Multi sample analysis
We would need:
- A means of providing multiple paired end sequences using:
- direct input to the command line
- a list of samples with their relevant paths (for very large studies)
-
After sample-wise analyses, "representative genomes" can be selected using dRep (http://biorxiv.org/content/early/2017/02/13/108142)
-
Reads from each sample should be remapped to the "representative genomes" generated by dRep
-
Other possible enhancements in this regard:
- Choice for a combined assembly (i.e. pooing reads from multiple samples). This strategy might work well for smaller number of samples, technical replicates, analytical replicates and/or highly similar samples