I have de novo assembled genomes of two strains that I have. One is wild type and the other is a mutant derived from the wild type through spontaneous mutation. I had sequenced the genome using Illumina Hiseq and Oxford Nanopore. de novo assembly was done using Nanopore reads. To compensate for any incorrectly called bases by the Nanopore technology, we mapped the genome with Hiseq reads using breseq which identified all the potential mutaions/SNPs. This creates a VCF file which let me go to all the specific positions in the contigs and correct them with the identified mutations leaving me with a "corrected" genome. Now that I had assembled both the genomes with this technique, I needed to identify the true differences or SNPs between them. So I cross mapped mutant reads over the wild type genome and vice versa. Theoretically both the methods should have given me exactly the same number of mutations but I'm not getting any such results. The WT genome against M reads does identify a huge portion of fragment which is missing in M that I know of, so I know it worked in this case but then the other SNPs are not identical in both the cases.
What could be the reasons for this? Is breseq's algorithm somehow flawed to handle it?
Please let me know if you want further clarification.
