Ribo-seq is still not up-to-date.

Also update submission pipeline to accomodate for possible absence of
libraries for an analysis (for instance, if they have been commented out
in the .yaml file because they have already been submitted).

Also changed directory layout and naming.

The read length to indicate in the data submission spreadsheet should be
the number of sequencing cycles. Some reads may be shorter than this.

It seems that submissions must be done in directories named after the
NCBI account of the submitter.
Meta data spreadsheet seem to be able to accomodate one type of library,
so one such directory is made per library type, with links to all data
of this type.
The workflow should fail if there are name conflicts due to repeated
file names once analysis directory is not present the path.

Files containg info about raw data in tab-separated format are created.
The content can be copy-pasted in the "RAW FILES" section of the
submission spreadsheet.

It creates links in a directory hierarchy from a configuration file.