Determine whether GEO will let us submit multiplexed FAST5s

[lkwhite]

The Deeplexicon barcoded data can only be efficiently demultiplexed at the sequence level. GEO generally doesn't accept multiplexed raw data, though they say they'll sometimes take it for single cell experiments. I wrote the GEO team to see if they'll let us submit FAST5s that contain multiple samples even though their submission sheet says each raw file name should only be listed once in the samples section. We'll see what they say. Novoa lab's solution seems to have been to submit the demuxed fastq files to GEO and the corresponding still-muxed FAST5 files to ENA, but I assume we cannot do that.

[lkwhite]

All files except these FAST5s currently live in $HOME/whitel/RNARePore/data/raw_temp and are captured in the SRA template uploaded today.

[lkwhite]

Closing, data is now public on SRA under BioProject accession number PRJNA910992.