Once all mci have been computed, qod partitions the central genome into common (classes 2 and 3) versus unshared (class 1) regions. In a partition, every base belongs to a single region; hence, unlike mci, the regions of the partition cannot overlap. To partition the central genome, qod simply scans the endpoints of mci from left to right and starts a new region at each position where the set of mci covering this position changes compared to the previous one. Clearly, if a new mci starts at this position then there is one more possible alignment for this region. This step's time complexity is linear in the mci number, while its space complexity is proportional to the maximum number of overlapping mci. Usually, the latter is small compared to the number of mci and the procedure is fast. At the same time, qod records all alignments associated with a region, which will serve for annotation transfer and to compute inter-genomic distances.
By combining the number of mci covering a region and the number of possible multiple alignments of an mci, qod classifies the partition regions of the center regarding their similarity to the compared genomes into three categories: 1) unshared, those that cannot be aligned with all other genomes 2) similar but with a unique possible alignment or, 3) similar with several possible alignments with at least another genome. As partition regions do not overlap, the genome coverage of each category delivers an overall measure of the central genome similarity with the compared genome set. Clearly, the coverage of similar regions (class 2+3) measures the core genome size, while that of class 2 regions indicates how much of the genome is shared and likely orthologous, provided the similarity level is high enough to ensure homology. Note the difference between unshared and genome specific regions. For instance, if three genomes are considered, a genome specific region must be unshared in a 3-way and in all 2-way comparisons with that genome as a center. Both notions are nevertheless relative to the set of genomes under consideration.