Open scottcain opened 6 months ago
Looking at https://github.com/alliance-genome/agr_blast_service_configuration/blob/main/conf/WB/databases.WB.WS292.json shows me that the tracks of interest are
"tracks" : [
"$ASSEMBLY_curated_genes",
"$ASSEMBLY_protein_motifs",
"$ASSEMBLY_classical_alleles"
],
@adamjohnwright : Do you have a preference for how these data a presented? I mentioned doing two tab delimited columns but I could also do json if you'd prefer, like:
{
"bioproject" : "$BIOPROJECT",
"tracks" : [
"$ASSEMBLY_curated_genes",
"$ASSEMBLY_protein_motifs",
"$ASSEMBLY_classical_alleles"
]
}
Probably both would work but I lean towards json
I'm going to suggest adding genblastg_cds_predictions as a track to add as well since the C. elegans Hawaiian strain doesn't have any of the other three tracks.
For the blast server, it would be good to have a file that gives a two column file (assemblies\tcomma delimited track list). Need to get the list of tracks desired (from @adamjohnwright)
The place to do this list building isn't immediately obvious--it could happen here (which would be after the config.json is created for each assembly) or in website-genome-browsers/jbrowse2, where the list would be created as the config.json is constructed. I'm leading towards the latter (the script that builds config.json already has S3 access built in to it--so it could just put this blast config file somewhere in the s3 bucket when placing the config.json.