Removal of mitochondrial and chloroplast seqs from the assembly

Question

[Ксения]

Hello, dear colleagues,

I'm a student trying to work with plant genomic data. I have an assembly of a plant genome consisted of contigs and scaffolds and included about 5 % of mt and chlpDNA. Could you recommend me any tool that could remove these DNA?

I previously deleted the contamination from reads, not from assemblies, by bbduk, but, as I understood, it doesn't fit to this task.

Thank you very much for your help in this matter.

Answer 1

[shelkmike]

In such cases, I do the following:
1) Calculate read coverage of your contigs.
2) Find completely assembled mitochondrial and plastid genomes of closely related species in GenBank.
3) Align these genomes to your contigs by BLAST.
4) Contigs that have coverage much higher than the coverage of nuclear contigs and that have BLAST matches, likely belong to mitochondrial and plastid genomes. Taking coverage into account is important to distinguish mitochondrial and plastid contigs from nuclear contigs that have NUMTs and NUPTs.

I think, using BBDuk is a bad idea because it can remove reads belonging to NUMTs and NUPTs.

[GenoMax]

bbduk is going to remove reads based on the reference provided and criteria used. It does not know kind of sequences you are working with. It is not clear if OP used bbduk in filter mode or trim mode to remove reads before doing the assemblies. If that was done prior to assemblies then I am not sure why there are 5% mtDNA remaining in the assembly.

[shelkmike]

Mitochondrial genomes of plants consist mostly of non-coding regions which are highly polymorphic. Maybe the reference that OP used differed too much from the mitochondrial genome of the studied species.