NEWS
genbankr 1.27.1
BUGFIXES
- parser now correctly reads the "D-loop" feature type (#22)
genbankr 1.3.2
BUGFIXES
- Circularity (from LOCUS header information) is now applied
to all sources in the file.
- Improved assertion related to id ordering in makeTxDbFromGenBank
which passed in previous R version but was failing in recent ones
genbankr 1.1.5
MINOR CHANGES
- Change vignette to use system file to retrieve sample file
in response to ROpenSci reviewer comment.
- Change vignette builder from knitr to rmarkdown in response
to ROpenSci reviewer comment.
genbankr 1.1.4
BUGFIXES
- Fix bug when mRNA features are present and not annototated with gene.
Addresses half of https://github.com/gmbecker/genbankr/issues/1 Will be
backported to 1.0.4
- Fix bug when joins include incomplete range (< or >).
Addresses second half of https://github.com/gmbecker/genbankr/issues/1
Will be backported to 1.0.4
genbankr 1.1.3
BUGFIXES
- Fix bug in .seqTypeFromLOCUS when accession has non-alphanumeric characters (will
be backported to version 1.0.3.
genbankr 0.99.9
USER FACING CHANGES
- use of genbankr and genbankr version are now listed in TxDb objects
created via makeTxDbFromGenBank.
BUGFIXES
- NEWS file entries are now in most-recent-first order.
genbankr 0.99.7
MAJOR USER FACING CHANGES
- gene_id is now taken from locus_tag when present (in any features of a
given type) and gene is now only used as a backup when locus_tag is
absent for all features of that type.
USER VISIBLE CHANGES
- gene_synonym added to set of attribute that are assumed multi-valued and
always returned as a CharacterList column (without a warning).
BUGFIXES
- Added Michael Lawrence as second author.
genbankr 0.99.6
USER VISIBLE CHANGES
- Added support for creating of TxDb objects from GenBankRecord objects
- Added support for genpept files to readGenBank and parseGenBank
BUGFIXES
- Fix bug where partial was not being passed down correctly when specified
in readGenBank call.
genbankr 0.99.5
MAJOR USER VISIBLE CHANGES
- Move to single-class model with nullable sequence slot, class name
changed to GenBankRecord. Removed methods, etc for old classes
- parseGenBank now accepts ret.anno and ret.seq to control whether the
the sequence and annotations are parsed.
- Change how repeated annotation fields within a single feature entry is
handled. Known multivalue fields (currently db_xref and EC_number)
*always* return CharacterList. Others return CharacterList in case of
duplicate entries only, with a warning.
BUGFIXES
- correctly support cases where arbitrary annotations appear more than once
within a single feature.
- fix regression problem with handling GenBank files with no sequence
information when ret.seq is TRUE
- fix problem where tests weren't being invoked properly during check
- More informative error message when non-existent filename is passed to
readGenBank or parseGenBank.
- Fix bug when *all* variation features in a file lack /replace (for
VRanges alt is a mandatory field).
genbankr 0.99.4
BUGFIXES
- Fix small documentation bugs to squash check warnings
genbankr 0.99.2
USER VISIBLE CHANGES
- Version number jump as part of Bioconductor submission process
- Added runnable examples to (most) help files
- Hook existing unit tests into Bioconductor testing harness
- Add fastpass to extract sequence, in the form of seq.only argument to
parseGenBank, and getSeq methods for GBAccession and GenBankFile classes
BUGFIXES
- Fix some accessor methods for GenBankFull which didn't call annotations()
to grab GenBankAnnot object before accessing slots.
genbankr 0.1.0
USER VISIBLE CHANGES
- Text argument now supports character vector of length 1 with embedded
newlines in addition to character vectors of length N without.
- Added GBAccession class and constructor
- If a GBAccession is passed to readGenBank's first argument (file), the
GenBank formatted content will be retrieved from NCBI using rentrez (if
available)
- rentrez added to Suggests
- added seqinfo capture and support for GenBankAnnot, GenBankFull, and
underlying G/VRanges
- Added support for parsing more than one GenBank file/content block at
once, e.g. when a GBAccession vector of length > 1 is passed to
readGenBank.
BUGFIXES
- Seqnames are no longer appended with ":1" when only one source
(/chromosome/plasmid/etc) is present in the file.
- Ensure correct (matching) transcript ids are manufactured when both mRNA
and CDS features are annotated in the file. Ensure generated transcript
ids include gene identity when known.
genbankr 0.0.8
USER VISIBLE CHANGES
- When gene label is NA in (any) CDS annotations, temporary grouping
variable is used to inferm "transcripts"
- non-valued attributes (e.g. "/pseudo") are now imported as logical
columns, with presence corresponding to TRUE and absence corresponding to
FALSE
BUGFIXES
- non-valued attributes are no longer appended to the value of the previous
attribute.
genbankr 0.0.7
USER VISIBLE CHANGES
- When "gene" label is missing in gene and CDS annotations (happens in
plasmid files), locus_tag is used as the gene identifier for both.
BUGFIXES
- Fix error when a subset of CDS annotations do not have translation
information.
genbankr 0.0.6
USER VISIBLE CHANGES
- Genbank files which do not contain origin sequence are now supported.
Variation features in such files are ignored with a warning