Closed victorlin closed 3 years ago
in the database it would be good to be more descriptive with what run
is. I agree sra_id
is the wrong name, we should go with run_id
. I don't like the term run
as it's ambiguous alone but it is the correct choice to use it.
Yeah run
is ambiguous but I suspect any database user should be familiar with Serratus being a SRA search tool, and if not then the column values should give it away. 2 tasks for me here:
sra_id
-> run_id
for all tables in the databaseThis isn’t an issue for serratus.io anymore, but I’ll cross reference and close this once the tasks are done.
Another thought: should database tables nsra
/psra
/rsra
be renamed to nrun
/prun
/rrun
?
naw that database names are fine because it refers to the actual SRA as a whole
There's a small mismatch between the new database and website routes in how they reference SRA run IDs.
sra_id
is the SRA run ID column across all tables. @ababaian this is from your SQL schema spreadsheet.run
is used in query URLs (e.g. https://serratus.io/explorer?run=ERR2756788). This is from previous discussion at https://github.com/serratus-bio/serratus-db/issues/2#issuecomment-666345859.We should pick one and be consistent. There are references to
serratus.io/explorer?run=
in the rebuttal so I want to get this ironed out.Opinion with example: take an official route https://trace.ncbi.nlm.nih.gov/Traces/sra/?run=ERR2756788. This is logical because it's scoped to
sra
+run
. For Serratus, I think it's reasonable to assume scope of SRA when usingrun
. Having/explorer?sra=ERR2756788
could prompt users to try/explorer?sra=ERX2769781
which doesn't make sense./explorer?srarun=ERR2756788
could work too.Also worth noting that database columns
family_name
/genbank_id
are simplyfamily
/genbank
in Serratus Explorer. Still different, but less of an annoyance to me.