#1_fastqc
conda activate wes
cd ~/wes_cancer/project
cat config | while read id
do
	fastqc --outdir ./3.qc/raw_qc/ --threads 10 ./1.raw_fq/${id}*.fastq.gz >> ./3.qc/raw_qc/${id}_fastqc.log 2>&1 
done 

multiqc  ./3.qc/raw_qc/*zip  -o ./3.qc/raw_qc/multiqc

## trim_galore.sh
cat config | while read id
do
	fq1=./1.raw_fq/${id}_1.fastq.gz
	fq2=./1.raw_fq/${id}_2.fastq.gz
	trim_galore  --paired -q 28 --phred33 --length 30 --stringency 3 --gzip --cores 8 -o ./2.clean_fq  $fq1  $fq2 >> ./2.clean_fq/${id}_trim.log 2>&1
done

nohup bash trim_galore.sh &

cat config | while read id
do
	fastqc --outdir ./3.qc/clean_qc/ --threads 16 ./2.clean_fq/${id}*.fq.gz >> ./3.qc/clean_qc/${id}_fastqc.log 2>&1 
done 

multiqc  ./3.qc/clean_qc/*zip  -o ./3.qc/clean_qc/multiqc

#2_bwa
conda activate wes
cd ~/wes_cancer/data
gunzip Homo_sapiens_assembly19.fasta.gz
time bwa index -a bwtsw -p gatk_hg19 ~/wes_cancer/data/Homo_sapiens_assembly19.fasta
cd ~/wes_cancer/project
## bwa.sh
INDEX=~/wes_cancer/data/gatk_hg19

cat config | while read id
do
	echo "start bwa for ${id}" `date`
	fq1=./2.clean_fq/${id}_1_val_1.fq.gz
	fq2=./2.clean_fq/${id}_2_val_2.fq.gz
	bwa mem -M -t 16 -R "@RG\tID:${id}\tSM:${id}\tLB:WXS\tPL:Illumina" ${INDEX} ${fq1} ${fq2} | samtools sort -@ 10 -m 1G  -o  ./4.align/${id}.bam -
	echo "end bwa for ${id}" `date`
done

samtools index case1_biorep_A_techrep.bam
samtools view -h case1_biorep_A_techrep.bam chr17 | samtools view -Sb - > small.bam
samtools index small.bam

## stats.sh
cat config | while read id
do
	bam=./4.align/${id}.bam
	samtools stats -@ 16 --reference ~/wes_cancer/data/Homo_sapiens_assembly19.fasta ${bam} > ./4.align/stats/${id}.stat

	plot-bamstats -p ./4.align/stats/${id} ./4.align/stats/${id}.stat
done

cat config | while read id
do
	qualimap bamqc --java-mem-size=10G -gff ~/wes_cancer/data/hg19.exon.bed -nr 100000 -nw 500 -nt 16 -bam ./4.align/${id}.bam -outdir ./4.align/qualimap/${id}
done


#3_facets
#! bin/bash

InputFileDir=./4.align/
Sample=` ls ./4.align/ | awk -F '.' '{print $1}' | uniq | sed -s 's/.$//g' | uniq `
OutputDir=./5.pileup_output

for i in $Sample;do
  InputFile1=$InputFileDir/$i\N.sort.markdup.bam
  InputFile2=$InputFileDir/$i\O.sort.markdup.bam
  InputFile3=$InputFileDir/$i\T.sort.markdup.bam
  echo $InputFile1
  echo $InputFile2
  echo $InputFile3
  perl ~/biosoft/PileUp/snp-pileup.pl $InputFile1 $InputFile2 $OutputDir/$i\_NO_output.facets
  perl ~/biosoft/PileUp/snp-pileup.pl $InputFile1 $InputFile3 $OutputDir/$i\_NT_output.facets

done
####---------------------R
library(facets)
set.seed(114)
AllFilePath <- list.files('~/wes_cancer/project/5.pileup_output')
AllFileDir <- list.dirs('~/wes_cancer/project/5.pileup_output')
AllFilePath <- paste(AllFileDir, AllFilePath, sep='/')
AllFilePath
final_df <- matrix(nrow=length(AllFilePath), ncol=3)
final_df <- as.data.frame(final_df)
for (i in 1:length(AllFilePath)) {
  xx=preProcSample(AllFilePath[i])
  oo=procSample(xx,cval=200) 
  fit=emcncf(oo)
  df <- c(fit$dipLogR,fit$ploidy, fit$purity)
  final_df[i,] <- df
}

colnames(final_df) <- c('dipLogR','ploidy','purity')
AllFilePath <- list.files('~/wes_cancer/project/5.pileup_output')
AllFilePath <- gsub('(G.._..)_.*','\\1',AllFilePath)
rownames(final_df) <- AllFilePath
write.csv(final_df,file = 'facet_result.csv', row.names = T, col.names = T, quote = F)

#4_GATK
GATK=~/wes_cancer/biosoft/gatk-4.1.4.1/gatk

cat config  | while read id
do
	BAM=./4.align/${id}.bam
	if [ ! -f ./5.gatk/ok.${id}_marked.status ]
	then
		echo "start MarkDuplicates for ${id}" `date`
		$GATK --java-options "-Xmx20G -Djava.io.tmpdir=./" MarkDuplicates \
		-I ${BAM} \
		--REMOVE_DUPLICATES=true \
		-O ./5.gatk/${id}_marked.bam \
		-M ./5.gatk/${id}.metrics \
		1>./5.gatk/${id}_log.mark 2>&1 
		
		if [ $? -eq 0 ]
		then
			touch ./5.gatk/ok.${id}_marked.status
		fi
		echo "end MarkDuplicates for ${id}" `date`
		samtools index -@ 16 -m 4G -b ./5.gatk/${id}_marked.bam ./5.gatk/${id}_marked.bai
	fi
done

GATK=~/wes_cancer/biosoft/gatk-4.1.4.1/gatk
snp=~/wes_cancer/data/dbsnp_146.hg19.vcf.gz
indel=~/wes_cancer/data/Mills_and_1000G_gold_standard.indels.hg19.vcf.gz
ref=~/wes_cancer/data/Homo_sapiens_assembly19.fasta
cat config  | while read id
do
	if [ ! -f ./5.gatk/${id}_bqsr.bam ]
	then
		echo "start BQSR for ${id}" `date`
		$GATK --java-options "-Xmx20G -Djava.io.tmpdir=./"  BaseRecalibrator \
		-R $ref  \
		-I ./5.gatk/${id}_marked.bam  \
		--known-sites ${snp} \
		--known-sites ${indel} \
		-O ./5.gatk/${id}_recal.table \
		1>./5.gatk/${id}_log.recal 2>&1 
		
		$GATK --java-options "-Xmx20G -Djava.io.tmpdir=./"  ApplyBQSR \
		-R $ref  \
		-I ./5.gatk/${id}_marked.bam  \
		-bqsr ./5.gatk/${id}_recal.table \
		-O ./5.gatk/${id}_bqsr.bam \
		1>./5.gatk/${id}_log.ApplyBQSR  2>&1 
		
		echo "end BQSR for ${id}" `date`
	fi
done

GATK=~/wes_cancer/biosoft/gatk-4.1.4.1/gatk
snp=~/wes_cancer/data/dbsnp_146.hg19.vcf.gz
indel=~/wes_cancer/data/Mills_and_1000G_gold_standard.indels.hg19.vcf.gz
ref=~/wes_cancer/data/Homo_sapiens_assembly19.fasta
bed=~/wes_cancer/data/hg19.exon.bed

cat config  | while read id
do
	echo "start HC for ${id}" `date`
	$GATK --java-options "-Xmx20G -Djava.io.tmpdir=./" HaplotypeCaller -ERC GVCF \
	-R ${ref} \
	-I ./5.gatk/${id}_bqsr.bam \
	--dbsnp ${snp} \
	-L ${bed} \
	-O ./5.gatk/${id}_raw.vcf \
	1>./5.gatk/${id}_log.HC 2>&1
	echo "end HC for ${id}" `date`

done

cd ./5.gatk/gvcf
for chr in chr{1..22} chrX chrY chrM
do

time $GATK --java-options "-Xmx20G -Djava.io.tmpdir=./" GenomicsDBImport \
-R ${ref} \
$(ls ./*raw.vcf | awk '{print "-V "$0" "}') \
-L ${chr} \
--genomicsdb-workspace-path gvcfs_${chr}.db

time $GATK --java-options "-Xmx20G -Djava.io.tmpdir=./" GenotypeGVCFs \
-R ${ref} \
-V gendb://gvcfs_${chr}.db \
-O gvcfs_${chr}.vcf

done

$GATK --java-options "-Xmx20G -Djava.io.tmpdir=./" GatherVcfs \
$(for i in {1..22} X Y M;do echo "-I gvcfs_chr${i}.vcf" ;done) \
-O merge.vcf


## mutect.sh
GATK=~/wes_cancer/biosoft/gatk-4.1.4.1/gatk
ref=~/wes_cancer/data/Homo_sapiens_assembly19.fasta
bed=~/wes_cancer/data/hg19.exon.bed

cat config2 | while read id
do
	arr=(${id})
	sample=${arr[1]}
	T=./5.gatk/${arr[1]}_bqsr.bam
	N=./5.gatk/${arr[0]}_bqsr.bam
	echo "start Mutect2 for ${id}" `date`
	$GATK  --java-options "-Xmx20G -Djava.io.tmpdir=./"  Mutect2 -R ${ref} \
	-I ${T} -tumor  $(basename "$T" _bqsr.bam) \
	-I ${N} -normal $(basename "$N" _bqsr.bam) \
	-L ${bed}  \
	-O ./6.mutect/${sample}_mutect2.vcf

	$GATK  FilterMutectCalls \
  -R ${ref} \
	-V ./6.mutect/${sample}_mutect2.vcf \
	-O ./6.mutect/${sample}_somatic.vcf
	echo "end Mutect2 for ${id}" `date`

	cat ./6.mutect/${sample}_somatic.vcf | perl -alne '{if(/^#/){print}else{next unless $F[6] eq "PASS";next if $F[0] =~/_/;print } }' > ./6.mutect/${sample}_filter.vcf
done


#5_ANNOVAR
cd ~/wes_cancer/biosoft
cd annovar
nohup ./annotate_variation.pl -downdb -webfrom annovar gnomad_genome --buildver hg19 humandb/ >down.log 2>&1 &

$ ./annotate_variation.pl -downdb -buildver hg19 -webfrom annovar avdblist hg19_list/
$ cat hg19_list/hg19_avdblist.txt 

cat config | while  read id
do
echo "start ANNOVAR for ${id} " `date`
~/biosoft/annovar/table_annovar.pl ./6.mutect/${id}_filter.vcf ~/biosoft/annovar/humandb/ \
-buildver hg19 \
-out ./7.annotation/annovar/${id} \
-remove \
-protocol refGene,knownGene,clinvar_20170905 \
-operation g,g,f \
-nastring . \
-vcfinput
echo "end ANNOVAR for ${id} " `date`
done

#6_maf
cat config | while read id
do 
	grep -v '^Chr' ./7.annotation/annovar/${id}.hg19_multianno.txt | cut -f 1-20 | awk -v T=${id} -v N=${id:0:5}_germline '{print $0"\t"T"\t"N}'  >./7.annotation/annovar/${id}.annovar.vcf 
done
head -1 ./7.annotation/annovar/case1_biorep_A_techrep.hg19_multianno.txt| sed 's/Otherinfo/Tumor_Sample_Barcode\tMatched_Norm_Sample_Barcode/' >./7.annotation/annovar/header
cat ./7.annotation/annovar/header ./7.annotation/annovar/*annovar.vcf >./7.annotation/annovar/annovar_merge.vcf

###R 
rm(list = ls())
require(maftools)
options(stringsAsFactors = F)
## annovar
annovar.laml <- annovarToMaf(annovar = "./7.annotation/annovar/annovar_merge.vcf", 
                     refBuild = 'hg19',
                     tsbCol = 'Tumor_Sample_Barcode', 
                     table = 'refGene',
                     MAFobj = T)


#7_maftools----------R
rm(list = ls())
library(maftools)
library(data.table)
library(ggplot2)
mergemafData <- read.maf(maf = 'snv.maf')
#dir.create("maf")
plotmafSummary(maf = mergemafData, rmOutlier = TRUE, 
               addStat = 'median', dashboard = TRUE, titvRaw = FALSE)

oncoplot(maf = mergemafData, top = 50, fontSize = 0.4,
         showTumorSampleBarcodes = T, 
         sampleOrder = mergemafData@clinical.data$Tumor_Sample_Barcode, removeNonMutated = FALSE)