When I was creating an Index file using HIV virus genome sequence with STAR, I realized with a lot of help, if the genome sequence I am indexing is too small or big, I needed to set a "genomeSAindexNbases" value based on log2(GenomeLength)/2 -1 or it would cause segmentation fault.
I was wondering the mechanism behind why it wouldn't just make make a faulty or inadequate index file. Is it by the creator's design? or is it just programmatically not possible?
I hope someone might have information on this matter.