Comparing SPAdes assemblies to each other?

Question

[laurenkleine18]

I would like to compare three WGS datasets to each other, without using a reference genome. I am planning on running each set of paired end fastqs through SPAdes independently and then comparing the spades.contig files to each other.

I don't know of an alignment tool that doesn't require a reference genome as an input. I essentially want each of these files to be a reference to each other. I am also hoping for some kind of output that states the % aligned, similar to bowtie2.

[remon_dulos]

I'm also looking for a tool that can do this, any luck finding it yet?

[Joe]

Use minimap2 or mummer to align the genomes in an all-vs-all pairwise manner. You can compute statistics on the alignments easily enough, but they will probably give you some already.

[GenoMax]

What is the expected size of the genome? Besides the tools mentioned by @Joe, you may need to look at LASTZ if these are eukaryotic genomes.

[young_bioinformatician]

I think, Quast work for you. All you need to do is to check the parameters. It has got alignment parameter.

Answer 1

[Istvan Albert]

See if using the interactive dot plotter works for your case. I recently discovered this implementation of dot plotting and found extremely usefuyl.

It uses nucmer alignments hence it is quite performant (unlike most dot plotters), plus it is interactive and fast

https://dnanexus.github.io/dot/

Answer 2

[h.mon]

You don't say if you are working with prokaryotes or eukaryotes, nor the estimated genome size of your organism.

In case you have bacterial assemblies: for such a small number of genomes, Mauve is a great option for alignment and visualization. Better than aligning draft assemblies, it can align annotated draft assemblies in GenBank format (produced by e.g. Prokka, and you can visualize the correspondence of annotations, when you zoom in.

In case you have eukaryotic assemblies: D-Genies is a project similar to Dot suggested by Istvan Albert , but uses minimap2 for the genome to genome alignment, which should make it even faster than Dot.