When using a large multi-fasta file (e.g raw sequencing reads) it'd be great if BRIG would use only a subsample of these for BRIG images, to save on run-time.
