{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "4ea103da",
   "metadata": {},
   "source": [
    "Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved. SPDX-License-Identifier: MIT-0"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ea103da",
   "metadata": {},
   "source": [
    "## Install required R packages"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "777e2cde",
   "metadata": {},
   "source": [
    "make sure to use R kernel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c3798984",
   "metadata": {},
   "outputs": [],
   "source": [
    "if (!require(\"pacman\")) install.packages(\"pacman\")\n",
    "\n",
    "pacman::p_load('R.utils', \"visNetwork\", \"data.table\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff3a0377",
   "metadata": {},
   "source": [
    "## Sync variant calling files to local directory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e852cc48",
   "metadata": {},
   "outputs": [],
   "source": [
    "system('aws s3 sync s3://roda3/results_breast_cancer/Annotation/ ./results-vcf', intern=TRUE)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7685f3b1",
   "metadata": {},
   "source": [
    "## Extract mutations from variant calling files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fde2e0ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "library(data.table)\n",
    "library(dplyr)\n",
    "library(stringr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "327dd140",
   "metadata": {},
   "outputs": [],
   "source": [
    "filenames <- Sys.glob(\"./results-vcf/*/VEP/*ann.vcf.gz\")\n",
    "filenames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d9d3a240",
   "metadata": {},
   "outputs": [],
   "source": [
    "extract_variants_snp <- function(filename){\n",
    "base = basename(filename)\n",
    "base = str_extract(base, \"SAMPLE_\\\\d\\\\d+\")\n",
    "\n",
    "dt = fread(filename, sep=\";\", header=FALSE)\n",
    "dt_mutations = dt %>% dplyr::filter(grepl(\"BRCA2\",V1)) %>%\n",
    "dplyr::mutate(gene = gsub(\".*MODIFIER|\",\"\",V1)) %>%\n",
    "                dplyr::mutate(gene = gsub(\"ENSG.*\",\"\",gene)) %>%\n",
    "                dplyr::mutate(SNP = gsub(\".*rs\",\"rs\",V1)) %>%\n",
    "                dplyr::mutate(SNP = stringr::str_extract(SNP, \"rs[0-9]+\")) %>%\n",
    "                dplyr::select(gene,SNP) %>%\n",
    "                dplyr::mutate(gene = gsub(\"\\\\|\",\"\",gene)) %>%\n",
    "                dplyr::filter(SNP != \"NA\") %>%\n",
    "                dplyr::filter(gene == \"BRCA2\") %>%\n",
    "                dplyr::mutate(name = base)\n",
    "dt_mutations\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1c73101",
   "metadata": {},
   "source": [
    "Extract and show mutations of single file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c638ca3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "extract_variants_snp(\"./results-vcf/SAMPLE_01/VEP/Mutect2_filtered_SAMPLE_01_VEP.ann.vcf.gz\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1423b074",
   "metadata": {},
   "source": [
    "Extract mutations from all files and combine to data.table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3c3b02df",
   "metadata": {},
   "outputs": [],
   "source": [
    "dt_all = lapply(filenames,extract_variants_snp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "28a6b44f",
   "metadata": {},
   "outputs": [],
   "source": [
    "dt_links = rbindlist(dt_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d46b24ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "#dt_links[,from:=name]\n",
    "#dt_links[,to:=SNP]\n",
    "#dt_links[,c(\"name\",\"SNP\",\"gene\"):=NULL]\n",
    "#dt_links"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eb2f75a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "dt_links"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f7ee8d4",
   "metadata": {},
   "source": [
    "## Use graph visualization to show gene mutation to sample relationships"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "80eb226f",
   "metadata": {},
   "outputs": [],
   "source": [
    "require(visNetwork, quietly = TRUE)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75e41cef",
   "metadata": {},
   "source": [
    "define the nodes of the graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9d475e51",
   "metadata": {},
   "outputs": [],
   "source": [
    "samples = unique(dt_links[[\"name\"]])\n",
    "nodes_samples <- data.table(id = samples, \n",
    "                    label = samples)                             \n",
    "\n",
    "nodes_samples[,group:= \"sample\"]\n",
    "nodes_samples[,value:= 1]\n",
    "nodes_samples[,shape:= \"dot\"]\n",
    "nodes_samples[,title:= \"title\"]\n",
    "nodes_samples[,color:= \"#a6cee3\"]\n",
    "\n",
    "mutations = unique(dt_links[[\"SNP\"]])\n",
    "\n",
    "nodes_mutations <- data.table(id = mutations, \n",
    "                    label = mutations)                                \n",
    "\n",
    "nodes_mutations[,group:= \"mutation\"]\n",
    "nodes_mutations[,value:= 1]\n",
    "nodes_mutations[,shape:= \"dot\"]\n",
    "nodes_mutations[,title:= \"title\"]\n",
    "nodes_mutations[,color:= \"#fb9a99\"]\n",
    "\n",
    "nodes = rbindlist(list(nodes_samples,nodes_mutations))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "287cc62b",
   "metadata": {},
   "source": [
    "define the edges (links) of the graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a902e4f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "edges = dt_links\n",
    "edges[, from:=name]\n",
    "edges[, to:=SNP]\n",
    "edges[,arrows:= \"to\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d48fe49d",
   "metadata": {},
   "outputs": [],
   "source": [
    "network = visNetwork(nodes, edges, width = 1000, height = 900) %>% \n",
    "  visLegend(position = \"right\" ,useGroups = FALSE, \n",
    "            addNodes = data.frame(label = c(\"patient sample\",\"mutation (SNP)\"), shape = c(\"dot\",\"dot\"), color = c(\"#a6cee3\",\"#fb9a99\")))\n",
    "network"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "071ac7bd",
   "metadata": {},
   "source": [
    "## Export graph visualization to html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "911d7c91",
   "metadata": {},
   "outputs": [],
   "source": [
    "dir_analytics = \"./results-analytics\"\n",
    "dir.create(file.path(dir_analytics), showWarnings = FALSE)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ccdff33c",
   "metadata": {},
   "outputs": [],
   "source": [
    "network %>% visSave(file = paste0(dir_analytics,\"/network-mutations.html\"), background = \"white\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5918368",
   "metadata": {},
   "source": [
    "## Export graph structure to file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e52e4e82",
   "metadata": {},
   "outputs": [],
   "source": [
    "fwrite(nodes, paste0(dir_analytics,\"/nodes.csv\"))\n",
    "fwrite(edges, paste0(dir_analytics,\"edges.csv\"))\n"
   ]
  }
 ],
 "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": "4.1.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}