The DifCover pipeline aims to identify regions in a reference genome for which the read coverage of a sample1 to the reference is significantly different from the read coverage of a sample2. “Significantly different” is determined by user defined threshold on a ration between average coverages of given samples. The pipeline allows to exclude from a consideration the under-represented fragments (with low coverage in sequencing of both samples) and/or the regions that carry repetitive sequences. Both cases can be misleading in the coverage analysis. The DifCover pipeline is specifically oriented to the analysis of large genomes and can handle very fragmented assemblies.