1. Fully mask genome

a. customize a repeat library if do not have one

$ BuildDatabase -name dbname -engine wublast refSeq.fasta

$ RepeatModeler -pa 32 -engine wublast -database dbname

b. Identify segmental duplications in reference genome

$ WGAC_SD.pl -i refSeq.fasta -l repeat.lib.fasta -t 48

c. thoroughly mask reference genome and get a unique genome

$ thorough_mask.pl -i ref.fasta -r TEseq.fasta -t 48 -s SD_interval.bed

2. Mapping reads to unique genome

mrfast

3. calculate and normalize read depth for each sample

4. call CNVs and calulate gene copy number

5. compare allele frequency