Right now the file we're using for conveying metadata about input sequences is sort of oddly defined, and gets interpreted differently in the context of fixed timepoint runs versus merged timepoint runs, as far as multiplicity aggregation. The way things are set up now, we should have input csv (or whatever) files with a fixed schema along the lines of sequence,timepoint,multiplicity. Will need to replace the old wonky assumptions of the aggregation step accordingly, and coordinate with @psathyrella on getting datascripts to spit this information out and link up to the yaml.
Right now the file we're using for conveying metadata about input sequences is sort of oddly defined, and gets interpreted differently in the context of fixed timepoint runs versus merged timepoint runs, as far as multiplicity aggregation. The way things are set up now, we should have input csv (or whatever) files with a fixed schema along the lines of
sequence,timepoint,multiplicity. Will need to replace the old wonky assumptions of the aggregation step accordingly, and coordinate with @psathyrella on getting datascripts to spit this information out and link up to the yaml.