recount::snaptron_query() is broken on BioC 3.16 and 3.17

[lcolladotor]

Hi,

Currently all snaptron-related tests at recount are failing. That is, https://github.com/leekgroup/recount/blob/7301c5fc09c968110d50617eef43a635b03fd4a2/tests/testthat/test-data.R#L176-L201.

Here's part of the output.

master* > snap <- snaptron_query(junctions)
2023-02-20 13:56:55 querying Snaptron
2023-02-20 13:56:55 processing results
2023-02-20 13:56:55 found no exon-exon junctions in Intropolis version srav1 matching your query: this version uses hg19 coordinates.
master* > snap_v2 <- snaptron_query(junctions_v2, version = "srav2")
2023-02-20 13:57:08 querying Snaptron
2023-02-20 13:57:08 processing results
2023-02-20 13:57:08 found no exon-exon junctions in Intropolis version srav2 matching your query: this version uses hg38 coordinates.
master* > snap_gtex <- snaptron_query(junctions_v2, version = "gtex")
2023-02-20 13:57:11 querying Snaptron
2023-02-20 13:57:11 processing results
2023-02-20 13:57:11 found no exon-exon junctions in Intropolis version gtex matching your query: this version uses hg38 coordinates.
master* > snap_tcga <- snaptron_query(junctions_v2, version = "tcga")
2023-02-20 13:57:12 querying Snaptron
2023-02-20 13:57:12 processing results
2023-02-20 13:57:12 found no exon-exon junctions in Intropolis version tcga matching your query: this version uses hg38 coordinates.
master* >     expect_equal(length(snap), 3)
Error: length(snap) not equal to 3.
1/1 mismatches
[1] 0 - 3 == -3

Maybe the syntax of snaptron changed and I need to update https://github.com/leekgroup/recount/blob/master/R/snaptron_query.R.

I don't think that this is related to the IDIES / AWS switch (which impacted leekgroup/recount#23). It's likely related to Stingray's move to IDIES.

Best,
Leo

[lcolladotor]

Hi,

I just wanted to make a note that this still seems to be an issue with BioC 3.19.

Best,
Leo