---
title: "Make VCF of gambl SSMs from all lymphoma genes"
output: html_document
editor_options: 
  chunk_output_type: console
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

# intro

For each BL, DLBCL and FL pathology, create a VCF file with SSMs from the bunbled data, from all lymphoma genes, genome seq type, and grch37 projection.


# fetch ssms from myc

```{r}
# load libraries
library(GAMBLR.data)
library(dplyr)

# regions
all_lymphome_gene_regions <- GAMBLR.utils::gene_to_region(lymphoma_genes_comprehensive$Gene)

# metadata
meta_all <- get_gambl_metadata(seq_type_filter = "genome") 
meta_by_pathology <- filter(meta_all, pathology %in% c("BL", "DLBCL", "FL")) %>% 
  split(.$pathology)

# retrieve ssms
maf_by_patology <- lapply(meta_by_pathology, function(meta_by_pathology_i){
  get_ssm_by_regions(these_samples_metadata = meta_by_pathology_i,
                     regions_list = all_lymphome_gene_regions, 
                     basic_columns = TRUE, streamlined = FALSE)
})

# save maf table to file
maf_dir <- "/projects/rmorin/projects/gambl-repos/gambl-vsouza/vlad_results/bundled_allLymphomaGenes_genome_grch37/"
dir.create(maf_dir)
maf_files <- paste0(maf_dir, names(maf_by_patology), ".maf")
mapply(write.table, x = maf_by_patology, file = maf_files, 
       quote = FALSE, sep = "\t", row.names = FALSE)
```


# convert maf file to vcf files

```{bash}
# define variables
INPUT_MAF_DIR=/projects/rmorin/projects/gambl-repos/gambl-vsouza/vlad_results/bundled_allLymphomaGenes_genome_grch37

# activate conda enve
conda activate /home/vsouza/software/miniconda3/envs/vlad_env

# get maf files
maf_files=(${INPUT_MAF_DIR}/*)

# convert maf to vcf
vcf_dir=${INPUT_MAF_DIR}/vcf
mkdir ${vcf_dir}
for maf_file in ${maf_files[@]} ; do
  # maf2vcf
  perl /home/vsouza/repos/lcr-modules/modules/vcf2maf/1.3/src/maf2vcf.pl \
    --input-maf ${maf_file} \
    --output-dir ${vcf_dir}
  
  # get vcf file path
  vcf_file=${vcf_dir}/`basename ${maf_file}`
  vcf_file=${vcf_file%.maf}.vcf
  
  # sorted vcf file path
  sorted_vcf_file=${vcf_file%.vcf}_sorted.vcf
  bcftools sort \
    ${vcf_file} \
    -o ${sorted_vcf_file}
  rm ${vcf_file}
  
  # compress
  bgzip ${sorted_vcf_file}
  
  # index
  tabix -p vcf ${sorted_vcf_file}.gz
done
```


# copy vcf files to gsc download dir

```{bash}
VCF_DIR=/projects/rmorin/projects/gambl-repos/gambl-vsouza/vlad_results/bundled_allLymphomaGenes_genome_grch37/vcf
DL_DIR=/gsc/www/bcgsc.ca/downloads/morinlab/vsouza/bundled_allLymphomaGenes_genome_grch37/vcf

mkdir ${DL_DIR}

cp ${VCF_DIR}/*.vcf.gz ${DL_DIR}/
cp ${VCF_DIR}/*.vcf.gz.tbi ${DL_DIR}/
```