August - September Changes in BioC Devel
Document Actions
August - September 2007
ABarray
* : Added parameter for none normalization.
aCGH
* : Bumped for LazyLoad.
ACME
* : 1.3.2
=====
* Major changes to class structures. Now based on ExpressionSets.
* Still needs some significant cleanup of other functions, but
classes seem to
work as does do.aGFF.calc
* : * fixed write.sgr and started working on plot functions again.
* : not quite there yet
* : 1.3.0
=====
This backs out all the new changes since last release to fix up
things for new release.
adSplit
* : Bumped for LazyLoad.
affxparser
* : Committing in for backup. Will commit the DESCRIPTION with
increased version number later when I know then new stuff works.
I am running tons of redundancy testing on it right now using
aroma.affymetrix.
* : OK, updating the devel version of affxparser to v1.9.3 (from
v1.9.2). For the actual updates, see the release version v1.8.3,
because that contains the same updates.
Version: 1.9.3 [2007-08-16]
o See updated made to release v1.8.3 below.
o The only difference between v1.9.3 and v1.8.3 is the
modification
of findCfd() in v1.9.2.
Version: 1.9.2 [2007-07-27]
o This BioC devel version requires R v2.6.0 due to a change of
how strings are handled in C by BioC.
o Modified findCdf() such that it is possible to set an
alternative
function for how CDFs are located.
* : Updated to Fusion 1.0.9 and bumped the version number. I also
decreased the dependency to R (>= 2.5.0), simply because I don't
knwo if we need this...
* : Only comments for now, version is still to be bumped, which
will not happen before new AffymetrixDataTestFiles package is on
BioC.
o Added argument 'recursive=TRUE' to findCdf(). Note, the current
working directory is always scanned first, but never recursively
(unless explicitly added to the search path). This is to avoid
endless scans in case the search path has not been set.
o findFiles() now do a breath-first search in lexicographic
order.
o Removed default search paths cdf/ and data/cdf/. We do not want
to enforce a standard path.
o Now the examples (as well as test scripts) utilize data
available
in the new Bioconductor AffymetrixDataTestFiles package. This
means that R CMD check now runs much more tests, which is good.
o CLEAN UP: Removed many of the old testscripts/ scripts. They
are now under tests/.
o BUG FIX: If the destination file already existed, convertCel()
would
correctly detect that, but would give the name of the source
file.
* : The recent updates to findFiles() did not handle filename
patterns correctly.
* : Bumped affxparser (devel) to the following version:
Version: 1.9.5 [2007-09-16]
o Added argument 'recursive=TRUE' to findCdf(). Note, the current
working directory is always scanned first, but never recursively
(unless explicitly added to the search path). This is to avoid
"endless" scans in case the search path has not been set.
o findFiles() now do a breath-first search in lexicographic
order.
o Removed default search paths cdf/ and data/cdf/. We do not want
to enforce a standard path.
o Now the examples (as well as test scripts) utilize data
available
in the new Bioconductor AffymetrixDataTestFiles package. This
means that R CMD check now runs much more tests, which is good.
o CLEAN UP: Removed many of the old testscripts/ scripts. They
are now under tests/.
o BUG FIX: findFiles() was not robust against broken Unix links.
o BUG FIX: If the destination file already existed, convertCel()
would
correctly detect that, but would report the name of the source
file.
affy
* : On Robert's request, removed all notes from the man pages that
were of the form:
\note{This function is part of the Bioconductor project at
Dana-Farber
Cancer Institute to provide Bioinfomatics functionalities through
R.}
* : small vignette correction
* : fix in computeExprSet (trouble with named ids,
mostly an S4 thing)
bumped version number
* : - fixed geneNames to featureNames (spotted by Robert -
presumably
missing in the move to ExpressionSet)
- bumped version number
* : added citation
* : Fixing ppsetApply example
affycomp
* : Fix bracket pairs in Rd
* : Add Suggests splines to remove warning.
affycoretools
* : Use require, not library to load optional packages in code
* : Backed out xlsReadWrite stuff and made some other changes.
* : Changes to package structure and vignettes to allow package to
pass R CMD check.
* : Kludgy fix in linksBM() to account for the fact that Ensembl
uses different filter and attribute names for the HG_U133av2
chip. Also adjusted the default 'annot' argument to
probes2table() to match limma2annaffy() and vennSelect(), which
were previously modified because the output of aaf.handler()
changed.:
AffyExpress
* : added the AffyExpress package
Maintainer: Xuejun Arthur Li
affyio
* : fix for read_bpmap
* : fix problem with too many file handles left open problem for
binary CEL files
* : Fix crash in affyio due to extra call to fclose
Also updating required version for oligo
* : add support for reading command console (aka generic) format
cel files
Also will read gzipped versions of the same.
* : fix some small compiler warnings
* : fix uninitialized variable warning.
affylmGUI
* : Version 1.11.5 now displays Citation information in a GUI
Message Box as well as in the R Console.
* : minor code formatting changes and remove INDEX file from source
tree
* : affylmGUI version 1.11.6. htmlReport now uses parameters set in
the topTable process.
main.R and htmlReport.R code reformatted to suit me.
affyPLM
* : move affydata from "depends" to "suggests"
* : fix vignettes for dealing with "Dilution" dataset from affydata
* : more "R_rlm_interfaces.c" from affyPLM to preprocessCore
* : fix some compiler warnings
* : clean up include files. Try to differentiate similar named c
source code files in affyPLM and preprocessCore by renaming these
files PLM_*
* : restructed medianPM and median_logPM code
* : commit the deletion and addition of a file (should have been
previously committed)
* : small doc fix
* : Add Sys.sleep in MAplots.Rnw, to give ghostscript on Windows a
few seconds to finish before Sweave removes necessary temporary
files
* : resync affyPLM with changes in preprocessCore
affyQCReport
* : added \usepackage{underscore} on user request - let's see if it
breaks anything
* : minor addition of additional latex package dependency
altcdfenvs
* : - cut out dependency to splicegear
- fixes in the documentation
* : - Fixes an unmatched "}"
- Bumped version number
annaffy
* : Fixed problem with primitive methods not dispatching for
annaffy S4 objects
Removed all dependencies on non-db hgu95av2
Renamed "LocusLink" to "Gene" in aaf.handler()
Updated aafLocusLink URL
Updated aafCytoband URL
aafCytoband now uses Gene ID instead of Accession Number for
lookup
Updated aafGOItem URL to show detail view on AmiGO
* : Remove classes and functions related to SUMFUNC maps
THe SUMFUNC annotation map was never properly implemented and has
been
removed from the annotation data packages that are candidates for
the
next release.
* : Forgot to change some hgu95av2 require() lines in the examples
* : Quick fix for is_dbpackage internal helper function
This function or similar should be in the AnnotationDbi package.
This
is a quick fix to get things going. Thanks to Jim MacDonald for
the
bug report.
* : Fixed is_dbpackage() to deal with recent db_conn changes in the
ann db packages. Version bump.
* : one more try
* : even better (+ small reorganizing)
* : Small change in getHTML() to remove warnings that didn't affect
the output.
AnnBuilder
* : On Robert's request, removed all notes from the man pages that
were of the form:
\note{This function is part of the Bioconductor project at
Dana-Farber
Cancer Institute to provide Bioinfomatics functionalities through
R.}
* : Fixed a small problem where a list was expected.
List was not always being returned from function
getList4GO() So switched sapply() to an lapply().
* : Passed check under R-2.6.0
* : Use the Artistic License instead of LGPL as default license
* : Switch to Artistic
* : we need to move much of the orgName stuff to annotate, since it
is used
there and the package dependencies go from there to here
* : modified man page accordingly
* : removed one more arg
* : Added dire warning to users of this function.
annotate
* : use featureNames rather than geneNames accessor
* : On Robert's request, removed all notes from the man pages that
were of the form:
\note{This function is part of the Bioconductor project at
Dana-Farber
Cancer Institute to provide Bioinfomatics functionalities through
R.}
* : Fix useProbeInfo.Rnw
It needed to load the annotate package. There is also a problem
with
Biostrings::nchar when given an empty view. This patch works
around
the error by calling length instead.
* : Allow getAnnMap to understand "foo" and "foo.db" as chip name
* : Remove defunct function installDataPackage
* : Teach lookUp to use getAnnMap
This makes lookUp flexible enough for use with env and SQLite
based
annotation data packages.
* : Remove doc for removed function
* : fix doc for lookUp
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
* : GOTerms class moved to AnnotationDbi, GONode removed
In order for GO and GO.db packages to be used interchangeably and
for
function in the annotate package to function properly, we must
have a
single shared class representing GO terms. This patch moves the
GOTerms class definition to AnnotationDbi and removes references
to
the GONode class which was intended as a replacement.
* : buildChromLocation now uses getAnnMap and is db ann pkg
compatible
* : getAnnMap now uses inherits=FALSE in interal get call for
safety
* : DB compatibility fixes for chromLocation class
Changed slot types for probesToChrom and geneSymbols from
environment
to ANY since these need to be able to hold AnnotationDbi objects
when
using db-based ann pkgs. A class union might be a possible
solution,
but I'm concerned about possible dispatch issues that such a
union
might introduce.
* : Remove commented code for cleanup
* : ACCNUMStats now uses getAnnMap
* : use annotation() accessor instead of slot access
* : No need to require annotate in the annotate package.
* : probesByLL uses getAnnMap
* : Version bump, use standard license
* : expand doc for getAnnMap
* : Fixes made to htmlpage() to use xtable() instead of building
the table by hand. Also made changes to man pages and
prettyOutput vignette for clarity.
* : More refactoring to htmlpage() to ensure correct handling of
input.
* : moving a few pieces of code around so that we don't have
undefined globals
* : one more global squashed
* : fixed a typo in a warning
* : found lots of references to exprSet, geneNames etc -
not sure why these did not cause build warnings
* : removed reference to exprSet from comment
* : fixed typos and removed LocusLinks
AnnotationDbi
* : Started the flatten.R file where all code related to flattening
maps (db-based and environments) will go.
* : cosmetic
* : Documentation for new ALIAS maps.
This adds documentation for the newly
added alias maps. The documentation for
ALIAS2EG should also be applicable for the
HUMAN.DB and RODENT.DB packates.
* : Update for manual files on the ALIAS mappings.
This just adds a note to these man pages to
indicate that in fact we mean to represent ALL
the possible gene symbols in this mapping,
including those already listed in the symbol
map.
* : When building an ann db pkg, the user has now full control on
the Title that goes into the DESCRIPTION file. Used this new
feature in the master index file to set a customized title for
some of the custom array pkgs. Some more tweaks to R/flatten.R.
Version bump.
* : small update in the master index file
* : small fix in AnnDbPkgSeed class definition
* : some improvements to various non-chip based pkg templates +
some reorganization and updates to the master index file
* : Added the definition of 2 new maps: ALIAS2PROBE in
createAnnObjs.HUMANCHIP_DB() and ALIAS2EG in
createAnnObjs.HUMAN_DB().
* : Removed broken "initialize" method for AnnDbPkgSeed objects
(not needed anymore).
* : fixed some typos and inaccuracies in various template man pages
* : small fixes and improvements in getSymbolValuesForManPages()
* : forgot to add ALIAS2PROBE map to the list of predefined maps in
createAnnObjs.HUMANCHIP_DB()
* : createMAPCOUNTS() now generates a MAPCOUNTS map where the map
names are in alphabetical order
* : Adds documentation for the RODENT EG pkgs.
This adds documentation for the rodent EG
pkgs. This should be the same as the documentation
for the human pkgs. Except for the OMIM which
I see has snuck in here and will be subsequently
removed...
* : This adds documentation for the human EG pkgs.
This documentation is largely redundant with the
Rodent documentation. there are only a few minor
differences (OMIM). I am going to discuss this
with Herve since this is a bad thing.
* : Removes bad/uneeded documentation files.
As promised, this removes the unwanted OMIM doc
and other documentation from the repository.
* : Corrected misunderstanding about naming.
This corrects a misunderstanding that I had
about how we were going to name things in the
EG packages. Next I will correct this mistake
for the rodent packages.
* : Correct misunderstanding in pkg names.
this just corrects a misunderstanding in the
way that the maps should be named for the
rodent EG packages. We want to use 'EG' instead
of 'ENTREZID'
* : Added the no.docs arg to the "makeAnnDbPkg" generic.
* : Better name it no.man instead of no.docs. Updated the man page
for makeAnnDbPkg().
* : added new portable (and commented) HUMANCHIP_DB schema under
inst/extdata/
* : Added the index creation to HUMANCHIP_DB.sql. Added
RODENTCHIP_DB.sql
* : centralized general comments about the schemas in
DBschemas-README.TXT
* : Added HUMAN_DB.sql
* : added RODENT_DB.sql
* : Corrected a bug in the way these man pages reference ea. other.
This just fixes some bugs that were holdovers from the initial
grotesque copy paste that was done with these man files (since
they
really should be nearly identical to the chip based ones...)
* : Moved man pages for reverse maps into same documents as
forward.
This will hopefully help to make things more manageable. I will
document all future reverse maps along with the forward maps.
* : More merges in the man pages.
This just merges one more man page that I missed earlier and
fixes a
couple of documentation errors.
* : Added man page entries for reverse maps where appropriate.
For each possible reverse map I have added a man page entry.
* : Respect ifnotfound=NA in mget
This patch allows users to specify ifnotfound=NA in calls to
mget.
The support for ifnotfound is minimal. Only '=NA' is supported,
but
this is the common use case we see in existing code.
* : deal properly with _any_ possible 'ifnotfound' value provided
by the user
* : added GO_DB.sql (is not ready yet, there is actually much more
ugliness/breakage in our current GO schema than I imagined)
* : Move in-development schemas to extdata/schema_next
extdata/schema will soon hold _the_ current version of the
annotation
data package schemas. extdata/schema_next is a place for in
progress
updates and changes.
* : new GO_DB.sql is ready (tested with SQLite and PostgreSQL)
* : added KEGG_DB.sql
* : commented data columns with their "real world types" in
HUMANCHIP_DB.sql and RODENTCHIP_DB.sql
* : commented data columns with their "real world types" in
HUMAN_DB.sql and RODENT_DB.sql
* : Renamed folders: schema_next -> schemas_1.0, schema ->
schemas_tmp. Started to put degraded version of the 1.0 schemas
under schemas_tmp/
* : cosmetic
* : Added more "degraded" versions of the 1.0 schemas under
schemas_tmp/
* : 1 more degraded schema
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
* : started using portable FOREIGN KEY syntax (compatible with
MySQL+InnoDB and PostgreSQL)
* : more schemas with portable FOREIGN KEY syntax (compatible with
MySQL+InnoDB and PostgreSQL)
* : more schemas with portable FOREIGN KEY syntax (compatible with
MySQL+InnoDB and PostgreSQL)
* : added new real world type "sequence name" (SQL type:
VARCHAR(20)) and use it in the "chromosome_locations" tables
* : small update
* : removed the "accessions" table from the HUMANCHIP_DB and
RODENTCHIP_DB schemas
* : renamed schemas_tmp/ -> schemas_0.9/
* : all schemas now use portable FOREIGN KEY syntax (compatible
with MySQL+InnoDB and PostgreSQL)
* : added KEGG_DB.sql to schemas_0.9/
* : added inst/DBschemas/ folder where all the schema related stuff
that is currently under inst/extdata/ will go
* : moved all schema related stuff to the new place
* : added SchemaGuidelines.txt
* : added DataTypes.txt to schemas_0.9/ and schemas_1.0/
* : renamed README.TXT file -> 00README.TXT under schemas_0.9/ and
schemas_1.0/ and updated them
* : one more column renamed (disruptive change) in schemas_1.0
* : started adding YEAST_DB.sql to schemas_0.9/
* : Added YEAST_DB.sql to schemas_0.9/ (DataTypes comments not
ready yet). Some cosmetic polishing to many schemas.
* : cosmetic
* : version bump
* : Add make_eg_to_go_map helper function
When dealing with chip packages programatically, we often need to
map
Entrez to GO and the reverse. This patch provides two helper
functions to achieve this. When map pluggability comes online,
these
functions can go away. This is a *temporary* solution.
* : Added data types used in YEAST_DB to DataTypes.txt. Commented
data column definitions in YEAST_DB.sql with their "real world
types" (aka "external data types").
* : Added YEASTCHIP_DB.sql to schemas_0.9/
* : small update
* : Added ARABIDOPSISCHIP_DB.sql to schemas_0.9/
* : Replaced the "names" methods for AnnDbMap objects by the "keys"
methods (new generic). Made all the renaming that this change
implies. Still need some testing before version bump.
* : Tested the "names"-to-"keys" changes. Found some minor
problems. Fixed them. Version bump.
* : Another difference with the other chip-based schemas:
probes.probe_id can't be a PRIMARY KEY for ARABIDOPSISCHIP_DB (it
is not unique i.e. the same probe id can be linked to several AGI
locus IDs.
* : Started to implement the new "AnnDbBimap" class and subclasses
+ some preliminary rework needed in R/FlatBimap.R. Package will
be temporarily broken.
* : Migrated most AtomicAnnDbMap predefined objects to the new
AtomicAnnDbBimap class. Package should work again but migration
needs to be completed before version bump.
* : End of migration from AnnDbMap to AnnDbBimap. Still need some
testing/troubleshooting (vignette does not build). Version bump
anyway.
* : renamed AnnDbMap-envirAPI.R -> AnnDbBimap-envirAPI.R
* : Fix AnnDbMapDemo.Rnw: we were feeding NA as a key
* : - Added support for new FLYCHIP_DB schema.
- Updated ANNDBPKG-INDEX.TXT with entries for the 3 new fly chip
db packages.
- Fixed small bugs in the template man pages (introduced by
recent change of
signature of the as.list method for AnnDbBimap objects).
- Version bump from 0.0.94 to 0.1.0
* : - Added map subsetting methods "left.keys<-" and "right.keys<-"
(now the
"subset" method is just a convenient wrapper that allows
subsetting
of the left and right keys in a single operation).
- The "show" method for AnnDbBimap objects now makes the
difference between
a map and a submap (a submap being the result of subsetting a
map).
- Fixed a few methods that were broken for submap objects
("nrow",
and the "*.mappedKeys" methods).
- Some improvements to the SQL query caching mechanism.
- Some simplification and cleaning in many places.
- Version bump.
* : - Renamed BimapAPI0 interface -> Bimap
- Added the "keys<-" method.
- Moved strictly Bimap stuff from FlatBimap.R to new Bimap.R
file.
- Started to document the Bimap interface.
* : more fixes, cleaning and documentation
* : one less warning
* : Lots of renaming (methods and class slots). In particular, all
left/right
methods now have "L" or "R" in their name instead of "left" or
"right".
Overall naming is more consistent, readable and convenient (names
are
shorter). Version bump.
* : cosmetic
* : more renaming: L2Rpath -> L2Rchain, L2Rbrick -> L2Rlink
* : - Some internal changes and improvements (clear distinction
between tags and
attribs).
- Improved show method for Bimap objects.
- Version bump.
* : - renamed Lcolname -> Lkeyname, Rcolname -> Rkeyname, Tcolname
-> tagname
and Rattrib_colnames -> Rattribnames
- fixed bug in initialize method for FlatBimap
- fixed bug in flatten (still doesn't work for Go3AnnDbBimap)
* : renamed Rattrib_colname -> Rattribname and collabels ->
colmetanames
* : - Renamed Lkeyname/Rkeyname slots for class L2Rlink ->
Lcolname/Rcolname.
- Fixed bug in makeGo3L2Rchain() that was breaking the
Go3AnnDbBimap
objects.
- Fixed and improved .makeSQLchunks().
- Version bump.
* : depends on RSQLite >= 0.5-6
* : ARABIDOPSISCHIP_DB schema:
- renamed current ENZYME map -> ARACYCENZYME (probes to AraCyc
enzyme names
mapping)
- replaced data currently in ENZYME map by probes to EC numbers
mapping
Version bump.
* : HUMAN_DB and RODENT_DB schema:
- removed GENENAME2EG amp (as suggested by Marc)
ARABIDOPSISCHIP_DB schema (from last commit):
- removed ENTREZID map (was an alias for the ACCNUM map, which is
a mapping
between probes and AGI locus IDs)
* : - Added support for FLY_DB schema.
- Updated master index ANNDBPKG-INDEX.TXT.
- Version bump.
* : tolerate missing manufacturerUrl when making a .db pkg
* : with a warning
* : YEAST_DB schema:
- Renamed *2PROBE maps -> *2SGD
- Removed GO2ALLPROBES map
* : - Moved master index ANNDBPKG-INDEX.TXT from inst/extdata/ to
inst/extdata/GentlemanLab/
- Added batch-script.R to inst/extdata/GentlemanLab/
* : - added [ operator for Bimap objects
- fixed the links/count.links methods for AnnDbBimap objects
(still
needs some testing)
- fixed problem with NULL data frames with 0 rows that was
affecting
code like mget("all", GOBPPARENTS, ifnotfound=NA) (found in
GOstats)
- version bumped
* : Replaced previous dirty fix for "NULL data frames with 0 rows"
problem
with cleaner fix.
Version bump.
* : - Changed semantic of colmetanames().
- Replaced flatten arg 'with.Rattribs' by 'drop.Rattribs'.
- Fixed bug in flatten method for Go3AnnDbBimap objects.
* : FLYCHIP_DB/FLY_DB schemas: renamed FB_id -> flybase_id
* : simplified the compareAnnDataIn2Pkgs() framework
* : renamed links() -> edges(), count.links() -> count.edges()
* : clean up
* : - added "Rattribnames<-" for Bimap objects
- version bump
* : bumped all db packages versions from *.99.0 to *.99.1 in the
master index file
* : renamed edges() -> links(), count.edges() -> count.links()
(back to their original names, because of a conflict with the
"edge" generic in the graph package). version bump.
* : added TODO.TXT file
* : - map[["invalidkey"]] now returns NULL (at the cost of a
performance
penalty)
- small bug fixes
* : added man page for environment-like API
* : - polishing the "ls" method
- small fix to the "eapply" method
- added a "sample" method for environments in addition to the
method for AnnDbBimap objects
* : added (empty) man page for AnnDbBimap objects
* : - start new man page for Bimap methods (other than key methods)
- suppressed a few warnings
* : - next .db roaster will have "The Artistic License, Version
2.0" (on Robert's
request)
- version bump
* : - No more attempt to define "is.na" methods for my objects (the
semantic
of "is.na" for environments does not align with what I would
expect,
looks like a silly semantic to me anyway). So added new "isNA"
generic
with methods for Bimap, environment and ANY objects (for ANYi,
it's just
an alias for "is.na").
- More doc.
- Version bump.
* : GOTerms class moved to AnnotationDbi, GONode removed
In order for GO and GO.db packages to be used interchangeably and
for
function in the annotate package to function properly, we must
have a
single shared class representing GO terms. This patch moves the
GOTerms class definition to AnnotationDbi and removes references
to
the GONode class which was intended as a replacement.
* : Fix makeAnnDbPkg so that dirs are created before file.copy is
called.
This patch makes the dest location more explicit rather than
relying
on the (empty) template dir.
* : Fix the dir creation fix. Must not error out if it already
exists
* : Changes to support legacy version of YEAST.
These changes are only here to just support the old YEAST package
so
that it can be "spoofed" to a classic format.
* : Call paste one less time to speed up .toSQLStringSet
sep and collapse can occur in a single paste call. For long
vectors
this is about 2x faster than two calls to paste.
* : Add toSQLStringSet exported helper function
Since users cannot use sQuote for building SQL, we export a
helper
function for building SQL value lists.
* : Remove bad constraint from probes table on chip packages
This change only affects packages for rodents and humans. It
drops
the not null constraint from the accessions collumn which is
problematic since lousy platforms will sometimes have probes that
are
measuring "who-knows what". Users ought to have a right to know
when
they are using a probe like this...
* : also removed NOT NULL constraint on probes.accession column in
schemas_1.0/HUMANCHIP_DB.sql and schemas_1.0/RODENTCHIP_DB.sql
(replaced NOT NULL by NULL)
* : Updated dual man pages for Enzyme and GO.
This just updates these two man pages so that they are no longer
broken.
* : added FLYCHIP_DB and FLY_DB schema definitions (version 0.9)
* : Man page updates for EG style packages.
These updates make corrections to the double man pages in the EG
style
packages.
* : Fixed error in ACCNUM and Added page for Flybase IDs.
This just corrects a typo in the ACCNUM page and adds a man page
for
the new Flybase Map.
* : Adds FLYBASE man page for fly chip based packages.
This just adds a flybase man page to the fly chip based packages.
Note that for these packages, the man page should NOT contain a
reverse map since there is no reverse map planned for this
package.
* : Update CHRLOC to make the env less confusing.
This just adds a sentence to the description of this environment
so
that users will be less confused about what the data represents.
* : Adds raw documentation for the YEAST package.
These files still need to be gone through and modified to make
them
accurate, but these will be the man pages for this package.
* : Removed unwanted man pages from this repository.
* : Updates directions for some of the man pages.
This updates man pages for a number of the YEAST.DB man pages.
* : More updates to the YEAST Man pages.
It has now been decided that the "2SGD" maps will become "2ORF"
maps
this is to prevent future confusion with SGD systematic IDs which
are
different from the classic ORF IDs that the YEAST package has
always
used.
* : This completes the addition of the man pages.
This completes the addition of the man pages for the
org.Sc.sgd.db
package. (aka as the revamped YEAST package)
* : This updates the rest of the double man pages in the EG based
packages.
The EG based packages had to be rearranged to reduce the
duplication
and this change does that.
* : - YEAST_DB schema: Renamed *2SGD maps -> *2ORF
- Updated Version entries in GentlemanLab master index file in
order to produce
the *.99.3 roaster
- Version bumped to 0.99.0 (we're getting close!)
* : This updates the man pages for REFSEQ
The REFSEQ pages were still inelegantly doubled up for FLY.DB,
HUMAN.DB and RODENT.DB. This fixes that.
* : - Renamed db_file -> @ANNOBJPREFIX@_dbfile in all package
templates.
- Renamed db_conn -> @ANNOBJPREFIX@_dbconn in all package
templates and made
it a function instead of just a variable.
- Modified man page templates consequently.
- Version bump.
* : Replaces COMMON2SYSTEMATIC man page with COMMON2ORF.
* : Improved Title and Description in organised-based package
templates.
* : YEAST_DB schema: renamed COMMON2SYSTEMATIC map -> COMMON2ORF.
* : small update to our in-house batch script
* : small fix
* : ARABIDOPSISCHIP.DB template: removed man page for ENTREZID map.
Version bump.
* : Added man page for ARACYCENZYME.
This just adds a man page for the new ARACYCENZYME map.
* : Adds man pages for GO2ORF and GO2ALLORFs for org.Sc.sgd.db
* : Added man pages for GO2EG and GO2ALLEG to organism pkgs
Added for FLY.DB, RODENT.DB and HUMAN.DB
* : oops, small bug in the vignette
* : Removed these man pages since they are covered already
These man pages are already covered by double man pages GO.Rd in
each
of the organism based packges.
* : fixed problem in make_eg_to_go_map() introduced by recent
replacement (and renaming) of db_conn by a function
* : - HUMAN, RODENT and FLY schemas: added support for the
GO2ALLEGS map
- YEAST schema: added support for the GO2ALLORFS map
- modified schemas definitions (0.9 and 0.1) consequently
- version bump
* : bumped db roaster from .99.3 to .99.4 in master index file used
at the Gentleman Lab
* : Added new Package for Robert.
This just adds an entry to the shared file so that AnnotationDbi
can
make the new annotation package that Robert has requested.
* : Bug fixes in GO2ALL man pages
This just fixes some typos in these man pages.
* : Edit for Roberts Ovarian cancer package.
* : - Replaced "mappedkeys" method for "ANY" by 2 methods: 1 for
"environment"
and 1 for "vector" (it doesn't seem to make sense to try to
support more
than that).
- Small fix in "mappedkeys" methods for "environment" and
"vector" when the
argument is of length 0.
- Fixed checkMAPCOUNTS() so it also works for maps that are just
unamed
vectors (so in fact they are not maps) like the REJECTORF
pseudo-map
defined in the YEAST_DB schema.
- Version numped.
* : - fixed bug in "Lkeys<-" and "Rkeys<-" methods for FlatBimap
objects
- centralized all Bimap methods in Bimap.R
- added the "keyname" generic to the Bimap interface
- got rid of 'Lkeys' and 'Rkeys' arguments in toTable's signature
- added toTable method for FlatBimap objects
- some cleaning, reorganization and cosmetic improvements in many
places
- version bump
* : - Renamed AnnDbMapDemo.Rnw -> AnnotationDbiDemo.Rnw
- Removed AnnDbMapAPIdemo-labmeeting20070123.txt
- Fixed bug in .dbUniqueMappedKeys.cached.symbol() that was
breaking AnnDbBimap
objects of (left or right) length 0.
- Small improvements to "show" methods for FlatBimap and
AnnDbBimap objects.
- Got rid of the 'keys' argument in "as.list" methods for
AnnDbBimap objects.
* : oops, forgot to adapt the vignette to the new signature of
"as.list" methods
for AnnDbBimap objects
* : some fixes in the man pages
* : some addition to the man pages
* : more work on the man pages
* : more work on the man pages (no more 'R CMD check' warnings)
* : temporarily disable checking of GO.db in examples given in man
pages for checkMAPCOUNTS()
* : some doc improvements
* : cosmetic
* : cosmetic
* : removed go_obsolete.term_id column (useless and leads to
confusion) from the GO_DB schema
* : replaced some remaining occurences of "R environment" by "R
object" in the man pages of the ann db pkg templates
* : Dropped an superfluous reference to gopher6.
* : updated TODO.TXT
* : glups! I forgot to add man/Bimap.Rd
* : small update
* : small improvement
* : version bump
* : - Renamed "db" generic -> "dbconn" (for consistency with the
recent addition
of "@ANNOBJPREFIX@_dbconn" in the ann db pkgs) and exported it.
- Added "dbfile" generic (for consistency with the recent
addition of
of "@ANNOBJPREFIX@_dbfile" in the ann db pkgs) and exported it.
- Added man page for "dbconn" and "dbfile".
- Removed @conn slot from AnnDbObj objects (not needed anymore)
and exported
the AnnDbObj class.
- Version bump.
* : - Polished the AnnDbBimap class hierarchy: removed
AtomicAnnDbBimap (all
previous AtomicAnnDbBimap instances are now just AnnDbBimap
instances) and
renamed GONodeAnnDbBimap -> GOTermsAnnDbBimap (to align with
recent renaming
of the GONode class -> GOTerms).
- Added internal convenience generics and methods: "fromCol",
"toCol" and
"tagCol".
- Moved all the "as.list" methods from AnnDbBimap-envirAPI.R to
Bimap.R (this
is temporary only, they will go in a .R file of their own).
- Added an "as.list" method for Bimap objects. Reworked (mainly
simplified) all
"as.list" methods by using new "fromCol", "toCol" and "tagCol"
convenience
methods. Fixed a bug in "as.list" method for AnnDbBimap objects
(affecting
objects with at least 1 multi-valued right attribute).
- Reworked "as.character" method for AnnDbBimap objects.
- Version bump.
* : - Moved all the "as.list" methods from Bimap.R to new
BimapFormatting.R file.
- Also moved "from.colpos", "to.colpos", "fromCol", "toCol",
"tagCol",
"toLList", "toRList" and "toList" internal methods +
".formatList" and
".alignAnnList" internal functions + "as.character" method for
AnnDbBimap
objects from Bimap.R to BimapFormatting.R
- Got rid of the 'direction' argument in "from.colpos" and
"to.colpos" generics.
- Removed "from.keys" and "to.keys" generics and methods.
- Renamed "foldListOfLists" internal function -> ".toListOfLists"
and moved
it to BimapFormatting.R
- Modified documentation consequently.
- Version bump.
* : updated TODO.TXT
* : bumped db roster from .99.4 to .99.5 in master index file used
at the Gentleman Lab
* : - Turned "fromCol", "toCol", "tagCol", "from.colpos",
"to.colpos" generics
into normal functions and prefixed them with a "." (they are not
used out
of the file they are defined in).
- Version bump.
* : - Fixed bug in "as.list" method for AgiAnnDbMap objects (e.g.
as.list(agACCNUM)
was containing the NA and "multiple" values only).
- Fixed brace problem in MULTIHIT.Rd and ARACYC.Rd
- Version bump.
* : small addition to TODO.TXT
* : small addition to TODO.TXT
* : a few additions to TODO.TXT
* : more edits to TODO.TXT
* : improved comments for internal .formatList()
* : Fixed examples in all the ann db pkg templates ("as.list" was
still used with extra keys arg, which has gone away recently).
* : call callNextMethod() with no args in "revmap" method for
AnnDbBimap objects
* : small addition to TODO.TXT
* : - @ANNDBOBJ@_dbfile is now a function instead of a variable
- added "dbschema" and "dbInfo" generics with methods for
DBIConnection and
AnnDbObj objects (implementation is minimalist for now)
- added documentation (in particular, the _dbconn.Rd man page in
the db pkg
templates was heavily reworked)
- version bump
* : checkMAPCOUNTS("GO.db") is back in the examples section of
AnnDbPkg-checker.Rd
* : Small bug fix in the db pkg templates. Version bump.
* : small edit
* : Fixed MAPCOUNTS.Rd in db pkg templates. Version bump.
* : renamed last remaining occurences of db_file -> dbfile
* : makeAnnDbPkg() now calls 'chmod 444' on the sqlite file so it's
read-only
* : - added "dbmeta" generic and methods (no doc for now, will come
soon)
- first real implementation of the "dbschema" and "dbInfo"
methods
- version bump
* : small addition to AnnDbObj-class.Rd
* : small fix
* : Improved dbschema(). Version bump.
annotationTools
* : corrected minor bug in ps2ps.R and bumped version number
aroma.light
* : Version: 1.5.2 [2007-08-10]
o Added encoding to some Rd files.
o Package pass R CMD check R v2.6.0.
arrayMagic
* : Bumped for LazyLoad.
arrayQualityMetrics
* : Debug the figure margins of heatmap and localisation plots
legends.
* : Replace the EBImage functions by graphics functions. No
dependencies of EBImage anymore.
* : fix heatmap and localisation legend problem
* : Added a TODO and fixed some wordings in the Figure captions
* : Corrected man pages
Add the function addXYfromGAL to handle arrays with multiple
blocks
* : - Creation of functions to avoid code duplication
- Use of lattice for the heatmap
* : - debug the main R code
- small modifications of the legends
- suppression of genes mapping and pm.mm per array (maybe to
reimplement
later with lattice)
- small corrections of man page
* : - Improvement of the Vignette
- Small correction of the code
* : - add the correct argument log.it in AffyRNAdeg
- remove sample names from boxplots when to many samples
- use of try() to call the plots to avoid partial reports
- precision added for Covariate in the Vignette
* : Better handling of NA
* : suppress warnings of dev.copy after qcStats plot production
beadarray
* : - added new options to createBeadSummaryData (see "method" and
"trim" arguments) to allow mean, median, trimmed mean or
winsorized means to be calculated to summarise the values from
each beat type on an array.
createBeadSummaryData.c code edited to allow new options
- "which" argument changed to "what" in getArrayData(). Also
added new arguments ("method" and "trim") for use by
createBeadSummaryData when what="residG", "residR" or "residM" to
allow control over summarisation methods. Updated arguments of
getArrayData() when called in other bead level plotting
functions.
- Added new arguments to boxplotBeads, imageplot,
plotBeadDensities, etc to accept these new arguments to
getArrayData (when whatToPlot="residG", "residR" or "residM")
- added FILL=TRUE to read.table() used in readBeadSummaryData()
in response to message on BioC list from Ina Hoeschele about
error trying to read in BeadStudio version 3 data caused by line
having incorrect number of entries
- updated help pages
* : - removed returnTIFF.c (not used anymore).
- NAs or Infs in imageplot() and createBeadSummaryData() now
excluded (previously set to zero).
- log2.na() (not exported) used in getArrayData() to avoid
throwing up a
warning message when 0's or negative values are log2.
- when useImages=FALSE in readIllumina(), x-, y-coordinates read
from file are
now adjusted to start at 0,0 (useful for image plotting).
- readBeadSummaryData() has new argument ... to take additional
arguments to read.table (quote="", fill=TRUE sometimes needed for
BeadStudio
3 data). Reverses changes to readBeadSummaryData made in 1.5.10
and 1.5.7.
- new argument 'legend' in imageplot() to display range and zlim
on plot.
- new argument 'arrays' in arrayNames() to get the name of a
selected array/s.
- new function getVariance() returns the variance for each
bead-type on each
array from an ExpressionSetIllumina object (calculated from
se.exprs and
NoBeads slots).
- SnpSetIllumina doesn't store se.exprs for each channel -
removed code
which assigns the se.exprs to a SnpSetIllumina object in
createBeadSummaryData()
- fixed broken link in BSData.Rd help page.
- default annoPkg argument in readIllumina() is
"illuminaProbeIDs" (not actually an annotation package, but is
there to tell the user that ProbeIDs are used to identify each
bead type (not Illumina IDs available in existing annotation
packages)
- createBeadSummaryData(): The phenoData slot is set to have 1
column 'sampleNames' which is set to the array names if no
phenoData information has been set. sampleNames are set to either
the 'sampleNames' from phenoData, or the array names.
* : - phenoData in BLData object created by readIllumina() contains
1 column 'arrayName' when no targets information is specified.
- example BLData.rda object in package updated with phenoData
information.
* : - example BLData.rda object updated
* : - fixed bug in C code in calculateBackground (readBeadImages.c)
function (was over-running end of array). Also made backgroudSize
adaptive (was set to fixed size of 17x17 pixels in C code) and is
more careful when at an edge (previously set background to 700 if
pixels went off the edge).
- fixed bug in findAllOutliers() (introduced in 1.5.11 when &&
and || added to if statements. Inadvertantly added to which at
the end of finaAllOutliers())
- fixed bug in SAMSummary() BeadChipSummary() - couldn't handle
negative summary values and removed unknown 'v' variable assigned
when missing_arrays argument was specified. Removed
missing_arrays argument, and added arguments 'samID', 'chipID'
and 'stripsPerChip' to allow user to choose which SAM/BeadChip to
plot and specify the correct number of strips per chip. Function
is more careful about plotting the data in the appropriate
row/column position, and returns a message it the user clicks out
of range. Missing arrays are coloured white.
- removed unknown 'v' variable assigned when missing_arrays
argument was specified in plotOnSAM().
- getArrayData() now checks for out of range or NA 'array'
argument and returns a message
- createBeadSummaryData() checks for arraynm in all full list of
array names in case 'array' has not been specified as integer.
* : Added new versions of bead summary and bead level analysis
vignettes
beadarraySNP
* : Documentation
* : Improving import of Beadstudio data
* : bug fixed in R-code of Vignette
bgafun
* : Add new contributed package bgafun
Maintainter: Iain Wallace
bgx
* : sokal now shlib; sqrted mcse; changed rankByDE; read mumcse
into ExpressionSet; new z-statistic for ranking
* : Use sokal() rather than acf() (faster...) for rankByDE
* : Fixed allocation bug for returning output dir; reduced default
burnin and iter
* : All ext C functions in bgx.so now. Registered sokal(). Minor
fix to problematic cex parameters in plotDEHistogram
* : Clean up screen presentation when reading in output
* : You can now combine output directories using readOutput.bgx
* : Include own version of boost header files
* : Check if removing -march=i386 solves liverpool linking problem
* : Actually, check if -march=i486 works..
* : No need for README anymore, since we no longer depend on
external boost
Biobase
* : On Robert's request, removed all notes from the man pages that
were of the form:
\note{This function is part of the Bioconductor project at
Dana-Farber
Cancer Institute to provide Bioinfomatics functionalities through
R.}
* : Add missing includes
* : should correct the DBL_MAX compilation error
* : no need for DBL_MAX
* : fixed a bit of the error messages
* : should correct the use of REAL macros in matchpt, optimization
* : matchpt algorithm optimized for single argument calls, 10x
speed up
* : Time for a version bump so people using R devel >= 2007-08-06
will be able to install Biobase again (version 1.15.23 currently
online is broken for them).
* : bug fix in "corrected" matchpt, also matchpt(x, numeric(0) now
returns NA indexes and Inf distances
* : man page correction for matchpt followin warning in check
* : my God, something happened to my logic: reverted most of the
changes in matchpt for the last two days - they were buggy. sorry
for that
* : I updated the man page of matchpt
* : matchpt to support integer for x and y
* : I removed the silly old cache_old.
matchpt now returns a data frame (index=integer, distance=double)
with
row names.
* : updated man page
* : Just adding and removing this to verify that I got write
permission to the SVN.
* : Removed dummy.txt; verified I've got write permission to
SVN/Biobase.
* : Fixed copySubstitute() so it raises an error when it tries to
substitute a symbol by an NA.
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
* : PUBLIC:
o Added anyMissing().
o Added support for missing values of rowMedians().
DETAILS:
o Now rowMedians() has an optimized C-level implementation. There
is one implementation for real values and one for integers. This
way an integer matrix does not have to be coerced to a double
(which would add 200% memory usage). Moreover, at the C-level
there is also an argument 'hasna'; if hasna=FALSE, missing values
are not looked for resulting in a faster processing. At the R
level, this argument is currently hidden, but may be added if
found useful.
* : NChannelSet bug fix
- "_ALL_" should be required as a level in 'channel' varMetadata;
it
was ommited during consturction (so channel info was labelled
'NA'
instead of '_ALL_' by default) and during channel selection.
- quietly add 'channel' to varMetadata
* : MIAME and phenoData bug fixes
- MIAME 'show' method uses 'notes' instead of 'other', as this is
the
accessor
- (deprecated) phenoData had validity check in the wrong place.
* : Add sublist_extract C function and subListExtract at R level
subListExtract provides a fast way to extract a named element
from
a sublist of a list. Code that does something like:
sapply(L, function(x) x$foo)
can instead do
subListExtract(L, "foo", simplify=TRUE)
The unit tests are in subListExtract_test.R.
* : Improve error message for subListExtract edge case (0-length
input)
When given list() as input, subListExtract will return list(),
unless
simplify=TRUE, in which case an error is raised. The reasoning is
that when the user asks to simplify, they expect a non-list to be
returned. We can't determine what sort of vector they expect so
an
error is raised.
* : Biobase::l2e now errors when given a list with NA in names(L)
Prior to this patch, l2e would return an environment in which an
element with name NA in the input was added to the environment
with
key "NA". This situation now raises an error.
Error messages for bad arguments were also improved.
* : l2e now gives a warning if input list has duplicated names
* : Accomodate changes in base::package_version
- use numeric_version rather than package_version
- internal .as.numeric_version
- stronger checks on valid version comparisons
* : added commentary
* : esApply for ExpressionSet
* made esApply(X, MARGIN, FUN, ...) a generic
* method for exprSet same as previous functoin esApply
(inappropriately modifies FUN environment, breaking lexical
scope)
* method for ExpressionSet equivalent to
with(pData(X), apply(exprs(X), MARGIN, FUN, ...))
* Documentation to follow shortly
* : Further esApply revisions
- conform closely to original behavior
- update documentation
* : Correctly construct AnnotatedDataFrame with $<- and new symbol
adf$sym <-,
added sym to data but did not add varMetadata["sym",]. Likewise
pData(adf) <- sym
would not add varMetadata for new cols in sym
* : (Silently) remove names from dimnames of assayData
Previously, named dimnames on assayData elements would produce an
error when trying to 'harmonize' assay and feature/sample names.
The
new behavior no longer produces an error, and silently removes
names
from dimnames. Thanks to Benilton Carvalho for the original
report (in
affy) and Laurent Gautier for pointing toward ExpressionSet
* : fixed doc bug in openPDF -- it wrote of options(pdfViewer) but
no cap V is desired
biocSurvey
* : tools to get overview of class usage etc for bioc software
review
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
biocViews
* : Fix bug in writeTopLevelView, unbound variable caught via
codetools
* : Fix inheritance for BiocView class
Don't double inherit Htmlized. Not sure why this was there in the
first place and it was causing a method dispatch warning.
* : Update usage of sp.between to match latest changes in RBGL
* : Fix htmlDoc methods: include ... in signature
The method definition sig must have ... if the generic does.
* : The URL field is now a proper link in pkg "homepage" HTML
* : Remove unused htmlValue method
* : Add installation instructions (biocLite) to package homepages
* : Update URL to point to release views
* : Output valid XHTML
* : biocViews is now a Biocore Team maintained package
* : Added terms for new annotation data pkg type.
These changes add new package names, a new ChipManufacturer and a
Microarray type for CpG Island arrays.
* : Updates the version number for the new terms.
bioDist
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
biomaRt
* : Update of getBM with uniqueRows argument. getBM now returns by
default only unique rows
* : Removed ensembl_transcript_id from getGene and getGO output to
get gene-level annotation and reduce the redundancy in the
output. Converted 1:length code in for loops with seq.
* : Added uniqueRows argument to getLDS function. By default getLDS
will return unique rows only
* : Defined methods on the Mart class to replace most @ symbols in
the code
* : set show method for objects of class Mart
* : added setReplacementMethods to remove final @ symbols in code
* : Removed braces that were not needed in Rd files
* : Moved XML from Depends to Imports in the package DESCRIPTION
file and imported XML in the NAMESPACE
* : Put the require(RMySQL) call into a do.call statement to avoid
warning message by R CMD check
* : Rewriting the vignette
* : Rewriting the vignette
* : Update of man pages
* : Fix in getBM when choosing list as output
BioMVCClass
* : removed all exprSet refs
Biostrings
* : - Small improvements to writeFASTA() implementation and doc.
writeFASTA() is not a generic anymore: writing a BStringViews
object to
a FASTA file can now be done with write.BStringViews() (see
below).
- Added BStringViews-io.R file and started putting all
BStringViews IO related stuff.
Contains so far: FASTArecordsToBStringViews(),
BStringViewsToFASTArecords(),
read.BStringViews() and write.BStringViews().
Deprecated "BStringViews" method for "file" object in favor of
function
read.BStringViews().
- Renamed arg 'sep' -> 'collapse'.
* : Check OK. Version bump.
* : - Added the "desc<-" generic with a replace method for
BStringViews objects.
- Got rid of a few more TODOs in the doc.
* : Fixed bugs in "nchar" method for BStringViews objects.
Revisited semantic and implementation of the "==" methods for
BString and BStringViews objects. Got rid of more TODOs in the
man pages.
* : small fixes in the doc to make R CMD check happier
* : small fix in the vignettes
* : Completed doc for constructor-like functions views(),
adjacentViews() and BStringViews().
* : Added man page skeleton for Longest Common
Prefix/Suffix/Substring family of functions (work in progress).
* : Added man page skeleton for the new BStringPartialMatches class
(work in progress).
* : Check OK. Version bump.
* : added a few comments in naive_fuzzy_search() C code
* : fixed small bug in "nchar" and "width" methods for BStringViews
objects
BufferedMatrix
* : Fix MoveStorageDirectory() for windows build
* : add rowMedians() (should only be used in RowMode)
buildBioC
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
CALIB
* : fix warning because of function "[ <-".
fix bug of missing value in spike intensity and gene intensity.
fix bug of empty clone id name when normalizing data.
* : bumping version
Category
* : Add first draft of cb_test
cb_test is a flexible hypothesis testing tools for discovering
interestings chromosome bands relative to a selected gene list.
The
tools supports local and global tests which can be carried out in
a
top down or bottom up fashion on the tree of chromosome bands.
* : Deprecate isConditional generic
Replace use of isConditional() with conditional(). This resolves
strange warning messages and check failure due to an improper
reassignment of the conditional generic.
* : Support for org.XX.eg.db annotation packages
* : Create a helper function to make hyperGTest methods more
flexible
* : Modifications to methods for KEGG and PFAM data to include
hyperlinks to KEGG and PFAM (Sanger institute) websites in output
from htmlReport().
* : Add digits and summary.args arguments to htmlReport
This allows user to specify the number of display digits for
numerical
fields. htmlReport no longer accepts '...' and uses summary.args
instead to pass additional arguments to the summary method.
* : Moved htmlReport and summary methods for KEGG and PFAM data
from GOstats to Category. Also changed the signatures for the
htmlReport to coincide with the refactored htmlReport method.
* : Update examples to use hgu95av2.db
* : Move GO and KEGG from Depends to Suggests
Actually, we now suggest the *.db versions. The code should work
if either is available. A downside of this approach is that the
dependence is implicit (we don't call require). So I've added a
comment with require("GO"), etc to help remind us.
* : Fixup some man pages to remove warnings during R CMD check. No
version bump.
* : Added species abbreviation to html links for htmlReport of KEGG
data. Currently only hsa, mmu and rnu are used with hsa as
default.
* : GO2AllProbes,DatPkg-method now only queries for GO IDs that
exist
In making Category/GOstats compatible with db-based ann pkgs, we
encountered some problems with AnnotationDbi raising an error
when
asking for a key that doesn't exist. We plan to fix the env-API
to Do
What Environments Do, but in this case it was easy to have the
code
simply not ask for things that won't be there.
* : Fix references in Category vignette
Thanks to Francois Pepin for the report.
* : Add a TODO list for Category
* : Add getGoToEntrezMap_db, big speed up for hyperGTest "over"
This patch adds DB-specific code (with custom SQL, unfortunately)
that
speeds up the categoryToEntrezBuilder step of hyperGTest for
GOHyperGParams when doing a test for over representation. The
under
representation case is special in that we need to keep the GO IDs
that
don't have any of the selected Entrez IDs annotated at them and
so
that is left as a TODO. We also need to add DB-specific code for
the other gene set collections (KEGG, PFAM, etc). Also, the
method of
dispatching on DB-based or not ann pkgs is not ideal.
* : Add some comments to getGoToEntrezMap_db
* : Add support for under-representation for getGoToEntrezMap_db
* : Use subListExtract in getUniverseViaGo
* : Add Org.XX.egDatPkg to class exports
This class is used only internally for providing a common
interface to
annotation data coming from the new org.*.eg.db packages. But it
should be exported so that others can make extensions of the
Category-based code.
* : Add more doc for cb_test
* : Add some doc for *_test_factory internal funcs and fix args
* : Rename tree_iter => tree_visitor
* : Rename topdown_iter and bottomup_iter
They are now called topdown_tree_visitor and
bottomup_tree_visitor
* : Add doc for internal func tree_visitor
* : add mouse chromosome band parsing for HyperG
* : Rename chrBandInciMat => MAPAmat
chrBandInciMat is still available, but deprecated. The behavior
of
MAPAmat has changed (see man page). Any users of MAPAmat will
need to
revisit their code.
* : Remove base_cbind workaround since Matrix no longer clobbers
base::cbind
* : Rename Category::gseaperm => gseattperm
This function is hard-wired to use rowttests. Eventually, more
flexible version should be written, but for now we will rename to
make
clear what this function does (and doesn't do).
* : add helper function for the clean procedure of the chromosome
band definition
add helper function to get organism name
rename parsing functions using Hs and Mm instead of hsa and mmus
simplify Mm parsing function
* : Cleanup MAPcode.R changes
- Fix deprecation of cb_parse_band_hsa.
- Remove debug print statement
- Call func directly instead of using do.call
- Add checkChrOrg to consolidate checking of organism
* : Fix helper function isDBDatPkg so it will work with non Affy
chip packages
Now we check for the name ending in '.db' rather than relying on
getAnnMap (which searches for DB if available first). Not sure
this
is best.
* : Fix annotation<- method so that it updates the datPkg slot
* : Add error message for org.Sc.sgd.db not supported
Hopefully this is temporary.
* : Fix annotation<- method, value must be character, not ANY
This patch also resolves the Rd warning.
* : Add a show method for ChrBandTree objects
* : Add treeLevels and level2nodes generics/methods for ChrBandTree
objects
* : Use toSQLStringSet instead of sQuote to build SQL statements
sQuote should NEVER be used for building SQL or anything other
than
display messages. In non-C locales, sQuote now provides fancy
quotes
which are therefore useless for single quoting strings. We now
use
the newly added toSQLStringSet from AnnotationDbi to created SQL
value
lists. This should resolve the error seen in GOstats on Windows.
* : Some enhancements to cb_test
Allow different p-value cutoffs to be specified for next level
testing
and for determining significant nodes for a conditional test.
* : Change maintainer from Seth to Robert
Category, GOstats, graph, graphPart, hypergraph
* : fixed up Category to deal with db_conn changes
* : one more try
* : fixed a typo
cellHTS
* : typo correction
* : correction
* : typo correction
* : typo correction of Rd files
* : Adding an extra check for the controls
* : Correction of 'writeReport' related with the case when there
are no positive controls in the screen.
* : Improvement of the function that does the tooltips. It's now
much quicker!
* : Add a import function for plate files obtained from EnVision
Plate Reader
* : typo correction
* : Specifying the error messages in "readPlateData".
Just to be sure: added a consistency check test for well format
of column "Well" of the gene annotation file.
* : typo corrections
* : typo correction
* : Typo correction
* : Updates of the main vignette due to updates in packages biomaRt
and Category.
* : correction
cellHTS2
* : Starting the package.
* : More updates
* : more updates
* : more updates
* : correction
* : correction
* : More updates
* : completing the package
* : improvements
* : more updates
* : Cleaning up
* : Adding the modified files (cellHTS class now extends
NChannelSet class)
* : An attempt to (slightly) clean up to links to S4 classes from
Biobase in the man pages.
* : Add ROC class, import function and example set for HTanalyser
input files
* : correction of links in man pages
* : Add getTopTable function
* : Adding function getMeasureRepAgreement
* : Add function to determine Z'-factor and dynamic range
* : Adding writeReport and imageScreen function
* : add function to plot plate effects
* : Typo corrections, added biomart query data set
* : Add example of batch usage
* : Adding vignettes and man page for biomart data set
* : Typo correction
* : typo correction
* : Adding functio to make a screen plot with the plate
configuration.
* : Correction
* : Adding a function called convertOldCellHTS that converts an old
S3 class cellHTS object to S4 class cellHTS object(s) suitable
for the cellHTS2 package.
* : typo corrections
CGHcall
* : added the CGHcall package
Maintainer: Sjoerd Vosse
cghMCR
* : Man page updated.
clusterStab
* : Bumped for LazyLoad.
codelink
* : 1.5.16
- FIXED weird return in codCorrect().
- ADDED experimental C code in src/ to fast reading of Codelink
files. The
function readCodelinkFiles() works pretty much the same as
readCodelink() but
is much faster. Currectly it reads data as a Codelink object. In
addition, it
reads also the feature_id field, that can be used for creating
suitable
CodelinkSet objects.
- ADDED this comments for the first time to the svn log.
- ADDED summary method for Codelink objects. Actually it outputs
a summary of
the flags.
convert
* : Add converters to NChannelSet
- RGList, marrayRaw
- bug fix to marrayNorm -> ExpressionSet
- Also updated documentation to indicate that -> exprSet
converters
are to deprecated classes
* : Tweaked RGList convert
- previously misunderstood 'targets' to contain a list of file
names,
but is actually a data frame.
* : split the R source of the package into several files, so they
can be
more easily copied to other places
DEDS
* : removeUnknownFunction
DNAcopy
* : Fixed help files to address warning messages.
* : Fixed the unused variable warnings from gfortran
* : This is version 1.11.2 with p-value code for the change-points
(3 new files).
* : p-value code files didn't go through. Now they should.
* : removed ostat print statement & used k=2
DynDoc
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
* : Bumped for LazyLoad.
EBarrays
* : major update of EBarrays, initial checkin of version 2.0
EBImage
* : ImageMagick requirement version bump. configure to follow
* : new, fully rewritten vignette
* : no info
* : minor typo fixes in man
* : version check of ImageMagick in configure script
* : awk is strange, split($1, x, ".") did not work, for $1="6.2.4",
neither did split($1,x,"\."). workaround
* : added drawtext. still devel, but working :)
* : Added possibility to draw text annotations on images: drawtext,
drawfont functions
* : bug in zernike, corrected (was: no summation over x,y)
* : more accurate drawtext initialization and !NULL checks for font
and text labels in the c-module for drawtext. trying to resolve
man pages error on Mac BioC build
* : forgot the new Windows DLL, added
* : drawtext example exexcution prevented on Mac to pass check
until the problem with this particular build can be tested (no
hardware available). The problem is most likely to be in font
allocation.
* : minor configure output modifications
* : I edited the Image() function and fixed its man page
* : reverted aliases in man::constructors.Rd that mess up index;
corrected Sys.getenv("R_PLATFORM") in drawables.Rd to determine
MacOS
* : fully refurbished man pages
* : man page for package contents
* : version bump because of deprecated dot-functions
* : corrected one forgotten deprecated function
* : I added a display method for matrices, currently it just
converts into a
greyscale image
* : added: animate,array-method, following the code by Wolfgang for
display,array-method. Added usage clause in the help entry
* : added display options
* : colorize argument for display if used with IndexedImage
* : updated examples in man pages, checked up FIXME things
ecolitk
* : - added a label to chunk 16 (there seems to be a bug
with affy)
- bumped version number
edd
* : Bumped for LazyLoad.
exonmap
* : Changes in splicing.index() : added a parameter for use of
median/mean in gene level expression and fixed the method of exon
selection for gene-level estimate.
* : fixed plotting bug in plot.gene
* : fixed hardoded config directory to look for environment
variable specifying the path
* : fixings in splicing anova
explorase
* : * Add quick search bar and column visibility panel.
* Add Canberra distance, meant for metabolomics data (ie
non-detects)
* Support hierarchical clustering with all distance measures
(angle, cor,
euclidean, and canberra) instead of just euclidean.
* Pattern finding cleaned up. GUI supports changing the
"unchanged"
fraction and updating the results. Pattern finder no longer
creates a
new result column each time it is run on the same set of
conditions.
Selecting a large number of conditions activates the scrollbars -
GUI no
longer reaches off-screen.
* Limma frontend works with user-selected subsets of conditions.
* Long titles in column headers handled better with ellipsis and
tooltips.
* More robust file loading with warning dialogs if problems
detected.
* Misc other bugs fixed.
* Initial support for network data and visualization - disabled
for
release.
externalVector
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
fbat
* : replaced 'SaveImage: Yes' with 'LazyLoad: true'
* : deleted an unmatched right brace in 'getFounders.Rd'
* : fixed a few warnings about 'uninitialized' variables:
gcc -std=gnu99 -I/home/biocbuild/bbs-2.1-bioc/R/include
-I/home/biocbuild/bbs-2.1-bioc/R/include -I/usr/local/include
-fpic -g -O2 -Wall -c FBAT.c -o FBAT.o
FBAT.c: In function 'countLines':
FBAT.c:2142: warning: 'tmpnc2' may be used uninitialized in this
function
FBAT.c: In function 'oneNuclearFamily':
FBAT.c:1859: warning: 'sex2' may be used uninitialized in this
function
FBAT.c:1859: warning: 'sex1' may be used uninitialized in this
function
FBAT.c: In function 'decomposePed':
FBAT.c:1986: warning: 'i' may be used uninitialized in this
function
FBAT.c: In function 'checkoneNuclearFamily':
FBAT.c:1667: warning: 'sex2' may be used uninitialized in this
function
FBAT.c:1667: warning: 'sex1' may be used uninitialized in this
function
FBAT.c: In function 'checkdecomposePed':
FBAT.c:1590: warning: 'i' may be used uninitialized in this
function
FBAT.c: In function 'checkCompatible':
FBAT.c:1404: warning: 'start2' may be used uninitialized in this
function
FBAT.c: In function 'fbat':
FBAT.c:43: warning: 'start2' may be used uninitialized in this
function
* : fixed the problem of 'no visible bindings'. Thanks Dr. Lazarus
for the help!
* : Replaced the tab with 7 whitespace to avoid the warnings:
arning: Nonconforming tab character in column 1 of line 1
Warning: Nonconforming tab character in column 1 of line 4
Warning: Nonconforming tab character in column 1 of line 5
Warning: Nonconforming tab character in column 1 of line 6
Warning: Nonconforming tab character in column 1 of line 9
Warning: Nonconforming tab character in column 1 of line 13
Warning: Nonconforming tab character in column 1 of line 48
Warning: Nonconforming tab character in column 1 of line 49
* : Bumped for Lazyload.
flowCore
* : Adding unit tests from Josef and Ryan.
* : added some code to allow for varLabels accessing and setting
* : modify the namespace to export the methods
* : remove parenId from class definition of gates
* : Many changes. Mostly related to filterSets, but there is a
bunch of other stuff as well. See the flowcyt-devel mailing list
archives for info
* : Adding extra files
* : fixed summary problems
* : fix documentation
* : Improvements to the filterSet filter() code and summary() code
flowSet(...,phenoData) convencience function to wrap
as(,"flowSet") and phenoData<-() into a single
operation if desired.
split() now takes an optional flowSet=TRUE argument to get a
flowSet instead of a list. Allows split
operations to optionally be "smart" (e.g. manyFilterResult and
filterSet put an indicator of WHICH filter
generated the flowFrame)
manyFilterResult implementation improvements
manyFilterResult documentation improvements
* : Fix read.FCS for case when non-UTF8 encoding (e.g. FlowJo) is
used in an FCS file
flowQ
* : fixed some minor documentation issues
flowUtils
* : reimplementation and refactoring of gating-ML
* : Adding extra files
* : fixed missing html links
flowViz
* : separated out panel function for xyplot.flowset, and removed
rgl as a dependency
* : fixed problem in last commit
* : changes in filter boundary for rectangleGate
* : attempt at filter boundary method for combined filters
* : more fixes for separated panel function
* : added missing parentheses
* : fixed some small documentation issues
* : displayFilter included in panel function, updated docs
* : commented out parallelCoordinates.rgl
* : documented prepanel and panel functions, updated version
* : documentation updates
genefilter
* : Fix nsFilter bug when var.filter=FALSE
Thanks to Swann for the bug report during BioC2007.
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
* : Use subListExtract instead of sapply to improve filtering speed
* : genefilter now depends on recent Biobase
* : Fix use of subListExtract, need simplify=TRUE
GeneMeta
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
geneplotter
* : added na.rm=TRUE in range() for axis limit computation
* : Added multiecdf and multidensity methods for matrix
(they are quite trivial, they call formula with x~col(x))
* : I fixed a bug that occur when argument was list of length 1
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
* : better treatment of xlim in multidensity
* : fixed man pages and default argument values
* : col parameter now gets passed to points in order to control the
point colors
* : Defaults of xaxs and yaxs now match those of plot.default
* : Fixed typo: seq(alog=...) --> seq(along=)
* : Update examples to use sample.ExpressionSet
Several examples used sample.exprSet (deprecated exprSet);
sample.ExpressionSet is not a 1:1 replacement for sample.exprSet,
so
example results are different (generally, two groups rather than
three).
* : Makesense now uses getAnnMap
* : Deprecate Makesense 'eSet' method
'eSet' method does not make sense for the virtual class (no
guarantee
of exprs method), but derived classes with defined exprs could
have
relied on for this method so cannot be removed immediately.
Added method for ExpressionSet.
* : Deprecate identifyLines
Not currently exported from the name space, not used internally
* : Clarify documentation to indicate ExprssionSet, rather than
eSet
* : geneplotter now imports AnnotationDbi
This is needed so that internal calls to mget dispatch properly
when
the object is a db map from AnnotationDbi rather than an
environment.
* : simple changes to pass check on 2.6
* : bump version for vignette visualize.Rnw fix
* : make plotExpressionGraph example work
GeneR
* : Supression of &.segSet and |.segSet shortcut
* : changes
GeneSpring
* : Bumped for LazyLoad.
GeneticsBase
* : replaced 'SaveImage: Yes' by 'LazyLoad: true' to fix the
following warning message:
WARNING: true value of 'SaveImage' is deprecated in favour of
'LazyLoad: true'
* : deleted an unmatched right brace on line 7 in
'makeMarkerInfo.Rd'
* : replaced 'print.xtable' with 'print' since there is no
'print.xtable' in the library 'xtable'.
* : (1) replaced 'fun.lo' and 'fun.hi' in 'binsearch.R' with
'val.lo' and 'val.hi', respectively, since no visible binding for
global variables 'fun.lo' and 'fun.hi'.
(2) replaced 'fname' with 'gformat' in 'readGenes.R' since no
visible binding for global variable 'fname'
(3) replaced 'maker' with 'nMarker' in 'makeCCodesAndTables.R'
since no visible binding for global variable 'maker'
* : replaced the unknown encoding 'µA' with 'mu' in 'xbat.Rd'. Also
added a few '$' to distinguish
math symbol and text.
* : added 'haplo.stats' to 'Depends' section in 'DESCRIPTION'
* : fixed a bug: readGenes.ped.cpp:86: warning: 'tmpnc2' may be
used uninitialized in this function
* : removed the comment: #importFrom(combinat, rmultz2)
* : deleted 'combinat' from 'Imports' field
* : Deleted 'haplo.stats' from 'Imports' section to avoid compiling
error
* : We added the missed argument 'founderOnly' to fix a bug pointed
out by Dr. Kouros Owzar. Thanks Dr. Kouros Owzar!
* : added a check if founder only data is required at the beginning
of the the function 'desMarker'
* : replaced the function 'geneSet2Ped2' by 'geneSet2Ped' so that
more meaningful error info will be output when error occurs.
GeneticsDesign
* : Fix documentation for power.case.ctl: swap description of kp
and alpha
* : Integrate changes to power.casectrl from Michael Man.
* : Rename files 'power.casectl.{R,Rd}' and 'plot.power.casectl.R'
to match the function names. 'casectl' --> 'casectrl'
* : Remove binsearch() and use (bug-fixed) gtools::binsearch()
instead
* : Expose power.genotype.conti() function for estimating power for
genetice studies using baseline measurements via simulation.
* : move pw() out of plot.power.casectl.R into a separate file.
Note that it needs to be documented before exporting.
GeneticsPed
* : cleanup
* : fixed a warning: sargolzaei.c:72: warning: 'N' is used
uninitialized in this function
* : generic strings
* : move
* : removing package GeneticsPed from depends
* : cleanup
* : subject --> individual
* : prune function
* : accesor functions
* : move
* : move
* : datasets
* : no need for t(); added utilities hwp() and gpLong2Wide
* : added utilities hwp() and gpLong2Wide
* : some formatting and note on utilities hwp() and gpLong2Wide
* : suggests genetics and note on use
* : fixes; I still have to learn how to use this function
* : now in datasets.Rd
* : we need better implementation
* : removal of codeUnit; genetics is in depends now
* : we have genetics in depends now
* : subject --> ind
* : better integration
* : note
* : use of slot accessors
* : commenting some funcs for the moment
GeneTraffic
* : Bumped for LazyLoad.
GEOquery
* : Fixed sort problems in GDS2eSet and GDS2MA
* : turned on the Windows build again
* : 2.1.9
=====
* removed R.utils dependency to fix windows build issue
* added gunzip() function (from R.utils), necessary because I
need to reset
the connection while reading
* : 2.1.10
======
* Windows parsing of GSEMatrix broken, so disabled it until it
can be fixed.
* : 2.1.11
======
Fixed vignette to deal with lack of GSEMatrix parsing on windows.
GGtools
* : for the ggexplorer.org site
* : found bug need to fix on plot_mlp
* : ok
* : site support resources
* : many changes to deal with reflectance of fitting function and
erroneous plotting of gene locations in trans analyses plot_mlp
* : full image of site
* : fix global in snps3Pto.
* : some man fixes
GlobalAncova
* : avoid strange error in examples
* : correction in globaltest function call
* : added C-implementation of permutation test
* : deleted needless comments
* : corrected .bib file
globaltest
* : - removed several tiny bugs
* : Added quotes to geneplot and sampleplot legends
* : Synchronized the definition of combine with Biobase
* : Corrected documentation of "combine"
GOstats
* : Deprecate isConditional generic
Replace use of isConditional() with conditional(). This resolves
strange warning messages and check failure due to an improper
reassignment of the conditional generic.
* : Specify required ver of Category
* : Modifications to methods for KEGG and PFAM data to include
hyperlinks to KEGG and PFAM (Sanger institute) websites in output
from htmlReport().
* : Add digits and summary.args arguments to htmlReport
This allows user to specify the number of display digits for
numerical
fields. htmlReport no longer accepts '...' and uses summary.args
instead to pass additional arguments to the summary method.
* : Moved htmlReport and summary methods for KEGG and PFAM data
from GOstats to Category. Also changed the signatures for the
htmlReport to coincide with the refactored htmlReport method.
* : Unify access to annotation maps
* : Refactor makeGOGraph, simLL for GO.db compatibility
Compatibility with GO.db requires a change in signature for these
functions because there is no longer a map to Entrez Gene IDs
within
the GO.db package. These functions now take a eg2go.map which is
expected to be an object with an environment-like interface.
* : Fixes for GO.db compatibility for shortestPath and makeGOGraph
example
shortestPath has new argument where the mapping from GO to EG
must be
specified.
* : GOstats vignette now using hgu95av2.db
* : Attempt to fix GOvis vignette for compatibility with db-based
pkgs
* : Backward compat fix for makeGOGraph, simLL for DB-based ann pkg
use
We introduce new arguments at the end of the signature, mapfun
and
chip, which allow for DB-based ann pkg compatibility without
breaking
existing code.
* : Move .get_eg_to_go_fun to zzz.R and add 'reverse' argument
The reverse argument can be used to obtain a go_to_eg map and is
used by shortestPath.
* : Backward compat fix for shortestPath and DB-compat
* : Use subListExtract in makeGOGraph
* : GOstats::GOLeaves is deprecated in favor of graph::leaves
An example of using leaves was added to GOviz.Rnw.
* : Change maintainer from Seth to Robert
Category, GOstats, graph, graphPart, hypergraph
* : fixed a typo
goTools
* : Bumped for LazyLoad.
gpls
* : Maintainer field of Biocore Team packages now points to the
BioC List
Since biocpkgs@lists.fhcrc.org seems to be a spam magnet and
isn't
getting much, if any, use we are changing the maintainer email
address
for BioC Core Team maintained packages to simply be the main
Bioconductor user email list. Most users report problems there
anyhow
and we feel that encouraging that single channel for bug reports
is
best.
graph
* : Improve error message when trying to remove nodes not present
in a graph
* : Add a graphData slot to class graph
graphData is intended to hold graph-level attributes. This patch
also
adds definitions for graphData generics, but no methods.
* : I improved the handling of the incoming matrix in the matrix ->
graphAM
coercion method.
There are a number of warning messages from R CMD check regarding
new
classes and methods that fall in someone else's reach.
* : Refactor multiGraph, do not extend graph
Since graph has an edgeData slot, multiGraph should not extend
it.
This patch adds an edgeData slot to edgeSet and a nodeData slot
to
multiGraph.
Also updated and added notes to the GraphClass document
* : More notes on multiGraph
* : Improve degree method and add unit tests
* : Add more unit tests for inEdges
These tests exercise common calling patterns where Nodes arg is
missing.
* : Add a leaves method to compute the leaves of a graph
A leaf of an undirected graph is a node with degree 1. For
directed
graphs, leaves are computed with respect to in or out edges;
leaves
are the nodes with no in edges or no out edges.
This patch adds a leaves generic and a method for graph objects.
* : Fix graphAM inEdges method defs: args were in wrong order
This fixes a unit test failure seen in inEdges for graphAM
objects.
The function def signature had the arguments in the wrong order.
* : remove commented code
* : Add alternative GXL even parser: graph_handler
This XML event parser handles GXL files using environments for
intermediate storage and only creating a graph instance after
parsing
is complete. This reduces copying and dramatically improves
runtime
performance.
* : Use getExportedValue instead of require where possible for XML
package
For functions that provide optional features based on the XML
package,
we now use getExportedValue instead of require("XML") to avoid
attaching the XML package to the search path and polluting the
user's
visible namespace. An exception is the toGXL function which
returns
an XML object. This function still calls require.
* : Change maintainer from Seth to Robert
Category, GOstats, graph, graphPart, hypergraph
* : serialized MAPK signaling pathway graph
* : trying to put some documentation in place
* : added man pages for the multigraph classes
graphPart
* : Change maintainer from Seth to Robert
Category, GOstats, graph, graphPart, hypergraph
GSEABase
* : incidence(<GeneSet>, ...) and (<GeneSetCollection>)
- also minor bug fixes
* : incidence reuse + minor speed refactor
* : Allow for missing attributes in Broad xml files
* : Preliminary KEGGCollection
* : GeneSetCollection constructor revision
- use idType, setType as well as 'object' in dispatch
- idType, setType _after_ ..., so unamed args are not
automatically
matched (e.g.., because we are constructing a gene set collection
from several different gene sets, and don't want/need to specify
idType, setType
- Multiple dispatch and allowing for 'missing' arguments probably
makes callNextMethod dispatch too complicated to work out
reliably.
* : Bug fixes on GeneSetCollection
* : GeneSetCollection from old or new annotation files
- User-transparent AnnotationEnvIdentifier,
AnnotationDbiIdentifier
* : More GeneSetCollection constructors
* idType=AnnotationIdentifier(), setType=GOCollection()
* object=ExprssionSet, setType=KEGGCollection() | GOCollection(),
to
construct sets just with AnnotationIdentifier's found in the
ExprssionSet
- Use evidenceCode in GOCollection to restrict to particular
evidence
Reverted to a single AnnotationIdentifier class
* : Minimally improved show method for GeneSetCollection
* : Bug in show
* : GeneSetCollection logic
- &, |, setdiff for character, GeneSet
Accessors for GeneSetCollection
- geneIds, geneIds<-
* : mapIdentifiers for GeneSetCollection, plus bug fix for
geneIds<-
* : Documentation updates
* : getBroadSets bug fix, and retrieve character vector of Broad
gene
card names
* : Add keggIds to KEGGCollectoin, [[,GeneSetCollection,GeneSet
Also:
- setters for GOCollection members
- slightly better warnings when maps are not 1:1
* : Add IdFactory, AnnotationIdFactory
- Not exported
- Construct identifier classes, perhaps derived from
AnnotationIdentifier
Bug fix to verbose m