[BioC] Looks like a bug in GenomicRanges::.breakTie

Paul Theodor Pyl paul.theodor.pyl at embl.de
Tue Mar 27 17:20:53 CEST 2012


I encountered (during a call to VariantAnnotation::locateVariants) an 
error that originated from GenomicRanges::.breakTie. From the source 
code of that function (svn revision 64480) it looks like equivalue can 
only ever have a length of 1 while equidist can be longer which breaks 
the call to DataFrame because the column lengths do not match.

Error in DataFrame(idx = c(mindist, eqidist), value = c(minvalue, 
eqivalue)) :
   different row counts implied by arguments

Enter a frame number, or 0 to exit
  1: locateVariants(insertions_granges, txdb19)
  2: locateVariants(insertions_granges, txdb19)
  3: .locateVariants(query, subject, ...)
  4: .intergenic(txCO, query, cache, map)
  5: follow(query, rng)
  6: follow(query, rng)
  7: .local(x, subject, ...)
  8: .findRanges(x, subject, ignore.strand, type = "follow", ...)
  9: .breakTie(sub1, res1, k1, dist1, sub2, res2, k2, dist2, type)
10: DataFrame(idx = c(mindist, eqidist), value = c(minvalue, eqivalue))

Kind Regards,

More information about the Bioconductor mailing list