- updating to GPL-2 license
- fixing bug in
extract_promotor_seqs_from_genome()
which causes genes on the minus strand that have end - start + 1 > 0
coordinates in their respective annotation file to be removed from the analysis. The new convention is now that genes will be filtered according to the strand
column
of the annotation file (hence: +
or -
) and it will be assumed that both, plus and minus strands use start
and end
coordinates according to their increasing direction, e.g. start = 1
and end = 5000
. (Many thanks for detecting this: Alexander Gabel)
- passing through
blast.path
to internal check function is_blast_installed()
- when performing
blast_best_reciprocal_hit()
, now users can specify tasks in both directions when selecting search_type = protein_to_nucleotide
(task = c("tblastn", "blastx")
) or search_type = nucleotide_to_protein
(task = c("blastx", "tblastn")
) #10. Many thanks to @wcfung14.
blast_test_reciprocal <- blast_best_reciprocal_hit(
query = system.file('seqs/qry_aa.fa', package = 'metablastr'), # protein sequence
subject = system.file('seqs/sbj_nn_best_hit.fa', package = 'metablastr'), # nucleotide sequence
search_type = "protein_to_nucleotide",
task = c("tblastn", "blastx"),
evalue = 0.000001,
output.path = tempdir(),
db.import = FALSE)