dfdffa57a180281078f1158f56d01a78e82bc3f9
hiram
  Tue Sep 27 10:09:55 2016 -0700
note field is actually a longblob not varchar refs #13673

diff --git src/hg/lib/ncbiRefSeqLink.as src/hg/lib/ncbiRefSeqLink.as
index f295ed0..44f8ab8 100644
--- src/hg/lib/ncbiRefSeqLink.as
+++ src/hg/lib/ncbiRefSeqLink.as
@@ -1,22 +1,22 @@
 table ncbiRefSeqLink
 "Metadata for NCBI RefSeq tracks"
     (
     string id;                  "id for this gene or curated item"
     string status;              "Inferred, Model, Predicted, Provisional, Reviewed, Validated, Unknown"
     string name;                "gene name"
     string product;		"product"
     string mrnaAcc;             "transcript_id"
     string protAcc;             "protein_id"
     string locusLinkId;         "locus link identifier, from Dbxref"
     string omimId;              "OMIM identifier, from Dbxref"
     string hgnc;                "HGNC identifier, from Dbxref"
     string genbank;             "genbank identifier from Dbxref"
     string pseudo;              "'true' if pseudo gene, or n/a"
     string gbkey;               "genbank key: Gene, mRNA, ncRNA, rRNA, tRNA, etc..."
     string source;              "source: RefSeq, tRNAscan-SE, Gnomon, Curated Genomic, BestRefSeq"
     string gene_biotype;        "bio type: protein_coding, pseudogene, C_region, J_segment_pseudogene, other"
     string gene_synonym;        "list of synonym names"
     string ncrna_class;         "type of RNA: miRNA, lncRNA, snoRNA, etc..."
-    string note;                "other notes from genbank record"
+    lstring note;               "other notes from genbank record"
     lstring description;        "description from rna gbff record via gbProcess"
     )