Sequence Comparison for Fungal Genomes

Question

[akaur]

Hi, I recently started working on some fungal diseases. There are a number of pathogen species and subspecies causing the targeted disease and another disease with very similar symptoms, making it hard to evaluate them. Draft genome sequence of 7 of these sps. has been published recently and they have almost 98% sequence similarity. Each of the draft genome assembly has almost 50-100 contigs.

I want to compare these draft sequences and find unique sequences. what could be the best option to start with for such kind of analysis.

Thanks.

Answer 1

[dthorbur]

To find regions unique to each genome could be achieved through pair-wise whole genome alignments between all pair-wise combinations of species. There are a few whole-genome alignment tools available like mummer, so have a look around.

You'd then be able to identify regions unique to each reference assembly, and also put some of the shared regions in context.

However, you should consider sequencing and assembly methodologies and publication date into account when considering what is truly "unique". Some of the regions may not exist in an assembly due to things like only short reads being used to assemble the genome. As opposed to more contemporary approaches mixing long reads, short reads, optical mapping, etc...

Answer 2

[colindaven]

You can look at a pangenome construction tool like PGGB or Minigraph-cactus to build a pangenome of these sequences. Consider renaming the contigs well with the genotype name before construction.

Then take the resulting odgi file and use odgi pavs on it to find unique regions. https://odgi.readthedocs.io/en/latest/rst/tutorials/presence_absence_variants.html

It should be quite a quick process for small genomes.