# Diagnostics of unmapped reads 
# note: the mapping efficiently of our TagSeq data was only 30% 

# The following commands were run under interactive mode in HISAT2 directory
# (1) 
# for loop calling all unmapped reads (-f 4) in each .bam file and moving this filtered .bam file to 
# the directory unmapped_diagnostics with the new name 'basenameUNMAPPED.bam'

for file in *.bam; do samtools view -b -f 4 $file > unmapped_diagnostics/${file/.bam/UNMAPPED.bam}; done 

# (2) 
# convert each binary bam to a sam file 

# (3) 
# build a master fasta file with all unmapped reads (~100 bp) 

# (4) 
# run blastn diamond