Trusted contigs with metaSPAdes

[ohickl]

Hi,

are there any plans to implement the use of trusted contigs with metaSPAdes? If not, why is it not possible?

Cheers Oskar

[asl]

Hello

No, nothing like this is planned. Will it be possible for you to describe your setup when trusted contigs are available / used?

[ohickl]

We have metaG and metaT data from the same samples and would like to use the rnaSPAdes assembly as trusted-read input for the metaG assembly. I imagine it would substantially help the graph construction. For now I would use the metaT assembly as long read input. Would you recommend one option over the other for that or is there no difference in the gap closure and repeat resolution mechanism of pacbio and nanopore input?

[VGalata]

I would also be interested in how to use additional conitgs in metaSPAdes, i.e. whether --nanopore or --pacbio make more sense here since the option --trusted-contigs is not supported.

[asl]

--trusted-contigs is certainly a fully supported option. The typical usage of it is follows: you're having and old assembly of the same sample, but you happen not to have these reads around. So, this way you could "mix" this assembly instead of the second library.

However, metaSPAdes for some other reasons has a limit of 1 input library.

[VGalata]

Dear @asl,

I know that the option --trusted-contigs can be used in general. What I meant is that it cannot be used together with --meta. Which is why it would be good to know whether --nanopore or --pacbio should/could be used instead and whether one of them is better than the other.

[nvpatin]

I wanted to chime in and say I am interested in @VGalata 's question as well, and see if there has been any change since October 2020.