Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- ## Each part of this pipeline was run separately
- ## This will probably not run as-is, I pulled each part of the pipeline into a single script for convenience
- #!/bin/bash
- base_dir=/scratch/asherrar/vg_test
- ref=/home/asherrar/t2t_sequence/v2.0/chm13v2.0.fa
- vardir=/scratch/asherrar/sgdp
- tmpdir=/scratch/asherrar/vg_tmp
- cd $base_dir
- variants=""
- for i in {1..22} X Y; do variants+="-v $vardir/SGDP.CHM13v2.0.chr$i.recalibrated.snp_indel.pass.vcf.gz "; done
- ## < Step 1 - Graph Construction > ##
- # construct base graph and index
- echo "constructing base graph"
- vg autoindex -r $ref -p $base_dir/t2t $variants -T $tmpdir -M 256G -V 2
- vg convert -p $base_dir/t2t.vg > $base_dir/t2t.xg
- vg convert -p $base_dir/t2t.xg > $base_dir/t2t.pg
- ## < Step 2 - Read Mapping > ##
- echo "mapping reads"
- vg map -m "long" -d $base_dir/t2t -f /scratch/asherrar/aicardi_longread/PAE47322.fasta > $base_dir/PAE47322.gam
- ## < Step 3 - Graph Chunking > ##
- mkdir -p $base_dir/chunks
- vg chunk -x $base_dir/t2t.xg -M -O pg -b $base_dir/chunks/t2t
- ## < Step 4 - Calling against Chunks > ##
- # run for each chunk:
- base_dir=/scratch/asherrar/vg_test
- mkdir -p $base_dir/chunk_steps
- mkdir -p $base_dir/chunk_vcf
- chunkdir=$base_dir/chunk_steps
- chunkname=chr${current_chunk}
- chunk=$base_dir/chunks/t2t_$chunkname.pg
- echo "calling variants on $chunkname"
- echo " - augmenting graph"
- vg augment -v -p -s -t 24 -m 4 -q 5 -Q 5 -A $chunkdir/${chunkname}_PAE47322_augmented.gam $chunk $base_dir/PAE47322.gam > $chunkdir/${chunkname}_PAE47322_augmented.pg
- echo " - calculating snarls"
- vg snarls $chunkdir/${chunkname}_PAE47322_augmented.pg > $chunkdir/${chunkname}_PAE47322_augmented.snarls
- echo " - packing graph"
- vg pack -x $chunkdir/${chunkname}_PAE47322_augmented.pg -g $chunkdir/${chunkname}_PAE47322_augmented.gam -o $chunkdir/${chunkname}_PAE47322_augmented.pack
- echo " - calling variants"
- vg call $chunkdir/${chunkname}_PAE47322_augmented.pg -r $chunkdir/${chunkname}_PAE47322_augmented.snarls -k $chunkdir/${chunkname}_PAE47322_augmented.pack -s PAE47322_reads > $base_dir/chunk_vcf/${chunkname}_PAE47322_readcalls.vcf
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement