimize 10 putative motifs\n",
      "\tFound mset for \"human\", will check against vertebrates motifs\n",
      "sh: 1: bed2pos.pl: not found\n",
      "sh: 1: checkPeakFile.pl: not found\n",
      "sh: 1: cleanUpPeakFile.pl: not found\n",
      "ls: cannot access '/root/homer/.//data/genomes/hg19//preparsed//hg19.*.cgbins': No such file or directory\n",
      "\tCould not find background files for 100 bp fragments\n",
      "\tBelow are the sizes that are already available prepared.\n",
      "\tHOMER will now create background files for 100 bp fragments\n",
      "\tTo CANCEL and rerun with a differet \"-size <#>\", hit <CTRL+C> now!\n",
      "\t\t5\n",
      "\t\t4\n",
      "\t\t3\n",
      "\t\t2\n",
      "\t\t1\n",
      "\tPreparsing genome for 100 bp fragments...(will probably take 1-5 min)\n",
      "sh: 1: preparseGenome.pl: not found\n",
      "wc: /root/homer/.//data/genomes/hg19//preparsed//hg19.100.gcbins: No such file or directory\n",
      "Use of uninitialized value $lineCount in numeric lt (<) at /root/homer/bin/findMotifsGenome.pl line 648.\n",
      "!!!! Might have something wrong with preparsed files\n",
      "!!!! Rerun and add \"-preparse\" to the command line to force recreation of the files\n",
      "sh: 1: resizePosFile.pl: not found\n",
      "sh: 1: homerTools: not found\n",
      "sh: 1: cleanUpSequences.pl: not found\n",
      "sh: 1: removePoorSeq.pl: not found\n",
      "\n",
      "\tNot removing redundant sequences\n",
      "\n",
      "sh: 1: homerTools: not found\n",
      "sh: 1: freq2group.pl: not found\n",
      "\n",
      "\tTotal sequences set to 50000\n",
      "\n",
      "\tChoosing background that matches in CpG/GC content...\n",
      "wc: /root/homer/.//data/genomes/hg19//preparsed//hg19.100.gcbins: No such file or directory\n",
      "cat: /root/homer/.//data/genomes/hg19//preparsed//hg19.100.gcbins: No such file or directory\n",
      "sh: 1: makeBinaryFile.pl: not found\n",
      "sh: 1: randRemoveBackground.pl: not found\n",
      "sh: 1: assignGeneWeights.pl: not found\n",
      "\tAssembling sequence file...\n",
      "sh: 1: filterListBy.pl: not found\n",
      "\tNormalizing lower order oligos using homer2\n",
      "sh: 1: homer2: not found\n",
      "\tFinished preparing sequence/group files\n",
      "\n",
      "\t----------------------------------------------------------\n",
      "\tKnown motif enrichment\n",
      "\t----------------------------------------------------------\n",
      "\tDe novo motif finding (HOMER)\n",
      "sh: 1: homer2: not found\n",
      "sh: 1: homer2: not found\n",
      "sh: 1: homer2: not found\n",
      "cat: 'ts_data/tau/homer_low_tau/homerMotifs.motifs*': No such file or directory\n",
      "sh: 1: compareMotifs.pl: not found\n",
      "\tJob finished - if results look good, please send beer to ..\n",
      "\n",
      "\tCleaning up tmp files...\n",
      "\n",
      "\n",
      "\tPosition file = ts_data/all_peak.bed\n",
      "\tGenome = hg19\n",
      "\tOutput Directory = ts_data/tau/homer_all_peak\n",
      "\tOperating in RNA mode\n",
      "\tMotif length set at 5,6,7,\n",
      "\tUsing 10 CPUs\n",
      "\tFragment size set to 100\n",
      "\tWill optimize 10 putative motifs\n",
      "\tFound mset for \"human\", will check against vertebrates motifs\n",
      "sh: 1: bed2pos.pl: not found\n",
      "sh: 1: checkPeakFile.pl: not found\n",
      "sh: 1: cleanUpPeakFile.pl: not found\n",
      "ls: cannot access '/root/homer/.//data/genomes/hg19//preparsed//hg19.*.cgbins': No such file or directory\n",
      "\tCould not find background files for 100 bp fragments\n",
      "\tBelow are the sizes that are already available prepared.\n",
      "\tHOMER will now create background files for 100 bp fragments\n",
      "\tTo CANCEL and rerun with a differet \"-size <#>\", hit <CTRL+C> now!\n",
      "\t\t5\n",
      "\t\t4\n",
      "\t\t3\n",
      "\t\t2\n",
      "\t\t1\n",
      "\tPreparsing genome for 100 bp fragments...(will probably take 1-5 min)\n",
      "sh: 1: preparseGenome.pl: not found\n",
      "wc: /root/homer/.//data/genomes/hg19//preparsed//hg19.100.gcbins: No such file or directory\n",
      "Use of uninitialized value $lineCount in numeric lt (<) at /root/homer/bin/findMotifsGenome.pl line 648.\n",
      "!!!! Might have something wrong with preparsed files\n",
      "!!!! Rerun and add \"-preparse\" to the command line to force recreation of the files\n",
      "sh: 1: resizePosFile.pl: not found\n",
      "sh: 1: homerTools: not found\n",
      "sh: 1: cleanUpSequences.pl: not found\n",
      "sh: 1: removePoorSeq.pl: not found\n",
      "\n",
      "\tNot removing redundant sequences\n",
      "\n",
      "sh: 1: homerTools: not found\n",
      "sh: 1: freq2group.pl: not found\n",
      "\n",
      "\tTotal sequences set to 50000\n",
      "\n",
      "\tChoosing background that matches in CpG/GC content...\n",
      "wc: /root/homer/.//data/genomes/hg19//preparsed//hg19.100.gcbins: No such file or directory\n",
      "cat: /root/homer/.//data/genomes/hg19//preparsed//hg19.100.gcbins: No such file or directory\n",
      "sh: 1: makeBinaryFile.pl: not found\n",
      "sh: 1: randRemoveBackground.pl: not found\n",
      "sh: 1: assignGeneWeights.pl: not found\n",
      "\tAssembling sequence file...\n",
      "sh: 1: filterListBy.pl: not found\n",
      "\tNormalizing lower order oligos using homer2\n",
      "sh: 1: homer2: not found\n",
      "\tFinished preparing sequence/group files\n",
      "\n",
      "\t----------------------------------------------------------\n",
      "\tKnown motif enrichment\n",
      "\t----------------------------------------------------------\n",
      "\tDe novo motif finding (HOMER)\n",
      "sh: 1: homer2: not found\n",
      "sh: 1: homer2: not found\n",
      "sh: 1: homer2: not found\n",
      "cat: 'ts_data/tau/homer_all_peak/homerMotifs.motifs*': No such file or directory\n",
      "sh: 1: compareMotifs.pl: not found\n",
      "\tJob finished - if results look good, please send beer to ..\n",
      "\n",
      "\tCleaning up tmp files...\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%%bash\n",
    "for peak in high_tau low_tau all_peak\n",
    "do\n",
    "mkdir -p ts_data/tau/homer_$peak\n",
    "/root/homer/bin/findMotifsGenome.pl ts_data/$peak.bed hg19 ts_data/tau/homer_$peak -rna -len 5,6,7 -p 10 -size 100 -S 10 \n",
    "done"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c1c07806",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R\n",
    "df_dist<-read.table('ts_data/tau_dist.txt',sep='\\t',header=T,stringsAsFactors=T)\n",
    "\n",
    "sep1=50.5\n",
    "sep2=100.5\n",
    "p<-ggplot(df_dist,aes(x=X,y=Y,color=Group,group=Group))+geom_line(size=1)+\n",
    "    geom_vline(xintercept=c(sep1, sep2), linetype='dashed', size = 0.5) +\n",
    "  annotate('text', label=c(\"5'UTR\", \"CDS\", \"3'UTR\"), size = 7,x=c( sep1/2, (sep1+sep2)/2, sep2+sep1/2 ),\n",
    "           y=max(df_dist[,3])*0.9)  +\n",
    "    ylab('Fraction')+xlab('Bin')+\n",
    "    scale_colour_Publication()+ theme_Publication()+\n",
    "    theme(panel.spacing = unit(0.5, 'lines'),\n",
    "        panel.grid.major = element_blank(),\n",
    "        panel.grid.minor = element_blank(),\n",
    "#         text = element_text(size=15),\n",
    "#         axis.text.y = element_text(size=15),\n",
    "        axis.title.y = element_text(size=15),\n",
    "        legend.position = c(0.17,0.5),\n",
    "        axis.ticks.x=element_blank(),\n",
    "        axis.text.x=element_blank())+\n",
    "    guides(color=guide_legend(ncol=1,title.position='top'))+xlab('')\n",
    "pdf('figures/Figure2B_distribution.pdf',height=4.5)\n",
    "print(p)\n",
    "dev.off()\n",
    "print(p)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b6a233c",
   "metadata": {},
   "source": [
    "# Figure S3A"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "39b28ff2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "png \n",
      "  2 \n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAHgCAMAAABKCk6nAAAAD1BMVEUQlkglXJnWKDn3syv///9mYRaNAAAcnklEQVR4nO3c4aKrKg4F4Fp5/2eeuae7SpKVEBQUafJjjlgLC75qbbvvvFLU1PW6O0BU3wrgySuAJ68AnrwCePIK4MkrgCevAJ68AnjyCuDJK4AnrwCevAJ48grgySuAJ68AnrwCePIK4MkrgCevAJ68AnjyCuDJK4AnrwCevAJ48grgyUsHXld0+P8LHbws+/b7v0p6O++YP5Z4WxR6AukdPgvuPTCXcr59yLbr59sr81nAIKEScKkAXvOOmwPD1HTMrdTF3idzCrjp+jUHXuGi6AH1RaHtlXTcGhinrgI25+IHbr1+fYBFQhxwqQZelWP7AOO5GMDbZE4CN1y/1sBrdUB1UUh7vQBYpK4BtufiBm6+fk8DXvGxJ4GV1AGsBuQJzwKvdwArc8GTWchkzgK3W7/GwGtNwGU0YJb6BuD26zcAsLYoGHiFx94OXJhLT2B7/QJYTx3ARkCW8CRw5nsdsLbYlwA3W7+2wNyhQUDe7w3AYC4q8ILn4gTusH43Ai/HgFd0bCNgknp0YN/6jQCsLEoA0zELQz4ZmPgOCVyaS1dgc/0CWE0dwGZAmvCBwOZi9wdutX5NgYVDg4Ci34mBe6zfA4FXcGxfYDkXHXiBc/lFYL4mQwBnqQP4ZEA/MPMNYDjk/cDsKUtpUQKYjqkPaa1fALcALs6lBfCx9QtgLfXdwGTQEYEbBRTdXgTc9sXqApbTPDVme+AuAWW/Aewb83HA+dbFwKfnEsBqwK3fbIQA9ox5M/ByBHjb7Aa8pT682AuaSwCrAelfJwewOmQAB7C+tzcwXBQEnAJYGzKA+wGX59IZ2Fi/AFZSB/BFATfTbIzfAM4HHQKY52sT8Cbgxi9WD3Cf9QvgAIb5ArhqLgGsBmTAf60ADmBW54DPzqUCmGydGbMXsC/g4g34POAFzCWA1YDMtzsw+U70JuBUHtO9fs2B+eaeL4Dt6rN+ARzAMF9jYLQoAUyHDGAZyTigIbBjLpXAfNLDAVcHDGA60Xbr1xq4ecDewFwygP+GvCpgAKcARgF5NQbOJ9MGmIwzInCngFcB069n8F5lMr2A90EHBW4SkAOnPsAJAoOpBHDjgNcBy895AXxBwOGAX1cAp2br1wnYGzDbsID5OM2B+QzuAlZa4wBXBkxjA/PNbC4A2JrLzwHvsVIh4O3A+O0GT+Zy4PIJEsCcct+qAybTugg4BbB6wLOB/et3KzDddALnjxkB0TNQDwGM8wWwOZkA3tckDQCchxafjQL4cMBsWw14I7DYm81FTIbO6xpguhnAnYDFXK4FLqzfKMBgUT7t9XZgOJcADuDfA65clOuBUwDvQx4IyBeFhb0FeG0LnM4B01aTE6QtcF1AFjaAV605ILAnIAsbwD3WbwBgbVFuBFbYjbkE8OTAhbmU8/F5Dgd8POD9wGISjwK21y+AVWBlagHcNmAAHx0zgAPYyBfAVXMJ4ACeCnh5FHCDxf5JYN4K4ABmj1kB0TNADwGs5AtgYzLdgSve4gL4scC8FcBl4DQW8NYO4AC+BbhFwCcCL78G/JoAWF3sROYSwCCgclkL4AA2AqJngB4eAXzmChjAZ4HBXFoA5zUkcEXAAJZDkjqzfgGMPtw5gLfJBHAz4HWrAA5gXgFs7f1FYG1uAdw24PXA623A2mMBfCHw65eBewQM4EmA5aJcAcyvOwG8jdk6YAAHcAAbY44HnD0ewAH8HSKA3WMGcAAb+QKYhi/MJYCHAlbYyVxuB84eD2BwQAAHMNtL5xLAjQIGcADDgLwC2NrbCJgsijNgAAcwDMgrgK29ATwE8JH1C2AHcHEuATw5cHEuPwa8DAucDeScy+jAfORewJ+UJJAvoFjswYHTiMBvvjeAA7gyYACfAybrF8CgZgKuWb8ADmCY76eB/2a0h88PGBv4HcApgP+NeVXAAG4PLK7bdwKLjgP4APAb7TTzBXAePj8ggB8PnJ4L/M53mvkCOA+fHxDAAYyqM/A72xfAUwGTUziAtyGmBN7O5rbArwC+Hzjbuh6YBmLtAG4DvFcaD5j/Vc0IwPlk8gPaA5+7AgZwY2A4lwAGg14FLI8N4AB+ELAQduUL4IGA8ZgBfAT4xYF9cxkJ2JcvgPsBr1vlYwbwyMAiLxwzHz6AA9g9ZmvgtgED+CHA6ZeA+XADAeM8AZz4XAI4gMVcjOq2ftvmqMB8uZ8D/FaboK4CVvIEcOJzCeAAVpugugLvwlqeAE58LgEcwGoTVF/gr7CaJ4ATn8uzgM1WCuCtHguc7DwBnPhcngZsHtsK+ND3tytZ7zWAfxG4vIA/D0zmFMD2T58PBKarGMBjAh//iZL26Fi/C4HpFfORwO9hgFfHsZ9BLwyY9/XdfgTwG7dg9QXmXQZwAF8YMOts2wxgPKTSXnmfjwCuuom5H9iM1xc48x0YeM23AhgPidv9gbOc9QFXUQ8DdlyhrwNe0cEB/C0vMEH9SWA5xQDONpsC88WTB98LLIVvB04BHMDbiO/zwPnQZ4FXcPBgwJ6AqHc5Wm/gN9vUK4CfCfzOt6zqCAyWL4CbAe9lxgvgcYDJv7cBKyOi9vjA4E3kecBmugCuDoh6l6P9EjD51b8l8Ks/sGMJ2wO/trkgYCFspqNjdgHOOpYHDwr8JofOA/xtNwfeNwNY6eaBwPnFQAEGCa8Flr93XQXMhg1gEPAqYNkj712O9gTgs+uXx9+6LZ4g8wFnA/QAttN1B+bdzgkse68FfgVw+4B7d1lUEsmzhKh37WLcBJgJ2+lGAKa93Q1MEgawNkMB/B1nXGDyaAAXZoiB1wGBE1nnFMC+GarA5GiU8C5g+ug72xwb2A6XOq4fXTb4JjwEMA+aLdwR4Gz2ef8tgYmwHS4F8KdH/ui/x31r+ATgFbZ/BhgcTUo9FPV2MbCdbRsItnsC61fAAA5gtCj1ATX2St87gJM7XLoN+K29xUFgO2AAW3UXsJYxgB3Azvu/fSDY/lFgkVB2aPX+s8D0u0p9/QLYA+z6ji0bCLbPrV8A9wcuJMsGgu0GwHyg7eEAPgn830CFYPlAsB3AQwN7axLgfVG8Ab3Asj+z9wAOYL5Xn4sBvN/6FuoyYPLwyMDJzFfqPYADmO/V5zIDsAh+LODNwG/afB5w/foF8OTAvF/2+GOA4RTkdHgrgAM4gPsG1IGTlU+ZDm/1Bn4FcClgK2DaUQCXroDTA69irz6XAO4ODGdgdH8psEv4emD7BBkBOBn5tOmwXgJ4TuBtexxg/oXEs4DT8YAuYDwBNB22fRg4zQUMkqN8VwMnPZ82Hbb908DmFTCAewCvqDkIsPkKTMcDmsD8G2arsqnsmwGsLeBIwEp8PBu6+RDgvbNtTGWCom0BW1fAQYCT2zebTDatA8CJTua3gDsELAD76y0qdQd+DQ5svMX5gV8B7Mp3HzCO7g7YF7jwax1/YgDLJynR3QG7Ai+LW/hG4CPfVV4ErNYYwMtyRjhVA78CWDY7Ai+ngL8p88ADAp9cvwKwGn8wYI/wIWBtLnMAG6s3AvASwGJ+oq1ejbbwuLfRgB3CBeAs9a8AW8s3APASwHJ+om0Am+t3GlgmPAdcFn4C8EqbdMwAtuse4Pyzs1mXA9vLdz8w8x0R+PUgYN7hA4G5MAjtAc5Kn0s+4UGBC+unAPsDtgImW3ZJ3wAeHjixTavuAz5wG30RcGKbe3ac7zTwm7ULwH875wNeS+unTE+0beC09Tcc8BLAaHqirQGX1q8DME/oAP7unAd4FY2Tb3Em8PdQF3Dle0gR+F0BrLymUf8BDIBBj0eAU1/gOmGUOYCz7EpAmfa3gdPIwPuxtwFrscsBUT0HeEXr1xJ4OQpsvQJbAhcDogrg/sDWogTwNu7VwPIutQ7YtygngH230Yn79gX+HtIKuHAFLALvwwTwzcDrStdv73Cb7M3ALGAAO2pXzS6ltcDiiyIDON/rB4aXmFGA013AHuFVFOkl/R6wGRAX8x0KWAqTXrJN1/rVAcvwar486jdVBfCbtAL43AnSA3hNAwPz69j8wKkHMEl4EPhdB+ytAP7XNlJ/wzvyHQrYGTjR7mYHTi2B2wTsDZxIbwGs1XOBZWTRbgf8/TeAt0g+4HJAXwWwVo8FpqvdE3gbKYDzSN+IEwHXflfZHphPZgDg953A/FV/E/DeTQD76ybgI8L7kycBTk8DfhlzCWDQQXYKB/BBYP5V76DAbIsfE8BDANcGJH8z1RN4Dz8v8LsDMAwI1iSAlZoW+Nu+AHgPsSfmM6gC1r80DeCthgRexV4xl9bASS5fAFfWxcD00GL1AEaT+R1g5WLcAhgcWqwAfgDwt30XMPsuHwMf+pOdAE55uxGwZ/3+Kgd+XwDsD/gLwNW30STwIMBHX4EBnA90GBh+l+8EFj12ARbC6ceAk1i+I18UsU3S3XcuYwBvu/gk2gJLyscD51frAH48sPUWZwDzLuhClfL1BS7fJCjFV3tYYBZ4QOADAfsDi9UeDDjx5WsATIS/26wLeUVU8gZwABsBAzix1WsIvJLNAJ4PeCW90z4OAB8JeALYCX4dMDzWUR2BCU5XYL4oswKTYR3ptrGysS8Blr4asO8mQQVOMp8BbN7m47oRuEYYJD4BLIW/z1lYB/m4RrziJcYJvO/Jjr0bmHSFRkbAtacwSNwSeHvOlMB//5PnTbx5J7BM3AU4F14qgF03Ce2BQUJYcwOjH2u2/Bh42TbqgO2AMwN/kxwF5keeASb5FeCteA4tYAnYWpTrgLf4BvDK9+rAIMlIwFw4e1Y9cOkmoRFwwstaqgDe53UUOBHfSuAk4lnA1gd1WK8A1oBFDl/c88Bc+BLg9AvAVFjk8MUNYHqwNyJtnwZGv+MmIixzHA3YBdjpqwFLytPAL7oO7YDN9TsMDHJ0CVgLbHwTY4S7CthYlVJE2u4FrHwE/ox7OqByD1wA3h/KnqSmxOGGB+bPPA8MfgXanwlXrj9wynck0tqB9a/ajHA3AfuETwGDH2vIPLmvlcMKKAL3A17IVjH1Z/zrgd2n8L5GZMchYNJxV2DRqAYG3AsrV2oBrFDOByy+Ytz34iBWQrVtLMrW4OkaA2f//KufAUaWxmltAqu3Fy2BC1+1aeF4JgN45XufDSxLufP6jFsTUDQOA78D+DtIdsx5YCncDxh+8CXfTEPg4myy4ZsBv/BchgSWi7SuhvB1wMp/FJmYsJJHhOOZ5gJOGjBYpvUu4MSyZXF1YCWODMczOYDXu4FFwwVM+gULtZrCPYFZvUWJ1GeBBaUDOFnAfBXuBkYrNS4wiK0/m4fjmR4LjL9HwMDoWrfeB8zWQgdGFx6rrgM+XP2Bv89b2WSZcEtgviasBwO44lvKvGMv8Er3vi4DlqPQRiUw8aXAcg2+Q5kBlXw+YHF5M4ALV3cQjoeqBJa92cCV3q/DwN/t4gm8aPNmScyESr4OwFW1dVwFvB4HrnsHSdcAL/usyCKwJHZCnK8NsDKyo64GFpdFV8LmwMwXAANhBVgE7A3MjzVrOzZ7FqQE71DngGvuEnoB0zuWkYD1n7F7A9PNWmB52ngSngQW3+V/c9BNJjoH8H5sDfBaAE5ysccCXkSkNDSwHA9HweFEqn7ACy13xOr1qwDenslFLwfW34TleDgKDidSSVSyvZcJjCZzE7D4sYYcPACwuEgl/St08GIwygKm5+pJ4H1WwwF/j70CWLyqSBek247A+ymcUTYErvsw/K/Lk8D81zh57CJBGwPjywbpgvbbH5hQtgPmm+WI54FpgWMBsExyKuBxYDneWMBbg6BWA7+yHbX3MKyeAQw+3WUHm0nB6EeAxVzGBaYx+wAXAtYDi1sH61gt8/dQ8iTo2wg45bvdCRsAZ895CrA2XIXwi9R372XADuEmwGxFmgJ7AzqAXWhngPnnpJmB8zH+tfeJqlFOBTSAxRmmBc0PNqOKoy8DpuF/FnjxAC/LAhapCpj59gXGF0ZPxN8EXrbi4/GbRiNyDTD6m7QTwM67rJ8FXrJCw5lRRc8/BYyvHTcDw49yQFgOZ0Z+CnAqrt8Cx8znmZQH80nbUYyAXuAlie3v5CzghXU3LHA+eFtg61NIohfAHsDegDYwexH+CnD2Tz4IaxnAdHZlYD7xbzeHA4prDLhCI+B9S17OzwArf+JfCcy/lPSBgojZP/kgrHU/8Gvfcxa46mw36lHA+vqRYWRjPGB0Ukpg1Hi9xAveqlPAci4QeNs+BJxNaExgV8BzwKk7sL43gIsB0aKUgFnCrAmGs+pKYLysnoRsoGcDZzdO7IBewKgRwDcAZ5fdVsDKUmnASe4dA1h8NKMLuF2o8kfFSBcCLzTf9/D9eQFMDygAf31t4IN/dHceeDucAueDEGA0hlLdgNENxc8Da99O1QDvhwwDvE0mgN3A+lIR4LLwKWDHXLJZdQO2P5ptR3QHTsWAIwGDc1Xsdcwlm9VZ4PL62cDGPWo+O6utArON5wEnfa8X2EztilgPnB+RnMD1/+kKWPHJgekd4z6Xw8CFqVUA4w8hZHasTbPgfM2A08sFvBc8BKfjrZmAX+MDo6mAwZhve+BUC5zP5RBwxfpli0bWb3Tg7Oha4KJwAOf1E8DwcxKdM9k5HLDnu3wtCZgsDXMuoAHMzDoB500nsBglgPWAfFGyhzsB82PyJvR9CDBdwO2A3wTW2nMBk39tYPqZgYY5F/A8MPdtD8x/SKsBFt+ge+rE+hV+jctH+bbJdMXcVWC24HxR2gLjBq4zwFk41MVAwGTh03ZkPgoBJr8J0zBKwKbAr+QEdggHcKLtbG5g7g7g17bLERB2MCAw7NID7AEHJ0h/YGXuCBjoGMs6EjCnxHsfDJwNvTcHAdaW5gJgdulSgdHliL8u64E5lDlkC2AWppjvILA4VDkXAvgssPaXDn89FfP1BdYmrtb4wGj9DgOrH4RVYJ7GF7ARsAwIJm4LoyTiLos/iez0AucxWepiQrR+lcDwSJzE8B0N2PNJ2KpJgLOj2UrjJLrvBcDf1t3A+hAWcIKpzWoEzHpgY7Mgqm8XYGUyTYHJevCJysPPASvnjVJV6wd964HBl9Dfrk4GTCCfASwCHgTmK7KVAmwO8STg7xGeIJ+uTgY8CfzSOzYKrclflX3PALuW9QpgMXV13RzALytgLbAWsKoM4ARvNEigA8BV500v4CyKnHolsHhiC+A9FqZxF14Uo+h1og44R/WN1x4Y3u7QIBXA8OWuAyc5ZAEYBVRvAkHxbsrC54GV80Yf7RQw7+EBwMYCJvNjHKjs2dsm/3+hpu2XOn3UMwSufTkxHn39KoCNS5c+ww7AYrASsPVFjCy5+qkA/NKnD/tm+Y4A80YPYN7fpcBKjzif9VWqLPrkv5YDmM0ULlQL4Mr1awgMha8FhstKvysvCjcBVpbKAVyK1woY9qcGuQgYdcB3yHh1wPzZh4AXVmi/cnQh3i8AV9bKyz68HvglgPO0JHkZuDyfnsD4jSV1ApYJ4YRLlakeBV5cwC/cRxm4boo9gIGwHPRqYG1JWFcEtRcwuNDBHrVzteYlfB6YHaxffcghJ4BfJrBy2VBf9QgYNnDhZS4CJwbMk4rQ8NVqR9uGOwyMeuBNFEO+C+0PFQKmY8DgdQme+V9R07JwNbBc4gPA/z1kxtqrN7A2pheYHzgfsHEuGsDu6g2MIGuBZbsSeGGlPDPB/zjuRmD9jcX9TfnMwNujeZOvFkvCRZsDb6OdAa75rrwSuOLHmq11ATBMuD1IWgMBv1Ag0bvh6xGGHn2BtzFQHDvfIWC2QI8HrvuuHPIYV0AXsPaxmI2B4hTyOYDFJ14AvOTP48Agll7XA9d9lcoB6oEFoQRmaorR5zEU0PHkpwDvgx0Frvsmtf4eBgLrXd4InPJ2digC3p57HfC2eQiYb6rVGZifIMYYf48dC2gB8wW8HTgrlE90Tw6o/Sr1DDDuwQWMnvd5zBMQvXlnVQAW3Wc9ej970MHk3hpg8wqDgRNuoToPXOy+I7DyNuQHpivdH5j5wjdhCLw9ehswfs/nV8B8SB+wXvqthnwLtoDzJFcDo2u0DK+ewM2Asx0QWL2vKwBLTrGHrdS+dMbd5NXA/p8LyZzxKQzC68Dle4Zq4Ko7d34FJH2dAxb3GlyYNIrAr6wn5VhcCNg+g9ngWyf/PU+cOkMAk0FJg58gpK9TwOBSlR2kv6fx4djtQCdg6IsXlYcFa12VrwGw8bbQDRi9kLWJG8AvA1h5f4fdi8HeVBgDa8IyPb1Y2nnwFEn7VQ9s9i6nBYf+7OQ7fMBWXQAs3w8I8LsMDO6X891bsxGwVYUTmPcuxni1ApbD4Xe+MnDqBPwJk21awFb36ltwj6oAFidIIq2hgF98Ml5g0v++/WZFhsmGdwlvjXpg9VOI+onXAOZrUgaW32v4Aopb6H87jgDTJIeBl31ryQa3gfl7AxyvF/BCS0yoAjiNCizfLY4Aa0vFeY1aWbHu2YGlPKQ0YBIVvM3UAqNPf3cBb+2+wDtxqYNVDGseWeqOlAIMTlrSsN73CfD33w7A8mMSBta/zAf3e/pkjNKudWguIB+bjIVY/wLE6wevymw6aqCtie+XX6R4nnuA6SmcD+c7Y3y+FrBzLi2BS7OhgdC3ShD41QQYv+oNYPjNVjNg5UMsZy+iyez6sf2A+Wdv8Bs0A046MAc99F10PTCPdRYYlXxnfhQw2bOy+uyF77atgMWYEni7w5HvMCjKPhM2M30ljMpct63nAuM/1QUnyDFg8LV9EsLbo3+PZ5v5JWfpA8znTtfor1WD5nu/LufTgQvCZq//CkseApbf6n4KvIjfeeWBF/FzHIpyCBhcvsCN3UjA5sck9mucWm2A8w+TjM1fCys9ygHgjHXb5GfIp+28RF8FbPxkLn354iWFku8sA8NvdWsL+eKAbLnLwPQJfy18CaxCowd/W9UvQAVYe9Hj8p8g4CMx7Qp/94501YDgbgD+3opfa7XAeMXxihnACA29sNsBW5/eGQs5ZG+UgcF3WcUfV7JhUHNlxVLyK4xssKe5gIvx/koBxhcBfbhqYPwf0ajFr63sJfBt4hOkEbByTSmcfvQ5HmBzseEYVj51sfnL0hiz9grDxvT70r/wyA/IgMmT6AZo/dsDxxQXE7yA6OWtT1+Jki9wYbG/h+DA4GD8vrqPxHgR+iFg8SnEOK9f/Aci7R0av9seAM4H+G7zQT9ti0OeHCbwf4cVz6bv4XII7a6F3zCSk0lePMBAtVeY7FNINqYKnC3Kd8sFrPylnQMYUYr9fw1ttvCkZGPvTf/l8nt04k80fq1ReVFm0a82GbPAa0oDRu+rVcDqqm57eDfauZDAhRsvtnIuuIGV0UlntG0By1IWWzmDa96vaZf5OMqYcE2U7Hj9wF72Ryy8H2NlRPHpftrKIuivtWpg8L5PV8XyNd4PMeUB4IoxZSkvVhcw+Csl1L07IJuvAM4XRX+t1S0fHBN8/51t0itPzWKDfMd+C4G3Xkrhj0l4/Y4Be7/VVW5B+CpAYO0UVqdNumOvoC04pcQncjUwFS7F068l4mZAS4Lud5T1qwcm58K26b6s7fu2Yz8t8e6QtytPD+WUEvcI9cDbqmkDFrPVvMXpSeSLVVm/I8AJLFXhAwcoLiBHJjtqL37eFQfntZwNCsTyVrz8Fl48sq8n9my8fuZp89nhyhz12ArgySuAJ68AnrwCePIK4MkrgCevAJ68AnjyCuDJK4AnrwCevAJ48grgySuAJ68AnrwCePIK4MkrgCevAJ68AnjyCuDJK4AnrwCevAJ48grgySuAJ68AnrwCePIK4Mnrf9IUd2AR0+HWAAAAAElFTkSuQmCC\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R\n",
    "demo_tissue<-c('high_tau','low_tau','all_peak') # Adult brain, adult heart and adult small intestine\n",
    "plot_data1 = function (tissue) {\n",
    "    seq_matrix<-read.csv(paste0('ts_data/tau/homer_',tissue,'/homerResults/motif1.motif'),skip=1,sep='\\t')\n",
    "    colnames(seq_matrix)=c('A','C','G','U')\n",
    "    seq_matrix<-t(seq_matrix)\n",
    "    ggplot() + geom_logo(seq_matrix, method = 'prob') + theme_logo() +\n",
    "    theme(axis.text.x = element_blank(),panel.spacing = unit(0.5, 'lines'),\n",
    "        axis.text.y = element_blank(),axis.title.y = element_blank(),axis.title.x = element_blank(),\n",
    "        plot.title = element_text(hjust = 0.5, size=30, face='bold'),legend.position = 'none') \n",
    "}\n",
    "\n",
    "figs <- lapply(demo_tissue, plot_data1)\n",
    "\n",
    "print(ggarrange(plotlist=figs,ncol=3))\n",
    "\n",
    "pdf('figures/Figure2_S3A_seqlog.pdf',width=14,height=4)\n",
    "print(ggarrange(plotlist=figs,nrow=1))\n",
    "dev.off()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "682131e6",
   "metadata": {},
   "source": [
    "# Figure 2C,D,E "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "df2c7aff",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Load xml in xxx to igv and export as SVG, then open SVG file in Adobe Illustrator"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a00a58f7",
   "metadata": {},
   "source": [
    "# Figure 2F"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "fcc35aa7",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"CirculatorySystem\"\n",
      "[1] \"DigestiveSystem\"\n",
      "[1] \"GlandularTissues\"\n",
      "[1] \"ImmuneSystem\"\n",
      "[1] \"NervousSystem\"\n",
      "[1] \"Other\"\n",
      "[1] \"ReproductiveSystem\"\n",
      "[1] \"RespiratorySystem\"\n",
      "[1] \"UrinarySystem\"\n"
     ]
    }
   ],
   "source": [
    "%%R\n",
    "### Generate significant groups\n",
    "intensity_all<-read.table('m6A_data/peak_intensity_m_1_renamed_gmean.csv',sep='\\t',header=T,row.names=1)\n",
    "intensity_h<-intensity_all[tau_idx[tau_idx$tau_index>=cut['90%'],]$X,]\n",
    "intensity_h<-intensity_h[complete.cases(intensity_h),]\n",
    "\n",
    "spl<-t(as.data.frame(strsplit(colnames(intensity_h),'_')))\n",
    "colnames(spl)<-c('Group','Age','Tissue')\n",
    "tissue_group<-as.vector(spl[,'Group'])\n",
    "tissue_age<-as.vector(spl[,'Age'])\n",
    "\n",
    "group_color=group_color=c(\"#386cb0\",\"#fdb462\",\"#7fc97f\",\"#ef3b2c\",\"#662506\",\"#a6cee3\",\"#fb9a99\",\"#984ea3\",\"#ffff33\")\n",
    "names(group_color)<-unique(tissue_group)\n",
    "\n",
    "column_ha=HeatmapAnnotation(\n",
    "  Age=tissue_age,\n",
    "  Group=tissue_group,\n",
    "  col=list(Age=c(\"Adult\"=\"pink\",\"Fetal\"=\"yellowgreen\"),Group=group_color)\n",
    "  )\n",
    "\n",
    "intensity_m<-intensity_h[,FALSE]\n",
    "groups<-unique(tissue_group)\n",
    "\n",
    "'%!in%' <- function(x,y)!('%in%'(x,y))\n",
    "group_t_test<-function(y1,y2){\n",
    "    ret<-c()\n",
    "    for(i in c(1:nrow(y1))){\n",
    "        test_<-t.test(y1[i,],y2[i,],alternative='g')\n",
    "        p_<-test_$p.value\n",
    "        ret<-c(ret,p_)\n",
    "    }\n",
    "    return(ret)\n",
    "}\n",
    "which_grp<-function(x)groups[which.min(x)[1]]\n",
    "for(group in groups){\n",
    "    group_idx<-which(tissue_group %in% c(group) & tissue_age %!in% c('Fetal'))\n",
    "    non_group_idx<-which(c(1:66) %!in% group_idx & tissue_age %!in% c('Fetal'))\n",
    "    \n",
    "    y1<-intensity_h[,group_idx]\n",
    "    y2<-intensity_h[,non_group_idx]\n",
    "    print(group)\n",
    "    intensity_m[,group]<-group_t_test(intensity_h[,group_idx],intensity_h[,non_group_idx])\n",
    "}\n",
    "intensity_m$sig_grp<-apply(intensity_m,1,which_grp)\n",
    "write.csv(intensity_m,file='ts_data/significant_grp.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "09220081",
   "metadata": {},
   "outputs": [],
   "source": [
    "%%R\n",
    "####### Perform enrichment analysis #######\n",
    "sig_grp<-read.csv('ts_data/significant_grp.txt',row.names=1)\n",
    "\n",
    "CirculatorySystem<-rownames(sig_grp[sig_grp$sig_grp=='CirculatorySystem',])\n",
    "DigestiveSystem<-rownames(sig_grp[sig_grp$sig_grp=='DigestiveSystem',])\n",
    "NervousSystem<-rownames(sig_grp[sig_grp$sig_grp=='NervousSystem',])\n",
    "\n",
    "peak2gene<-read.csv('m6A_data/peak2gene.txt',sep='\\t',header=F,stringsAsFactors=F,row.names=1)\n",
    "write.csv(unique(peak2gene[NervousSystem,'V2']),file='ts_data/NervousSystem.txt',row.names=F,quote=F,col.names=NA)\n",
    "write.csv(unique(peak2gene[DigestiveSystem,'V2']),file='ts_data/DigestiveSystem.txt',row.names=F,quote=F,col.names=NA)\n",
    "write.csv(unique(peak2gene[CirculatorySystem,'V2']),file='ts_data/CirculatorySystem.txt',row.names=F,quote=F,col.names=NA)\n",
    "\n",
    "system('python3 scripts/hyper_geometry.py ts_data/NervousSystem.txt ts_data/DigestiveSystem.txt ts_data/CirculatorySystem.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "7ec93057",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%R\n",
    "library(rlist)\n",
    "tau_idx<-read.csv('ts_data/tau_index_dir.txt',header=T,stringsAsFactors=F)\n",
    "sig_grp<-read.csv('ts_data/significant_grp.txt',row.names=1)\n",
    "cut<-quantile(tau_idx$tau_index,probs=c(0.1,0.9),na.rm=T)\n",
    "\n",
    "intensity_all<-read.table('m6A_data/peak_intensity_m_1_renamed_gmean.csv',sep='\\t',header=T,row.names=1)\n",
    "intensity_h<-intensity_all[tau_idx[tau_idx$tau_index>=cut['90%'],]$X,]\n",
    "intensity_h<-intensity_h[complete.cases(intensity_h),]\n",
    "\n",
    "spl<-t(as.data.frame(strsplit(colnames(intensity_h),'_')))\n",
    "colnames(spl)<-c('Group','Age','Tissue')\n",
    "tissue_group<-as.vector(spl[,'Group'])\n",
    "tissue_age<-as.vector(spl[,'Age'])\n",
    "\n",
    "group_color=group_color=c(\"#386cb0\",\"#fdb462\",\"#7fc97f\",\"#ef3b2c\",\"#662506\",\"#a6cee3\",\"#fb9a99\",\"#984ea3\",\"#ffff33\")\n",
    "names(group_color)<-unique(tissue_group)\n",
    "\n",
    "column_ha=HeatmapAnnotation(\n",
    "  Group=tissue_group,\n",
    "  Age=tissue_age,\n",
    "  col=list(Group=group_color,Age=c(\"Adult\"=\"pink\",\"Fetal\"=\"yellowgreen\"))\n",
    ")\n",
    "\n",
    "\n",
    "col_fun = colorRamp2(c(-4, 0, 4), c(\"blue\", \"white\", \"red\"))\n",
    "\n",
    "mat<-t(scale(t(as.matrix(intensity_h))))\n",
    "CirculatorySystem<-intersect(rownames(sig_grp[sig_grp$sig_grp=='CirculatorySystem',]),rownames(mat))\n",
    "DigestiveSystem<-intersect(rownames(sig_grp[sig_grp$sig_grp=='DigestiveSystem',]),rownames(mat))\n",
    "NervousSystem<-intersect(rownames(sig_grp[sig_grp$sig_grp=='NervousSystem',]),rownames(mat))\n",
    "sig_dif<-c(CirculatorySystem,c(DigestiveSystem,NervousSystem))\n",
    "mat<-mat[sig_dif,]\n",
    "clst_c<-hclust(as.dist(1-cor(mat, method=\"spearman\")),method='complete')\n",
    "\n",
    "Sig_Grp<-c(rep('CirculatorySystem',times=length(CirculatorySystem)),\n",
    "           rep('DigestiveSystem',times=length(DigestiveSystem)),\n",
    "           rep('NervousSystem',times=length(NervousSystem)))\n",
    "row_ha=rowAnnotation(\n",
    "  Sig_Grp=Sig_Grp,\n",
    "  col=list(Sig_Grp=c(\"NervousSystem\"=\"#662506\",\"CirculatorySystem\"=\"#386cb0\",\"DigestiveSystem\"=\"#fdb462\"))\n",
    ")\n",
    "\n",
    "align_to = split(seq_len(nrow(mat)), c(rep(1,times=length(CirculatorySystem)),\n",
    "                                       rep(2,times=length(DigestiveSystem)),\n",
    "                                       rep(3,times=length(NervousSystem))))\n",
    "\n",
    "keywords<-read.csv('ts_data/GO_enrichment.txt',sep='\\t',header=T,stringsAsFactors = F)\n",
    "keywords<-keywords[,c('Group','GO_Term','E_ratio')]\n",
    "keywords_n<-list()\n",
    "keywords_n<-list.append(keywords_n,'1'=keywords[keywords$Group=='CirculatorySystem',c('GO_Term','E_ratio')])\n",
    "keywords_n<-list.append(keywords_n,'2'=keywords[keywords$Group=='DigestiveSystem',c('GO_Term','E_ratio')])\n",
    "keywords_n<-list.append(keywords_n,'3'=keywords[keywords$Group=='NervousSystem',c('GO_Term','E_ratio')])\n",
    "fontsize_range = c(4, 16)\n",
    "gbl = lapply(names(align_to), function(nm) {\n",
    "  kw = keywords_n[[nm]][, 1]\n",
    "  freq = keywords_n[[nm]][, 2]\n",
    "  fontsize = scale_fontsize(freq, rg = c(1, max(10, freq)), fs = fontsize_range)\n",
    "  \n",
    "  word_cloud_grob(text = kw, fontsize = fontsize)\n",
    "})\n",
    "names(gbl) = names(align_to)\n",
    "margin = unit(8, \"pt\")\n",
    "gbl_h = lapply(gbl, function(x) convertHeight(grobHeight(x), \"cm\") + margin)\n",
    "gbl_h = do.call(unit.c, gbl_h)\n",
    "gbl_w = lapply(gbl, function(x) convertWidth(grobWidth(x), \"cm\"))\n",
    "gbl_w = do.call(unit.c, gbl_w)\n",
    "gbl_w = max(gbl_w) + margin\n",
    "panel_fun = function(index, nm) {\n",
    "  # background\n",
    "  grid.rect(gp = gpar(fill = \"#FFFFFF\", col = NA))\n",
    "  # border\n",
    "  grid.lines(c(0, 1, 1, 0), c(0, 0, 1, 1), gp = gpar(col = \"#AAAAAA\"), \n",
    "             default.units = \"npc\")\n",
    "  gb = gbl[[nm]]\n",
    "  # a viewport within the margins\n",
    "  pushViewport(viewport(x = margin/2, y = margin/2, \n",
    "                        width = grobWidth(gb), height = grobHeight(gb),\n",
    "                        just = c(\"left\", \"bottom\")))\n",
    "  grid.draw(gb)\n",
    "  popViewport()\n",
    "}\n",
    "col_fun = colorRamp2(c(-4, 0, 4), c(\"blue\", \"white\", \"red\"))\n",
    "\n",
    "colnames(mat)<-spl[,'Tissue']\n",
    "ht<-Heatmap(mat,name='z-score',cluster_columns=clst_c,cluster_rows = F,bottom_annotation = column_ha,show_row_name=F,\n",
    "            col=col_fun,show_row_dend =F,right_annotation = row_ha,use_raster=FALSE)\n",
    "ht = ht + rowAnnotation(keywords = anno_link(align_to = align_to, \n",
    "                                             which = \"row\", panel_fun = panel_fun, \n",
    "                                             size = gbl_h, gap = unit(2, \"mm\"), \n",
    "                                             width = gbl_w + unit(5, \"mm\"), # 5mm for the link\n",
    "                                             link_gp = gpar(fill = \"#FFFFFF\", col = \"#AAAAAA\"), \n",
    "                                             internal_line = FALSE))\n",
    "# pdf('figures/Figure2F_tissue_specificity.pdf',width=14,height=8)\n",
    "draw(ht)\n",
    "# dev.off()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "77745cbb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Figure S3 "
   ]
  }
 ],
 "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": "3.10.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}