nanopore output files: demultiplexed or non

[oatesa]

This may be a little simple to question to aks but still learning the processes ... I am hoping to use multipl2/miniasm to perfrom de novo genome assembly on nanopore output files from mixed bacterial communities.

(1) With the minimap2 overlap script should i use the fastq files or demultiplexed fastq files? (or does this not matter?)

and

(2) for the miniasm part am i correct in assuming you would use the demultiplexed fastq file with the output file from minimap2 and do this for each sample?

Thanks in advance :)