Assembling multiple samples separately and merging in Trinity

[Lhouda S]

Hello,

I’m assembling RNA-seq data from the same tissue and species, with several biological replicates (each sequenced across multiple lanes of the same library).

Due to computational limits, assembling all replicates together in a single Trinity run might not be feasible. I’m considering two possible strategies and would appreciate advice on which is best practice:

1. Option A:

Select one representative biological replicate per category, combine those reads, and assemble a single reference transcriptome.

But my Concern is: the chosen replicate may not capture all transcript diversity or might be a bad representative,

2. Option B:

Run Trinity separately for each replicate and then merge the resulting assemblies to create a single refrence transcriptome.

However, i am not sure if this would be correct?  IBetween these two options, which is preferred when computational resources are limited? Do you have any other suggestions?

If per-replicate assemblies are merged later, is there a recommended or example merging approach ?

Any Trinity parameters you’d suggest for handling large data sets efficiently? 

Thank you!!