# GATK 4.2.2.0
# MutectVersion 2.2

# Gatk commands ran after alignment of reads to genome of reference (hg38) with bwa-mem:

## MarkDuplicates ##
# Ran for tumor and normal samples

gatk MarkDuplicates \
--arguments_file mark_dup_list.txt \
--OPTICAL_DUPLICATE_PIXEL_DISTANCE 2500 \
-MAX_FILE_HANDLES 1000 \
--REMOVE_SEQUENCING_DUPLICATES false \
--DUPLICATE_SCORING_STRATEGY SUM_OF_BASE_QUALITIES \
--VERBOSITY INFO \
--TMP_DIR tmp/ \
-M ${sample}.duplicate_metrics \
-O ${sample}_merged_dedup.bam


## Samtools alignment stats and tags, outputfiles used in following steps ##
# Ran for tumor and normal samples

samtools sort ${sample}_merged_dedup.bam -o ${sample}_merged_dedup_sorted.bam

picard CollectAlignmentSummaryMetrics R=Homo_sapiens_assembly38.fasta \
I=${sample}_merged_dedup_sorted.bam \
O=${sample}_merged_AlignmentMetrics.txt &

picard SetNmMdAndUqTags  R=Homo_sapiens_assembly38.fasta \
I=${sample}_merged_dedup_sorted.bam \
O=${sample}_merged_dedup_fixed.bam


## BaseRecalibrator ##
# Ran for tumor and normal samples

gatk --java-options "-Xmx16g -Xms16g" \
BaseRecalibrator \
-I ${sample}_merged_dedup_fixed.bam \
-R Homo_sapiens_assembly38.fasta \
--known-sites Homo_sapiens_assembly38.known_indels.vcf.gz \
--known-sites resources_broad_hg38_v0_Homo_sapiens_assembly38.dbsnp138.vcf \
-L timomas_hg38.interval_list --ip 100 \
-O ${sample}_recal_data.table


## ApplyBQSR (base recalibration)
# Ran for tumor and normal samples

gatk --java-options "-Xmx16g -Xms16g" \
ApplyBQSR \
-R Homo_sapiens_assembly38.fasta \
-I ${sample}_merged_dedup_fixed.bam \
--bqsr-recal-file ${sample}_recal_data.table \
-L timomas_hg38.interval_list \
--ip 100 \
-O ${sample}_bqsr.bam


## MUTECT2 call somatic variants ##

gatk Mutect2 \
-R Homo_sapiens_assembly38.fasta \
-L timomas_hg38.interval_list --ip 100 \
-I $sample/${sample}_bqsr.bam \
-I ${sample_normal}_bqsr.bam \
-tumor $sample \
-normal $sample_normal \
-germline-resource somatic-hg38_af-only-gnomad.hg38.vcf.gz \
-pon normals_tc_hg38_pon.vcf.gz  \
--genotype-germline-sites true --genotype-pon-sites true \
--f1r2-tar-gz ${sample}_f1r2.tar.gz \
-O ${sample}_unfiltered.vcf
       

## LEARNREADORIENTATIONMODEL ##

gatk LearnReadOrientationModel \
-I ${sample}_f1r2.tar.gz \
-O ${sample}_read-orientation-model.tar.gz


## PILEUP ##
# Ran for tumor and normal samples

gatk GetPileupSummaries \
-I ${sample}_bqsr.bam \
-V small_exac_common_3.hg38.vcf.gz \
-L small_exac_common_3.hg38.vcf.gz --ip 100 \
-O tumor_getpileupsummaries.table


## CALCULATE_CONTAMINATION ##

gatk CalculateContamination \
-I tumor_getpileupsummaries.table \
-tumor-segmentation segments.table \
-matched normal_getpileupsummaries.table \
-O calculatecontamination.table
	    

## FILTERMUTECTCALLS ##

gatk FilterMutectCalls \
-V ${sample}_unfiltered.vcf \
-R Homo_sapiens_assembly38.fasta \
--tumor-segmentation segments.table \
--contamination-table calculatecontamination.table \
--ob-priors ${sample}_read-orientation-model.tar.gz \
--stats ${sample}_unfiltered.vcf.stats  \
-O ${sample}_filtered.vcf