isoform_r.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "library(SGSeq)\n",
    "rm(list = ls())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sample_name <- c(\"day_0.Rep1\",\"day_3.Rep1\",\"day_5.Rep1\")\n",
    "file_bam <- c(\"day_0.Rep1.genome.bam\",\"day_3.Rep1.genome.bam\",\"day_5.Rep1.genome.bam\")\n",
    "\n",
    "si <- data.frame(sample_name,file_bam)\n",
    "colnames(si) <- c(\"sample_name\",\"file_bam\")\n",
    "\n",
    "path <- \"/project/Neurodifferentiation_System/NanoporeResults/IGV\"\n",
    "si$file_bam <- file.path(path,si$file_bam)\n",
    "\n",
    "si\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "si$sample_name <- as.character(si$sample_name)\n",
    "si$file_bam <- as.character(si$file_bam)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "bi <- getBamInfo(si)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "library(TxDb.Hsapiens.UCSC.hg19.knownGene)\n",
    "txdb <- TxDb.Hsapiens.UCSC.hg19.knownGene\n",
    "seqlevelsStyle(txdb) <- \"UCSC\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "chr <- 1:22\n",
    "chr <- as.character(chr)\n",
    "chr <- c(chr, \"X\",\"Y\")\n",
    "\n",
    "seqlevels_in_bam <- seqlevels(BamFile(bi$file_bam[1]))\n",
    "\n",
    "#sgfdb2 <- keepSeqlevels(sgfdb, seqlevels_in_bam)\n",
    "\n",
    "#txdb <- keepSeqlevels(txdb, chr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "txf_ucsc <- convertToTxFeatures(txdb)\n",
    "head(txf_ucsc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#type(txf_ucsc)\n",
    "#head(txName(txf_ucsc))\n",
    "#head(geneName(txf_ucsc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "sgf_ucsc <- convertToSGFeatures(txf_ucsc)\n",
    "head(sgf_ucsc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sgfc_ucsc <- analyzeFeatures(bi, features = txf_ucsc)\n",
    "sgfc_ucsc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "colData(sgfc_ucsc)\n",
    "rowRanges(sgfc_ucsc)\n",
    "head(counts(sgfc_ucsc))\n",
    "head(FPKM(sgfc_ucsc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df <- plotFeatures(sgfc_ucsc, geneID = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sgfc_pred <- analyzeFeatures(si, which = gr)\n",
    "head(rowRanges(sgfc_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sgfc_pred <- annotate(sgfc_pred, txf_ucsc)\n",
    "head(rowRanges(sgfc_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "seqlevels_in_bam"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "ir"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}