How to reassembly metagenomics bins - need advice??

Question

[David]

Hello,

I´m trying to assembly a community Mock composed of 10 species at the moment. The steps I have follows so far are

1/ Trimming of reads (Q>20)

2/ Removal of phiX contaminants

3/ Removal of human contaminants

4 / Assembly using MEGAHIT

The assembly range from 88% (E.coli) up to 99%. I would like to try a reassembly but i´m not sure on the procedure. This is what i thought

5/ Create bins using Metabat

6/ Assembly each of the bins but i´m not sure which reads i should take. Should i take the reads from each bin and try a Megahit assembly again for each of the bins ? (will reads for each assembly will improve the assembly, not sure about this) ?? Also 2 bins may be lined to the same species so i might even get worse reassembly in such case ???

Thanks for your advice..

[Asaf]

If you map the reads to the scaffolds, divide them to bins accordingly and re-assemble you should get better results. Try to use other assemblers such as SPAdes, it might give better results. As for your concerns that a species will be divided into two bins, you should avoid this by restricting the binning.

Answer 1

[David]

I have tried SSPADES but it´s not performing well in my hands with metagenomics samples.