---
file_transformation: 
changes: 
  - 2018-03-03 add MANIFEST files
  - 2018-09-15 make prefixing in fasta consistent with IDs in gff3 (roughly, s/tripr.MilvusB.v2.Tp57577_TGAC_v2_/tripr.MilvusB.gnm2.ann1./)
  - 2018-10-14 fixed the files tripr.MilvusB.gnm2.ann1.DFgp.cds_primaryTranscript.fna.gz tripr.MilvusB.gnm2.ann1.DFgp.protein_primaryTranscript.faa.gz to use appropriate primary IDs in headers (ie not the gene IDs) and also to correctly limit to a single isoform per gene; noticed that fastas are missing any representation for gene with ID=tripr.MilvusB.gnm2.ann1.gene29308 ; this appears to be due to a bug of some sort in gffread but as the CDS of that gene is only 12 bp long, it seems not worth worrying about any more than I already have
  - 2021-07-03 Added tripr.MilvusB.gnm2.ann1.DFgp.legfed_v1_0.M65K.gfa.tsv.gz
  - 2021-07-29 adf: generated transcript files using gffread (fixes https://github.com/legumeinfo/datastore-issues/issues/51). also added cds and protein entries for tripr.MilvusB.gnm2.ann1.mRNA30314 after noticing discrepancies in the counts traceable to the absence of this transcript from the other files