combine metagenomic assemblies using cat?

Question

[goatsrunfaster]

I am working on metagenomic project where I have replicates for each sampling location. I originally wanted to combine the forward and reverse .fastq files and create one assembly for each site, but the problem I am having is that the combined files are too large for assembly (I get unresolvable errors on my universities HPC related to temporary memory)

Is there any reason I cant assemble each replicate on its own and then just concatenate them into one larger assembly?

If there are better work around ideas here they are appreciated.

[andres.firrincieli]

Hey,

some assemblers are pretty demanding in terms of memory. I would try megahit. Alternativelly, ask for more RAM.

Is there any reason I cant assemble each replicate on its own and then just concatenate them into one larger assembly?

This is perfectly fine however, you might loose low coverage genomes/contigs.