Core 1.4.0 1.5.0 delta

These are detailed notes on changes made between bioperl-release-1-4-0 and bioperl-release-1-5-0.

Bio::Align::DNAStatistics

Change API to return Bio::Matrix::MatrixI rather than arrayrefs - this takes a step closer to having pure-perl replacement for dnadist (eventually protdist)

Fixed TajimaNei, added Uncorrected

Code cleanup - use substr rather than arrays for speed. Do some indenting cleanup

Update some docs

Some more doc updates

Chris Dwan's fixes

Support for calculating Kimura variance per Wu and Li 1995 and Kimura 1980

Bio::Align::PairwiseStatistics

Quiet undef warnings

Documentation update

Made SYNOPSIS section to compile, helped by "cd maintenance; ./modules.pl --synopsis"

Simpler routine for calculating number of gapped columns -- fixes Issue #1676

Bio::Align::ProteinStatistics

Perl implementation of protein distance

Bail when distance is going to be too far to calculate

Bio::Align::Utilities

98% speedup for aa_to_dna_aln, operate on strings instead of sequences, and just walk down each sequence rather than going through the alignment structure. Old implementation is left intact and renamed OLD_aa_to_dna_aln

Faster reverse translate

Add bootstrap_replicates method which will generate replicates for non-parametric bootstrap. Also remove old aa->codon alignment implementation, new one is good enough.

Make sure Id gets set for new randomized sequence in bootstrap

Some more doc updates

Tiny bug- only gap character - was considered, it is possible to expect . as well

Transform gaps

If we've gone beyond the end we're in same state as a gap

Deal with missing trailing gaps

Bio::AlignIO

A flag for allowing a forced set_displayname_flat flag to be set on initialization

Remove repetitive, correct examples, clarify

Made SYNOPSIS section to compile, helped by "cd maintenance; ./modules.pl --synopsis"

Bio::AlignIO::bl2seq

Some documentation update to record the -report_type information

Various HEAD changes committed to the branch

Bail if no HSP - how we got here I have no idea

Bio::AlignIO::clustalw

Modified to accomodate concatenated alignments

Various HEAD changes committed to the branch

Match SQUID clustalw 1.5 output

Better differentiation from SQUID's Clustal 1.5 and real Clustalw

Use the new flag which will allow one to specify displayname should be flat (and used in bp_sreformat with the --special option

Change regular expression to address Issue #1637

Grab the right number of bases off of the matchline as well. Allow clustalw version to be customized (global variable for now, can be tweaked later on if we want this)

Bio::AlignIO::emboss

Push score into the SimpleAlign object as well

Bio::AlignIO::fasta

Started adding sequence alignments that store sequences in a file. coding by Albert Vilella

Various HEAD changes committed to the branch

Force displayname flat

Bio::AlignIO::largemultifasta

Started adding sequence alignments that store sequences in a file. coding by Albert Vilella

Remove unnecessary length() calls, more docs

Minor doc and cosmetic changes

Various HEAD changes committed to the branch

Bio::AlignIO::maf

Added strand designation to the Bio::LocatableSeq object as per Issue #1721.

Brad F's maf Issue #1772

Fixed maf.pm so that co-ordinates are one-based rather than zero-based.

Bio::AlignIO::mega

Nathan Haigh's suggestion for making next_aln call unmatch on an alignment before returning it

Bio::AlignIO::meme

So far so good, but I don't know if the meme output file used in test has all the conditions we need

Capture strand column when it's there

Clean up - there were some warnings in 'make test'

Bio::AlignIO::msf

Wes Barris's suggested fix to make msf have stackpack like location information

Various HEAD changes committed to the branch

Format

Bio::AlignIO::nexus

Add documentation for _initialize/new, add a feature to disable writing of 'symbols' part of nexus header in order to support Mr.Bayes nexus flavor

Parse treebase nexus files a little better

Various HEAD changes committed to the branch

Better nexus parsing

Avoid dealing with undefs if we hit the end of the file first

Nathan's patch

More from Nathan

Nathan's additions

Nathan's fix

Dos2unix

Bio::AlignIO::phylip

Started adding sequence alignments that store sequences in a file. coding by Albert Vilella

Various HEAD changes committed to the branch

Proper parse of non-interleaved files now

Adding tag_length and flag_SI, and taking care of line_length as parameters to pretty write_aln to phylip

Not zero for tag_length - nonsense

Bio::AlignIO::po

Matthew Bettis's po alignment parser/writer

Matthew's patch

prints 'VERSION=bioperl' if the alignment source field is blank. (poa chokes on blank 'VERSION=' information.)

Bio::AlignIO::selex

Support '*' in the alignment as produced by matches to model in HMMER. Also retool loops to be simpler and capture the description (DE) from Stockholm

Bio::AlignIO::stockholm

Support '*' in the alignment as produced by matches to model in HMMER. Also retool loops to be simpler and capture the description (DE) from Stockholm

Bio::AnalysisI

Fix Revision string problems

Bio::AnnotatableI

. Bio::SeqFeatureI inherits Bio::AnnotatableI NOT Bio::AnnotationCollectionI . *_tag_* methods are in Bio::AnnotatableI, and internally defer to Bio::AnnotatableI->annotation->some_analagous_mapped_function() . deprecation warnings commented until 1.6 . Bio::AnnotatableI now keeps a tag->annotation_type registry to allow new tags to be created (see Bio::SeqFeature::AnnotationAdaptor). . Bio::SeqFeature::AnnotationAdaptor is now not very useful, as *_tag_* methods map directly onto Bio::AnnotationI's Bio::AnnotationCollectionI instance. . Unflattener and Unflattener2 tests pass with no changes.

<rant> No no no no no. What the hell! This was supposed to be code working towards a stable branch! Don't add a dependancy that makes people have to install Graph::Directed to parse sequence files or use SearchIO. Bio::Seqfeature::Generic is a CORE part of the toolkit -- all changes should at least preserve its functionality and all the things it depends on. </rant> Points to the guts-l folks and anyone who actually is reading log commit messages in the future.

FTHelper uses SeqFeatureI calls rather than SeqFeature::Generic-specific calls.

Bio::Annotation::Comment

Fixed two minor bugs as reported by Peter v. Heusden.

Bio::Annotation::DBLink

Included version and possibly optional ID into the as_text() method.

Bio::Annotation::Reference

Allow initialization of all the components in new and rely on inheritance for the tagname field

Added recognition and capability of dealing with RG line in swissprot format.

Bio::Annotation::SimpleValue

Fixed two minor bugs as reported by Peter v. Heusden.

. Bio::SeqFeatureI inherits Bio::AnnotatableI NOT Bio::AnnotationCollectionI . *_tag_* methods are in Bio::AnnotatableI, and internally defer to Bio::AnnotatableI->annotation->some_analagous_mapped_function() . deprecation warnings commented until 1.6 . Bio::AnnotatableI now keeps a tag->annotation_type registry to allow new tags to be created (see Bio::SeqFeature::AnnotationAdaptor). . Bio::SeqFeature::AnnotationAdaptor is now not very useful, as *_tag_* methods map directly onto Bio::AnnotationI's Bio::AnnotationCollectionI instance. . Unflattener and Unflattener2 tests pass with no changes.

Bio::Annotation::Target

Much better support for Targets in Bio::FeatureIO::gff via a new Bio::Annotation::Target object.

Allen was right--Bio::Annotation::Target should inherit from Bio::Range

Changes to work with scott's test gff. target processing seems borken.

Fixed minor bugs in Target.pm and Target related bugs in gff.pm

Made SYNOPSIS section to compile, helped by "cd maintenance; ./modules.pl --synopsis"

Bio::AnnotationCollectionI

Doc fix

Updated Bio::AnnotationCollection to implement *_tag_* methods with deprecation warning. these were taken from Bio::SeqFeatureI and Bio::SeqFeature::Generic. *_tag_* methods in Bio::SeqFeature::Annotated are now implemented by explicit pasthru to the conttained Bio::Annotation::Collection instance.

. Bio::SeqFeatureI inherits Bio::AnnotatableI NOT Bio::AnnotationCollectionI . *_tag_* methods are in Bio::AnnotatableI, and internally defer to Bio::AnnotatableI->annotation->some_analagous_mapped_function() . deprecation warnings commented until 1.6 . Bio::AnnotatableI now keeps a tag->annotation_type registry to allow new tags to be created (see Bio::SeqFeature::AnnotationAdaptor). . Bio::SeqFeature::AnnotationAdaptor is now not very useful, as *_tag_* methods map directly onto Bio::AnnotationI's Bio::AnnotationCollectionI instance. . Unflattener and Unflattener2 tests pass with no changes.

Bio::AnnotationI

Added an ontology document registry. ontology store is now usable outside the Bio::OntologyIO subsystem as a class to create ontology structures on demand -- just ask the store for what you want by name, e.g. my $store = Bio::Ontology::OntologyStore->get_instance(); my $so = $store->get_ontology('Sequence Ontology'); file fetching happens behind the scenes with the help of recent modifications to Bio::Root::IO to support URLs, and Bio::Ontology::DocumentRegistry (just commited).

Bio::Assembly::Contig

Continuing to make Singlets work with the Assembly framework.

Bio::Assembly::IO::ace

Added a routine to read singlets in the next_seq method. Added (well, uncommented really) a routine for reading from the BS line to get chromat fine information from read information.

Creating the Singlet object. I'm not sure what has changed in sequencetrace.t but I'll look into that right away.

Singlet now works as expected. See t/singlet.t for examples of that I am trying to accomplish.

Continuing to make Singlets work with the Assembly framework.

Bio::Assembly::Scaffold

Added a routine to read singlets in the next_seq method. Added (well, uncommented really) a routine for reading from the BS line to get chromat fine information from read information.

Creating the Singlet object. I'm not sure what has changed in sequencetrace.t but I'll look into that right away.

Bio::Assembly::Singlet

This is a module to model a Singlet. I proposed creating this module on the mailing list in April 2004 and there were no objections so here it is.

Creating the Singlet object. I'm not sure what has changed in sequencetrace.t but I'll look into that right away.

Singlet now works as expected. See t/singlet.t for examples of that I am trying to accomplish.

Continuing to make Singlets work with the Assembly framework.

Bio::Biblio

Access pubmed via eutils, see: http://eutils.ncbi.nlm.nih.gov/entrez/query/static/eutils_help.html it would be nice to make Bio::DB::Biblio return Bio::Biblio::Ref object with Bio::DB::MeSH attached to them, but the Bio::Biblio::IO::pubmedxml parser doesn't currently handle enough of the pubmed xml elements to do this.

Edits

Fix Revision string problems

Bio::Biblio::IO::medline2ref

Fix Revision string problems

Bio::Biblio::IO::medlinexml

Fix Revision string problems

Bio::Biblio::IO::pubmed2ref

Fix Revision string problems

Bio::Biblio::IO::pubmedxml

Fix Revision string problems

Bio::Cluster::UniGene

Updated species.

Adding species update to branch.

Added parsing of HOMOL tag, added new species, tidied up tests

Merely whitespace formatting.

Bio::Cluster::UniGeneI

Added parsing of HOMOL tag, added new species, tidied up tests

Bio::ClusterIO::unigene

Added parsing of HOMOL tag, added new species, tidied up tests

Fixed ClusterIO::unigene not to set version unless it could parse one out.

Bio::CodonUsage::Table

Probable_codons() added - returns codons with frequency above a spcified threshold

Bio::Coordinate::Collection

Some documentation fixes Cood -> Coord

Docu and warning update from main trunk

Bio::Coordinate::GeneMapper

Pass verbose flag through to creation of locations as well - helps when debugging

Also match on stop codon, not just aa/nt - this removes error where coordinate::pairs were not mapping correctly from a pairwise alignment

Push verbosity status down to child objects

Bio::Coordinate::Graph

Old docu update

Bio::Coordinate::Pair

Also match on stop codon, not just aa/nt - this removes error where coordinate::pairs were not mapping correctly from a pairwise alignment

Prettyness update

Bio::Coordinate::Utils

Cleanup for other types of gap chars

Don't match '.' - that is special code here - only match '-'

Because we detect gaps via '-' translate gaps to '-' before getting gap line - some MSAs have this

Also match on stop codon, not just aa/nt - this removes error where coordinate::pairs were not mapping correctly from a pairwise alignment

Test with warning message

New code to map from sequence to position in alignment rather than position in the other sequence for a pair

New method to generate coordinate mapper from sequence to the alignment space- merged from main trunk

Fix method name in docu

Bio::DB::Biblio::biofetch

Fix Revision string problems

Bio::DB::Biblio::eutils

Access pubmed via eutils, see: http://eutils.ncbi.nlm.nih.gov/entrez/query/static/eutils_help.html it would be nice to make Bio::DB::Biblio return Bio::Biblio::Ref object with Bio::DB::MeSH attached to them, but the Bio::Biblio::IO::pubmedxml parser doesn't currently handle enough of the pubmed xml elements to do this.

Rather use offset of zero for cursor

Fixed a bug in get_next(). Made sure you get an exception if you try this again.

Added ability to search other dbs, such as pmc, in addition to pubmed, which was the only option before.

More docs

Bio::DB::Biblio::soap

Fix Revision string problems

Bio::DB::BiblioI

Fix Revision string problems

Bio::DB::BioFetch

Little code cleanup after the testing

Bio::DB::EMBL

Nathan's additions

Bio::DB::Failover

Fixed Bio::DB::Failover to properly pass get_seq_by_version() method, and fixed Bio::DB::Flat::BDB to properly implement it

Bio::DB::Fasta

Workaround for Win32 glob() failures due to long file names that contain whitespace

Lincoln's Win32 glob changes merged onto branch

Close and reopen index file to possibly avoid corruption issues on Windows platforms

Quenched uninitialized variable warning

Speculative changes to the search_notes() feature which will increase sensitivity but cause hits to non-full words as well

Removing cvs conflict messages (where there didn't appear to be a conflict anyway)

Killed uninit variable warning when reindexing a database for the first time

Bio::DB::FileCache

Issue #1628 - don't try and store things we didn't find

Issue #1628 - don't try and store things we didn't find

Bio::DB::Flat

Issue #1642 patch

Rearrange format

Remove warnings

Odd spacing

Bio::DB::Flat::BDB

Fixed Bio::DB::Failover to properly pass get_seq_by_version() method, and fixed Bio::DB::Flat::BDB to properly implement it

Bio::DB::Flat::BDB::embl

Fixed Bio::DB::Failover to properly pass get_seq_by_version() method, and fixed Bio::DB::Flat::BDB to properly implement it

Bio::DB::Flat::BDB::genbank

Fixed Bio::DB::Failover to properly pass get_seq_by_version() method, and fixed Bio::DB::Flat::BDB to properly implement it

Bio::DB::Flat::BDB::swiss

Fixed Bio::DB::Failover to properly pass get_seq_by_version() method, and fixed Bio::DB::Flat::BDB to properly implement it

Bio::DB::Flat::BDB::swissprot

Fixed transcript glyph so that nonstranded features are not automatically treated as + strand

Bio::DB::Flat::BinarySearch

Documentation errors

Fixed Bio::DB::Flat::BinarySearch to respect get_seq_by_version()

Pass format

Eliminate warnings by providing a regexp

Bio::DB::GFF

Patch problems that occur on win32 platforms with long names that contain spaces

Lincoln's Win32 glob changes merged onto branch

Fixed awful error which turns 0 attributes into undefined ones

Applied patch for precedence in GFF2 group assignment

Added a workable mechanism to change the grouping behavior in the ninth column of GFF2

Fixed bad merge; could someone check that the grep() solution is any faster than the original loop?

Fixed documentation errors in synopsis section

Fixed nasty bug in Bio::DB::GFF->new() routine which will cause all subsequent new() attempts to fail if the first one fails. This is a lesson not to rely on $@ being undef after a success!

Performance improvements in way that group field is handled during GFF2 parsing

Modified GFF3 compatability so that order of ID preference is "Target", "Parent" and "ID"

Quenched a bug that prevented Bio::DB::GFF from loading attributes whose values are zero.

Changed behavior of features() method so that aggregation occurs even if the type is not specified

Made Bio::DB::GFF more RandomAccessI compliant (apparently this hadn't been tested!)

Can now specify the "require_whole_object" behavior of aggregators in constructor

Moved parse_types() method from GFF.pm to DasI.pm it's a utility method, and unassuming enough to be in the interface class.

Fixed bad bug in GFF3 group handling

Bio::DB::GFF::Adaptor::dbi

Keep from croaking with invalid SQL when unknown feature types are requested

Squashed bug which prevented features added via add_feature() from appearing during the render() call

Speculative changes to the search_notes() feature which will increase sensitivity but cause hits to non-full words as well

Folded in Aaron Mackey patch for wildcard matching on source and method

Bio::DB::GFF::Adaptor::dbi::mysql

Modified GFF3 compatability so that order of ID preference is "Target", "Parent" and "ID"

Speculative changes to the search_notes() feature which will increase sensitivity but cause hits to non-full words as well

Added warnings regarding the maxbin value

Bio::DB::GFF::Adaptor::dbi::pg

Speculative changes to the search_notes() feature which will increase sensitivity but cause hits to non-full words as well

Added a few indexes to the postgres GFF adaptor that were missing

Bio::DB::GFF::Adaptor::memory

Added simple bumping and horizontal whitespace control per user request

Close and reopen index file to possibly avoid corruption issues on Windows platforms

Quelched a non-numeric argument warning

Added get_feature_by_name() and search_notes() methods

Squashed bug which prevented features added via add_feature() from appearing during the render() call

Made the description returned by search_notes() a little more specific

Fixed a bug in memory adapter that caused segments found by wildcard searching to be (sometimes) flipped in wrong orientation

Bio::DB::GFF::Aggregator

Quelched a non-numeric argument warning

Fixed error which caused method() to return undef on custom-created aggregators

Old docu update

Fixed a problem in which features with same groups but different sources got aggregated together even when user requests sources explicitly

Can now specify the "require_whole_object" behavior of aggregators in constructor

Bio::DB::GFF::Aggregator::alignment

Updated for arrival of new methods in elegans release WS121

Bio::DB::GFF::Aggregator::clone

Updated for arrival of new methods in elegans release WS121

Keys on the left or right side now adjust padding automatically

Bio::DB::GFF::Aggregator::coding

Suppress uninit variable warnings in the make_link() function

Bio::DB::GFF::Aggregator::processed_transcript

Quelched a non-numeric argument warning

Bio::DB::GFF::Feature

Remove leftover debugging

Use these variable we computed score into - handles undef problems for us

Fixed missing strand when start==stop

Fixed problem with gff3 dumping in which ID and Parent attributes were missing from features that had Target attributes

Fixed longstanding apparent bug in the merged_segments() call

Removed extraneous debugging code

Setting refseq() of top-level feature should change refseq of subfeatures as well, no?

Folded in Aaron Mackey patch to shrink parent features if subfeatures dont fill it

Bio::DB::GFF::RelSegment

Fixed biographics side key so as not to be cut off for short tracks

Quashed uninit variable warnings

Added utility routines to help gbrowse

Bio::DB::InMemoryCache

Issue #1628 - don't try and store things we didn't find

Issue #1628 - don't try and store things we didn't find

Add Id line

Bio::DB::NCBIHelper

Worked on the genbank query fetcher and got it to return more than 500 records, but maybe it is because NCBI decided to cooperate and not because of anything I did :-(

Added gbwithparts key to %FORMATMAP to deal with some genbank nucleotide records that by default don't contain all CDS features, such as L42023. To force the Bio::DB object to get all the features, you can do the following: my $gb = new Bio::DB::GenBank; $gb->request_format('gbwithparts');

Bio::DB::Query::GenBank

Unset retmax - Brad Chapman's patch.

Fixed date format that is now expected by NCBI. Also, the query produces a different result now. Fixed the expected lengths as a temporary hack.

Actually make efetch work for >500 seqs?

Fixed the featurefile class to allow for link generation when building imagemaps

Worked on the genbank query fetcher and got it to return more than 500 records, but maybe it is because NCBI decided to cooperate and not because of anything I did :-(

Bio::DB::Query::WebQuery

Fixed date format that is now expected by NCBI. Also, the query produces a different result now. Fixed the expected lengths as a temporary hack.

Bio::DB::RefSeq

Nathan's additions

Bio::DB::Registry

Need to eval{} getpwuid() call since some Perl's don't implement it.

Don't rely on HOME

Little code cleanup after the testing

Bio::DB::SwissProt

Nathan's additions

Bio::DB::Taxonomy

Add API for supporting retrieval of child nodes

Bio::DB::Taxonomy::entrez

Fix so that case doesn't matter and apply Andreas Kahari's fix Issue #1583

Taxonomy::Node can now masquerade as a Bio::Species

Move dependancy check for XML::Twig to where it is actually needed

Bio::DB::Taxonomy::flatfile

Add API for supporting retrieval of child nodes

Make sure we remove the handle before untie-ing

Get species name correct when building classification

Deal with uninit and unknown taxa requests with some grace

Bio::DB::WebDBSeqI

Worked on the genbank query fetcher and got it to return more than 500 records, but maybe it is because NCBI decided to cooperate and not because of anything I did :-(

The retry code seems to be working now

Chomp the right variable

Fixed bug in _stream_request() that was causing Bio::DB::BioFetch to fail when attempting to read FASTA format.

Nathan's patch for Windows

Nathan's additions

Format

Bio::Das::FeatureTypeI

Implementing a Bio::Das::FeatureTypeI class, this required touching several files during debug. I think the only changes made are formatting and switchable debug warningses.

Bio::Das::SegmentI

The contained_in() and contains() methods were handling list arguments incorrectly

Bio::DasI

Moved parse_types() method from GFF.pm to DasI.pm it's a utility method, and unassuming enough to be in the interface class.

Bio::Expression::FeatureSet::FeatureSetMas50

Dos2unix

Bio::Factory::AnalysisI

Fix Revision string problems

Bio::Factory::FTLocationFactory

Avoid warnings when location specified with start > end

Deal with non-numeric start/end separately

Deal with ? as start or end as well, okay this is getting a little ridiculous, just to handle reversed start/end...

Yuck. lookaheads and balanced parens. But this is a problem that has been around for a while, glad to finally fix it. Issue #1674 describes the behavior. Couldn't previously handle nested join(join()) properly

Bio::Factory::LocationFactoryI

Some documentation for the interface

Update email documentation

Bio::FeatureHolderI

See bp maillist for full desc the script seq/unflatten_seq will now generate GFF3 - the unflattener module is used to build the 'feature graph' connecting genes, transcripts, exons and CDSs together. This means we can have GFF3 for anything in genbank! As far as I'm aware, the only other sensible output formats to use here (ie formats that support feature graphs/containment hierarchies) are: chado, chaos, and the write-only asciitree. This feature graph is written out in the GFF3 using the ID and Parent tags. To do this there is an extra intermediate step - the bioperl FeatureHolderI hierarchy is traversed and ID/Parent tags are generated. Here is a description of the changes I have made: [unless you're a bioperl hacker you don't really need to read the rest of this] You can get the context of what I'm on about from this thread: http://bioperl.org/pipermail/bioperl-l/2003-December/014150.html Two new public methods: FeatureHolderI->set_ParentIDs_from_hierarchy sets both ID and ParentID from FeatureHolder hierarchy SeqFeatureI->generate_unique_persistent_id this is required by the above method Lincoln wanted this to be private, but I think it has to be called from outside FeatureHolderI->create_hierarchy_from_ParentIDs the inverse of set_ParentIDs_from_hierarchy

Moved ID methods to new class IDHandler see thread 'new GFF3 support methods' http://bioperl.org/pipermail/bioperl-l/2004-March/thread.html

On rare occasions, subfeatures might appear more than once in the FeatureHolder hierarchy. E.g. an exon might be a subfeature of two transcripts, which both are subfeatures of the same gene. Fixed 'get_all_SeqFeatures' method to take this into account.

Bringing SeqFeature::Annotated up to be Bio::SeqFeatureI compliant. added add_SeqFeature() and remove_SeqFeature() methods to Bio::FeatureHolderI

'EXPAND' option in 'add_SeqFeature' now documented.

Bio::FeatureIO

Initial commit of FeatureIO subsystem

Doc fix

Adding BED module. write-only, and not yet tested. http://www.genome.ucsc.edu/goldenPath/help/customTrack.html#BED

Cvs commit

Interpro sets 'source' of feature to value of match 'evidence' attribute, which is the program name that made the match. also adds friendly name of interpro accession as a 'comment' annotation. gff modified to support writing of 'Note' attributes.

Doc update, also _guess_format() tweak to remove Bio::Tools dependency

Bio::FeatureIO::bed

Adding BED module. write-only, and not yet tested. http://www.genome.ucsc.edu/goldenPath/help/customTrack.html#BED

Added header functionality to BED writer.

Bio::FeatureIO::gff

Initial commit of FeatureIO subsystem

Basic support for gff i/o

Gff parser wasn't initializing IO properly

Added support for more canonical gff attributes. these are represented using Bio::Annotation::SimpleValue objects

Added handling of more gff3 attributes. added test gff3 file to t/data. will add unit tests soon.

Adding BED module. write-only, and not yet tested. http://www.genome.ucsc.edu/goldenPath/help/customTrack.html#BED

Now writes GFF v2.5 (GTF)

Added a source() method to Bio::SeqFeature::Annotated. updated warning messages to state that tag* methods are deprecated

Two things: * adding SOFA as an available ontology to DocumentRegistry.pm * modifying FeatureIO::gff to use SOFA to validate, and to parse Ontology_term

Modifications to allow url fetch of gene ontology. this was not easy b/c of the multiple .ontology files for each aspect. Bio::SeqFeature::Annotated objects now instantiate Ontology_term tags as Bio::Annotation::OntologyTerm objects, not Bio::Annotation::SimpleValue objects (Scott!)

Patch from Steffen Grossmann to parse fasta (with no result)

Trying to add Target handling to FeatureIO::gff--it doesn't work yet, but is no more broken than before. Also added a phase method to Bio::SeqFeatureAnnotated and fixed confusion over frame vs phase.

Fixed some bugs in the usage of the 'source' entry. Implementation now is based on the 'source' method of Bio::SeqFeature::Annotated.

Added the ability to handle non-reserved word tags; still on the list of things to do is handling of Target tags and sequence-region and fasta directives.

Added stuff to support fasta and target processing. The quesion remains what to do with this data once you have it--particularly the fasta data. Should there be (or is there) a next_sequence() method?

Cleanup to use annotation shortcut methods

Much better support for Targets in Bio::FeatureIO::gff via a new Bio::Annotation::Target object.

Allen was right--Bio::Annotation::Target should inherit from Bio::Range

Changes to work with scott's test gff. target processing seems borken.

Added a 'next_feature_group' method as a starter for retrieving more than flat arrays of features from GFF3 files. Moved some of the functionality from 'next_feature' into the new method '_next_feature_or_directive'. Existing functionality is not changed by this.

'gff-version' directive is now placed on top of a file which is opened for output. There was a strange behaviour when I tried to use 'mode' to detect whether the file is writeable. I got some extra output which must have been related to the fact that 'mode' tries to access the file. Therefore I did it differently (certainly not optimal). Any suggestions/comments?

'source' in Bio::SeqFeature::Annotated gives back a Bio::Annotation::SimpleValue object. Make sure to print its value.

Bio::SeqFeature::Annotated: Unified the implementation of attribute accessor methods. 'seqid', 'type', 'source', 'score', 'frame' and 'phase' now all use the AnnotationCollection directly and give back Bio::AnnotationI implementing objects. Scalar values as well as appropriate Bio::AnnotationI implementing objects can be used on set. Default values ('.') are returned when called without previous setting. 'seqid' and 'source' uri_escape their values on set with scalars. 'seqid' itself is new and should replace 'id' for better compatibility with gff2 specs, but I left 'id' untouched. Bio::FeatureIO::gff: Adapted to the changes in Bio::SeqFeature::Annotated. Especially switched back to the use of 'seqid' instead of 'id'.

I'm trying to 'fix' the constructor of Annotated.pm and the write_feature method of gff.pm so that I can create an arbitrary Annotated sequence that I can use to generate a line of gff. I'm sure I did several things along the way.

Interpro makes features returned of type 'region'. gff3 produced now correctly uri-escapes attribute column, and additionally displays Ontology_term and Dbxref attributes.

Interpro sets 'source' of feature to value of match 'evidence' attribute, which is the program name that made the match. also adds friendly name of interpro accession as a 'comment' annotation. gff modified to support writing of 'Note' attributes.

Consolidated seq_id/seqid/id to seq_id.

Interpro SO accession did not match name. gff writer now supports 'Target' annotations.

- patched in validation code from Rob Edwards. - added code to parse ##sequence-region (based on Rob Edwards patch). sequences given as ##sequence-region directives are buffered onto the next_feature() stream. - added a next_seq() method for reading the FASTA section (if any) at the bottom of the file. this will only ever return objects after all features have been read. introduces dependency on Bio::SeqIO - refactored next_feature_group() to use recursion rather than clunky _next_feature_or_directive() method (removed).

Bio::FeatureIO::gtf

Now writes GFF v2.5 (GTF)

Bio::FeatureIO::interpro

Moved over from Bio::SeqIO::interpro

Interpro makes features returned of type 'region'. gff3 produced now correctly uri-escapes attribute column, and additionally displays Ontology_term and Dbxref attributes.

Interpro sets 'source' of feature to value of match 'evidence' attribute, which is the program name that made the match. also adds friendly name of interpro accession as a 'comment' annotation. gff modified to support writing of 'Note' attributes.

Consolidated seq_id/seqid/id to seq_id.

Interpro SO accession did not match name. gff writer now supports 'Target' annotations.

Bio::Graph::Edge

Initial commit

Made SYNOPSIS section to compile, helped by "cd maintenance; ./modules.pl --synopsis"

Bio::Graph::IO

Initial commit

Debug

Threshold option

Threshold parameter specifiable

Bio::Graph::IO::dip

Intial commits

Write_network() method added

Debug for test

Debugw

Threshold option

Debug

Bio::Graph::IO::psi_xml

Intial commits

Debug

Bio::Graph::ProteinGraph

Initial commit of protein network module

Debug for test

3 new methods added: remove_nodes(), clustering_coefficient(), remove_dup_edges()

New method - unconnected_nodes()

New method - articulation_points()

Algoritm improvement to articulation_points

Revised articulation_points() now faster

Bugfix

New methods edge_count(), node_count(), debugged union()

New method, neighbour_count()

Improved articulation_point() now independent of implementation of SimpleGraphTraversal

Made SYNOPSIS section to compile, helped by "cd maintenance; ./modules.pl --synopsis"

Get rid of 'redeclare' warning

Bio::Graph::SimpleGraph

Commit of Nat Goodman's new SimpleGraph module

Debug for test

Small fixes to the names of the modules (Bio::Graph::SimpleGraph vs SimpleGraph)

Tidy up

Made SYNOPSIS section to compile, helped by "cd maintenance; ./modules.pl --synopsis"

Bio::Graph::SimpleGraph::Traversal

Initial commit of Nat Goodman's graph traversal module

Debug for test

Algoritm improvement to articulation_points

Tidied up

Test that get_all deals with non-hash based nodes

Dependencies on node being a hashref removed

Bio::Graphics::Feature

Backed out Jason's very sensible paranoia because it breaks ability to deliberately make an undef setting

Fixed the cds and translation glyphs so that they honor the -flip argument correctly

Added get_feature_by_name() and search_notes() methods

Keys on the left or right side now adjust padding automatically

Added utility routines to help gbrowse

Bio::Graphics::FeatureFile

Paranoia to avoid dereferencing undefinded values

Backed out Jason's very sensible paranoia because it breaks ability to deliberately make an undef setting

Added code for generating clickable imagemaps in CGI scripts

Quashed uninit variable warnings

Hopefully fixed all the linkrule problems

Fixed more confusion in the make_link() method (how I wish I\'d never started this!

Quelched a non-numeric argument warning

Removed trailing whitespace from each line before parsing. I hope nobody wants/needs those.

Suppress uninit variable warnings in the make_link() function

Added get_feature_by_name() and search_notes() methods

Auto-label attribute should now work correctly, at cost of some memory bloat

Speculative changes to improve performance of uploaded 3d party features in genome browser

Performance improvements in way that group field is handled during GFF2 parsing

Fixed the MAX_REMAP variable so as to be a more rational value

Squashed bug which prevented features added via add_feature() from appearing during the render() call

Fixed the featurefile class to allow for link generation when building imagemaps

Fixed handling of reverse strands in the super-short version of the featureFile renderer

Fixed a bug in memory adapter that caused segments found by wildcard searching to be (sometimes) flipped in wrong orientation

Change to allow a prototyped coderef to be passed in from a configuration (ie, to allow something like 'sort_order = sub ($$) {...}'

Bio::Graphics::Glyph

Fixed a problem that occurs when an unrecognized font is passed to the glyph

Conditional use of ellipse() and filledEllipse() for backwards support of gd 1.8.4. NB: this might fail if user has mismatched gd/GD versions as it checks for can() of above methods.

Merged conditional use of gd2-specific methods, enabling backward compatibility with gd 1.8.4

Paranoia to avoid dereferencing undefinded values

Added simple bumping and horizontal whitespace control per user request

Added code for generating clickable imagemaps in CGI scripts

Quashed uninit variable warnings

Quelched a non-numeric argument warning

Keys on the left or right side now adjust padding automatically

Resolved (at least some) clipping bugs for features that extend over the end of the panel

Added recognition of 'maxdepth' option. if declared for a track and > 0, the glyph only attempts to render subfeatures down to level 'maxdepth'.

Improved alignment of bases when segments are showing a multiple alignment; still problems involving ragged ends and dynamic realignment

Fixed highmag segments glyph crash (uncommitted changes from yesterday)

Implementing a Bio::Das::FeatureTypeI class, this required touching several files during debug. I think the only changes made are formatting and switchable debug warningses.

Bio::Graphics::Glyph::Factory

Resolved (at least some) clipping bugs for features that extend over the end of the panel

Added the EMBL/Genbank entry renderer to the scripts directory, since it is probably of general interest

Implementing a Bio::Das::FeatureTypeI class, this required touching several files during debug. I think the only changes made are formatting and switchable debug warningses.

Bio::Graphics::Glyph::anchored_arrow

Keys on the left or right side now adjust padding automatically

Issue #1641 - documentation fix

Issue #1641 - documentation fix

Bio::Graphics::Glyph::arrow

Patched in Aaron's flipping fixes

Issue #1641 - documentation fix

Issue #1641 - documentation fix

Bio::Graphics::Glyph::broken_line

Added new glyphs

Bio::Graphics::Glyph::cds

Fixed the cds and translation glyphs so that they honor the -flip argument correctly

Fixed color problems in reverse-strand-flipped 6-frame translation glyph

Bio::Graphics::Glyph::christmas_arrow

Added new glyphs

Bio::Graphics::Glyph::dashed_line

Glyphs requested by KEGG, courtesy Simon

Added new glyphs

Bio::Graphics::Glyph::diamond

Fixed track clipping problems

Bio::Graphics::Glyph::dna

Fixed the cds and translation glyphs so that they honor the -flip argument correctly

Fixed color problems in reverse-strand-flipped 6-frame translation glyph

Bio::Graphics::Glyph::dot

Conditional use of ellipse() and filledEllipse() for backwards support of gd 1.8.4. NB: this might fail if user has mismatched gd/GD versions as it checks for can() of above methods.

Merged conditional use of gd2-specific methods, enabling backward compatibility with gd 1.8.4

Bio::Graphics::Glyph::dumbbell

Glyphs requested by KEGG, courtesy Simon

Fixes to the dumbbell glyph from Simon

Added new glyphs

Bio::Graphics::Glyph::flag

Glyphs requested by KEGG, courtesy Simon

Bio::Graphics::Glyph::generic

Fixed problem of right-side of glyph labels colliding with right-side key

Fixed a crash that occurs when drawing right & left keys and no default background specified for panel; added magic code to display tags of type "note" or "description" as default descriptions

Bio::Graphics::Glyph::lightning

Added a fairly ridiculous new lightning glyph. April showers bring May flowers...

Bio::Graphics::Glyph::pentagram

Added new glyphs

Bio::Graphics::Glyph::ragged_ends

Initial import

Bio::Graphics::Glyph::repeating_shape

Glyphs requested by KEGG, courtesy Simon

Bio::Graphics::Glyph::saw_teeth

Glyphs requested by KEGG, courtesy Simon

Bio::Graphics::Glyph::segments

Improved alignment of bases when segments are showing a multiple alignment; still problems involving ragged ends and dynamic realignment

Rolled back changes

Rewrote the way multiple alignments are displayed from scratch

Improved display of inserted bases

Fixed crash in segments glyph when drawing multiple alignments at high mag

Still problems in boundary conditions when no segment is in current display window

Fixed problem of phony DNA appearing when zoomed into a gap at high mag

Fixed highmag segments glyph crash (uncommitted changes from yesterday)

Bio::Graphics::Glyph::text_in_box

Added new glyphs

Bio::Graphics::Glyph::three_letters

Glyphs requested by KEGG, courtesy Simon

Bio::Graphics::Glyph::tic_tac_toe

Added new glyphs

Bio::Graphics::Glyph::track

Fixed problem of right-side of glyph labels colliding with right-side key

Fixed track clipping problems

Bio::Graphics::Glyph::transcript

Performance improvements in way that group field is handled during GFF2 parsing

Fixed transcript glyph so that nonstranded features are not automatically treated as + strand

Bio::Graphics::Glyph::transcript2

Performance improvements in way that group field is handled during GFF2 parsing

Resolved (at least some) clipping bugs for features that extend over the end of the panel

Bio::Graphics::Glyph::translation

Workaround for the dreaded difference between a Bio::SeqFeature and a Bio::Seq

Fixed the cds and translation glyphs so that they honor the -flip argument correctly

Added the EMBL/Genbank entry renderer to the scripts directory, since it is probably of general interest

Fixed problem of translation glyph not aligning to flipped coordinates; issue of colorization of flipped six-frame translation still pending

Fixed colors of translation glyph so that when the image is flipped, the colors of the frames flip appropriately

Another attempted fix of the padding problem

Fixed color problems in reverse-strand-flipped 6-frame translation glyph

Removing a redundant line to eliminate a warning

Bio::Graphics::Glyph::two_bolts

Added new glyphs

Bio::Graphics::Glyph::wave

Added new glyphs

Bio::Graphics::Glyph::weighted_arrow

Added new glyphs

Bio::Graphics::Glyph::xyplot

Added defaults to xyplot graph type

Some people are using -graphtype option rather than -graph_type, and if you can't beat 'em join 'em

Formatting fixes for xyplot, including drawing labels which were previously neglected

Removed a debugging statement

Resolved (at least some) clipping bugs for features that extend over the end of the panel

Fixed bad bug in GFF3 group handling

Bio::Graphics::Panel

Paranoia to avoid dereferencing undefinded values

Added simple bumping and horizontal whitespace control per user request

Fixed biographics side key so as not to be cut off for short tracks

Added code for generating clickable imagemaps in CGI scripts

Hopefully fixed all the linkrule problems

Fixed more confusion in the make_link() method (how I wish I\'d never started this!

Keys on the left or right side now adjust padding automatically

Improved documentation of boxes() result

Squashed bug which prevented features added via add_feature() from appearing during the render() call

Resolved (at least some) clipping bugs for features that extend over the end of the panel

Fixed problem of right-side of glyph labels colliding with right-side key

Fixed track clipping problems

Fixed a crash that occurs when drawing right & left keys and no default background specified for panel; added magic code to display tags of type "note" or "description" as default descriptions

Fixed mis-registration of grid with tickmarks when viewing flipped regions; problem of translation glyph not aligning is still pending

Fixed scrambled "bottom" key when viewing flipped regions; problem of translation glyph not aligning is still pending

Small fix brought out by recent changes to GBrowse; user may now specify image class simply using SVG instead of fully qualified GD::SVG package

Bio::Graphics::Pictogram

Need to include SeqIO

Bio::Index::Abstract

Add warning if overwriting an existing value for sequence id -- to turn off the warning set verbose < 0

Format

Bio::Index::AbstractSeq

Nathan's additions

Bio::Index::Blast

Note on id_parser

Bio::Index::Fasta

Format

Note on id_parser

Bio::Index::GenBank

Format

Bio::Index::Hmmer

Adding hmmer index

Probably want to put this after we keep track of headers - per Marc Logghe individually reported bug

Made SYNOPSIS section to compile, helped by "cd maintenance; ./modules.pl --synopsis"

Note on id_parser

Bio::Index::Qual

Added Mark's Bio::Index::Qual module

Note on id_parser

Bio::Index::Swissprot

Add id_parser method

Suppress a warning

Note on id_parser

Bio::LiveSeq::IO::Loader

Jouni's patch

Bio::LocatableSeq

Escape the special characters out of paranoia

Also match on stop codon, not just aa/nt - this removes error where coordinate::pairs were not mapping correctly from a pairwise alignment

More fixes to handle stop codons in alignments and remapping positions

Bio::Location::Atomic

Documentation fix, Issue #1731 fix

Bio::Location::AvWithinCoordPolicy

Documentation fix, Issue #1731 fix

Bio::Location::Fuzzy

Documentation fix, Issue #1731 fix

More cleanup for Issue #1731

Bio::Location::FuzzyLocationI

Documentation fix, Issue #1731 fix

Bio::Location::NarrowestCoordPolicy

Documentation fix, Issue #1731 fix

Bio::Location::Simple

Documentation fix, Issue #1731 fix

Bio::Location::Split

Documentation fix, Issue #1731 fix

Bio::Location::SplitLocationI

Documentation fix, Issue #1731 fix

Bio::Location::WidestCoordPolicy

Documentation fix, Issue #1731 fix

Bio::LocationI

Documentation fix, Issue #1731 fix

Bio::Map::Clone

Jamie Hatfield et al's Bio::Map stuff for FPC Marker data

Bio::Map::Contig

Jamie Hatfield et al's Bio::Map stuff for FPC Marker data

Bio::Map::FPCMarker

Jamie Hatfield et al's Bio::Map stuff for FPC Marker data

Bio::Map::Physical

Jamie Hatfield et al's Bio::Map stuff for FPC Marker data

Bio::MapIO::fpc

Jamie Hatfield et al's Bio::Map stuff for FPC Marker data

Bio::Matrix::PSM::IO

Added psiblast to list of allowable file formats in IO.pm.

Fixed a problem in IO.pm: moved throw to $class rather than $self, as $self isn't defined yet.

Fix problems with ISA inheritance

Dos2unix

Adding a matrix fasta- 'masta' file format Only DNA support for now

Bio::Matrix::PSM::IO::mast

Spelling

Bio::Matrix::PSM::IO::masta

Adding a matrix fasta- 'masta' file format Only DNA support for now

Bio::Matrix::PSM::IO::meme

Now parses the log-odds

Bug fixed- supports -revcomp option now and gives info on the strand of the instance parser will throw an exception now if it can't find the right number of columns

Get rev strand properly

Critical bug- when instance sequence at the end of the input sequence

Critical bug- when instance sequence at the end of the input sequence

+ to 1/- to 0 for the strand

Gap char . added to the alphabet

Bio::Matrix::PSM::IO::psiblast

Initial commit of ProtMatrix.pm, a module for handling position-specific matrices with protein alphabets by implementing SiteMatrixI.pm (basically a protein analogue to SiteMatrix.pm). Not yet fully functional, but very close. Initial commit of psiblast.pm, a parser for ASCII-formatted PSI-BLAST matrices. Returns a Bio::Matrix::PSM::ProtMatrix object. See "Gapped BLAST and PSI-BLAST: a new generation of protein database search programs." by Altschul et. al (1997) for more information.

Changed documentation in psiblast.pm to reflect use of ProtPsm rather than Psm module. Added alphabet subroutine to ProtMatrix.pm.

Cleaned up whitespace in psiblast formatter, removed unnecessary break condition in file parsing loop.

Added PsmHeader and Bio::Root::Root to @ISA.

Fix problems with ISA inheritance

Bio::Matrix::PSM::IO::transfac

Parses any reference data, stores it into the Psm object that is returned

Bio::Matrix::PSM::InstanceSiteI

Dos2unix

Bio::Matrix::PSM::ProtMatrix

Initial commit of ProtMatrix.pm, a module for handling position-specific matrices with protein alphabets by implementing SiteMatrixI.pm (basically a protein analogue to SiteMatrix.pm). Not yet fully functional, but very close. Initial commit of psiblast.pm, a parser for ASCII-formatted PSI-BLAST matrices. Returns a Bio::Matrix::PSM::ProtMatrix object. See "Gapped BLAST and PSI-BLAST: a new generation of protein database search programs." by Altschul et. al (1997) for more information.

Added the IUPAC method in ProtMatrix.pm to comply with SiteMatrix.pm interface. Added my name to the AUTHORS file. :)

- Made regexp methods in ProtMatrix.pm accept a threshold. - Fixed sequence_match_weight method in ProtMatrix.pm. - Added testing for regexp and sequence_match_weight methods in ProtMatrix.t.

Changed documentation in psiblast.pm to reflect use of ProtPsm rather than Psm module. Added alphabet subroutine to ProtMatrix.pm.

Fixed documentation in ProtMatrix. Added another column to psiblast header in PsmHeader.pm.

Cleaned up docs in ProtMatrix.

Removed unimplemented method get_compressed_freq and get_compressed_logs. Fixed PWM->PSM in Perldoc.

Bio::Matrix::PSM::ProtPsm

Initial commit of ProtPsm.pm, a module for holding ProtMatrix and SiteMatrix information on protein-specific PSM's.

Need to 'use' a module before declaring inheritance with ISA

Bio::Matrix::PSM::Psm

Now isa Bio::Annotation object

Psm has to initialize as Annotation::Collection, done

When calling matrix method log-odds were lost

Unnecessary Bio::Root::Root in ISA

Bio::Matrix::PSM::PsmHeader

Added psiblast information to PsmHeader.pm.

Fixed documentation in ProtMatrix. Added another column to psiblast header in PsmHeader.pm.

Bio::Matrix::PSM::PsmHeaderI

Dos2unix

Bio::Matrix::PSM::SiteMatrix

Regexp was incorrectly created

Compression of frequencies and log-odds: one position, one base is a single char

Added checks and documentation

Simple consensus can be calculated now using threshold value

New method- get the score for a sequence match, calculate from the log-odds

Method added: get_all_vectors, all possible seq to satisfy the PFM under a give threshold

Aaaron's suggestions

Aaaron's suggestions

Uninitialized variable-fixed

Next_pos will return weights as well now

Method added to convert to PSM(PWM) from PFM

Sequence_match_weight will produce more info to locate a possible problem

Bio::Matrix::PSM::SiteMatrixI

Added checks and documentation

Simple consensus can be calculated now using threshold value

New method- get the score for a sequence match, calculate from the log-odds

Method added: get_all_vectors, all possible seq to satisfy the PFM under a give threshold

Aaaron's suggestions

Doc update

Method added to convert to PSM(PWM) from PFM, documentation added for calc_weight

Bio::Matrix::PhylipDist

Shorten code, debug warnings when missing a value

Bio::Ontology::DocumentRegistry

Added an ontology document registry. ontology store is now usable outside the Bio::OntologyIO subsystem as a class to create ontology structures on demand -- just ask the store for what you want by name, e.g. my $store = Bio::Ontology::OntologyStore->get_instance(); my $so = $store->get_ontology('Sequence Ontology'); file fetching happens behind the scenes with the help of recent modifications to Bio::Root::IO to support URLs, and Bio::Ontology::DocumentRegistry (just commited).

Fixed duplicate definition of get_instance().

Two things: * adding SOFA as an available ontology to DocumentRegistry.pm * modifying FeatureIO::gff to use SOFA to validate, and to parse Ontology_term

Modifications to allow url fetch of gene ontology. this was not easy b/c of the multiple .ontology files for each aspect. Bio::SeqFeature::Annotated objects now instantiate Ontology_term tags as Bio::Annotation::OntologyTerm objects, not Bio::Annotation::SimpleValue objects (Scott!)

Experimental changes for caching ontology files

Bio::Ontology::Ontology

Removed useless line of code, possibly a left-over from somebody testing.

Implementing a Bio::Das::FeatureTypeI class, this required touching several files during debug. I think the only changes made are formatting and switchable debug warningses.

Experimental changes for caching ontology files

Bio::Ontology::OntologyStore

Added an ontology document registry. ontology store is now usable outside the Bio::OntologyIO subsystem as a class to create ontology structures on demand -- just ask the store for what you want by name, e.g. my $store = Bio::Ontology::OntologyStore->get_instance(); my $so = $store->get_ontology('Sequence Ontology'); file fetching happens behind the scenes with the help of recent modifications to Bio::Root::IO to support URLs, and Bio::Ontology::DocumentRegistry (just commited).

Cache downloaded ontology

Doc fix

Modifications to allow url fetch of gene ontology. this was not easy b/c of the multiple .ontology files for each aspect. Bio::SeqFeature::Annotated objects now instantiate Ontology_term tags as Bio::Annotation::OntologyTerm objects, not Bio::Annotation::SimpleValue objects (Scott!)

Experimental changes for caching ontology files

Oops, keep as url on head

Experimental ontology caching

A little protection

Bio::Ontology::RelationshipType

Added support for '~' (related_to) relationships. ran into this one in the current Cell Ontology. '~' relationships are described in the format spec: http://www.geneontology.org/GO.format.html note: there is still a lack of support for '<' (opposite of is_a) and '!=' (opposite of '=' or is_a in the identical sense). will add as needed.

Bio::Ontology::SimpleGOEngine

Added support for '~' (related_to) relationships. ran into this one in the current Cell Ontology. '~' relationships are described in the format spec: http://www.geneontology.org/GO.format.html note: there is still a lack of support for '<' (opposite of is_a) and '!=' (opposite of '=' or is_a in the identical sense). will add as needed.

Temporary workaround -- this won't necessarily work forever....

Bio::Ontology::SimpleOntologyEngine

Issue #2004 richard at uwc dot net Files: Bio/Ontology/SimpleOntologyEngine.pm Bio/Ontology/Term.pm Bio/OntologyIO/InterProParser.pm Bio/OntologyIO/Handlers/InterProHandler.pm t/InterProParser.t

Fixed bug in erroneously expecting a scalar instead of an array.

Bio::Ontology::Term

Dblinks come with context

Issue #2004 richard at uwc dot net Files: Bio/Ontology/SimpleOntologyEngine.pm Bio/Ontology/Term.pm Bio/OntologyIO/InterProParser.pm Bio/OntologyIO/Handlers/InterProHandler.pm t/InterProParser.t

You can't just cut the definition to an arbitrary length - only because your schema or whatever imposes that limit. Reverting this change. BTW it leads to lots of uninitialized value warnings being thrown.

Polished up the InterPro XML SAX handler and added persistence handler properties in order for it to be useable under load_ontology.pl (which wants to install its own persistence handler since they are highly configurable). Aliased the handler as interprosax in OntologyIO.

Experimental changes for caching ontology files

Bio::OntologyIO

Polished up the InterPro XML SAX handler and added persistence handler properties in order for it to be useable under load_ontology.pl (which wants to install its own persistence handler since they are highly configurable). Aliased the handler as interprosax in OntologyIO.

Bio::OntologyIO::Handlers::BaseSAXHandler

Polished up the InterPro XML SAX handler and added persistence handler properties in order for it to be useable under load_ontology.pl (which wants to install its own persistence handler since they are highly configurable). Aliased the handler as interprosax in OntologyIO.

Bio::OntologyIO::Handlers::InterProHandler

Issue #2004 richard at uwc dot net Files: Bio/Ontology/SimpleOntologyEngine.pm Bio/Ontology/Term.pm Bio/OntologyIO/InterProParser.pm Bio/OntologyIO/Handlers/InterProHandler.pm t/InterProParser.t

Bio::OntologyIO::Handlers::InterPro_BioSQL_Handler

Polished up the InterPro XML SAX handler and added persistence handler properties in order for it to be useable under load_ontology.pl (which wants to install its own persistence handler since they are highly configurable). Aliased the handler as interprosax in OntologyIO.

Bio::OntologyIO::InterProParser

Issue #2004 richard at uwc dot net Files: Bio/Ontology/SimpleOntologyEngine.pm Bio/Ontology/Term.pm Bio/OntologyIO/InterProParser.pm Bio/OntologyIO/Handlers/InterProHandler.pm t/InterProParser.t

Bio::OntologyIO::dagflat

Fixed the dagflat ontology parser. Yes you do suffer if your regexps become too promiscuous. Also added tests so that this doesn't happen anymore.

Migrated dagflat parser fix to the stable branch.

Added support for '~' (related_to) relationships. ran into this one in the current Cell Ontology. '~' relationships are described in the format spec: http://www.geneontology.org/GO.format.html note: there is still a lack of support for '<' (opposite of is_a) and '!=' (opposite of '=' or is_a in the identical sense). will add as needed.

Bug in defintions file parser caused a def node and a term node to be created (2x the number of objects needed). further, the defs were not associated with the terms. :\

Modifications to allow url fetch of gene ontology. this was not easy b/c of the multiple .ontology files for each aspect. Bio::SeqFeature::Annotated objects now instantiate Ontology_term tags as Bio::Annotation::OntologyTerm objects, not Bio::Annotation::SimpleValue objects (Scott!)

Bio::OntologyIO::simplehierarchy

Little improvements, mostly to accommodate the latest eVoc formats.

Removed cruft that was lying around commented out. Improved the meaningfulness of an error message.

Fixed omitted related_to relationship from setting the ontology etc.

Bio::Perl

A little more detail

Bio::Phenotype::OMIM::OMIMentry

Perl 5.8 doesn't like "return my @array = ()" and calls it bizarre. Well, it's certainly not the most efficient way to program perl ...

Merged fixes from main trunk to make perl 5.8.4 happy.

Add get_clinical_symptom_organs and query_clinical_symptoms

Bio::Phenotype::OMIM::OMIMparser

Make parser a little more robust to badly formatted data

Pitch in _finer_parse_symptoms method

Bio::Phenotype::Phenotype

Perl 5.8 doesn't like "return my @array = ()" and calls it bizarre. Well, it's certainly not the most efficient way to program perl ...

Merged fixes from main trunk to make perl 5.8.4 happy.

Bio::PopGen::Genotype

Fix some documentation

Change regexp

Bio::PopGen::HtSNP

Adding Tag and HtSNP modules

Made SYNOPSIS section to compile, helped by "cd maintenance; ./modules.pl --synopsis"

Bio::PopGen::IO::csv

Initial contribution by Rich, cleaned up by Jason

Bio::PopGen::IO::hapmap

Initial contribution by Rich, cleaned up by Jason

Update Rich's email addr

Bio::PopGen::IO::phase

Initial contribution by Rich, cleaned up by Jason

Update Rich's email addr

Bio::PopGen::IO::prettybase

Warn if did not parse line correctly

Bio::PopGen::Individual

Documentation update

Fix some documentation

Force individual id to be same for genotype when it is add to an ind

Bio::PopGen::PopStats

Documentation update

Bio::PopGen::Population

Convert a diploid (or any-ploid really) population to a haploid individuals

Documentation update

Fix some documentation

Bio::PopGen::PopulationI

Fix doc for interface

Bio::PopGen::Statistics

Fix Fu and Li's F, warn that F* is still broken, allow for methods to directly calculate the statistics from the raw counts rather than inferring from a population. These are named XX_counts methods

Some docu fix and force haploid population calculation

Fu and Li's F* now working properly. Thanks to Charla Lambert for pointing out the corrections in Simonsen et al (1995). Also checked values against other available libs. K.thornton and D.Ardell

Warn if the full complement of markers is not seen for every individual (probably need to adjust and remove these markers from consideration eventually)

Hanging use Data::Dumper deleted

Docu update

Promised example in the SYNOPSIS

Added citation to MBE paper using Bio::PopGen modules

Bio::PopGen::TagHaplotype

Adding Tag and HtSNP modules

Bio::PopGen::Utilities

Documentation update

Convert alignments into population

Missing needed use statements

Bio::PrimarySeq

Fixing docs to indicate that these objects have get/set on the alphabet call

Allow unsetting of the primary ID.

Nathan says correct regexp

Fix Issue #1670 on the branch

Add X to class of ambiguous characters, _guess_alphabet

Bio::Range

Adding new constructor, unions() that takes a list of Bio::Range as input and returns a list of non-overlapping Bio::Range as output. this method is strand agnostic

Commented warning

Get rid of warnings

Bio::Restriction::Analysis

Method needs to be cut(), not cuts(). Issue #1596

Method needs to be cut(), not cuts(). Issue #1596

Yes, Type I and Type III enzymes cut outside their recognition sequences, there should be no exception

Fixed according to Andrew Nunberg's comments.

Bio::Restriction::IO::withrefm

Remove some warnings

Bio::Root::AccessorMaker

A bioperl customized accessor maker. working now but still under construction for more features

This module is removed from bioperl and will be used an external module

Bio::Root::IO

Dave Howorth's fix for dealing with mac/unix/dos linefeeds when reading from files produced on different platforms

Added code which, if a file to read from doesn't exist and the name looks like a web url, attempts to download the file and use a local tempfile of it instead. this allows: Some::IO->new(-file=>'http://server.my/test.gff') to pull down a gff file, cache it locally, and read from it. it may be better to have a special -url option or something to keep this code separated out, but this commit is just to get it out there.

Network fetch functionality in _initialize_io() activated with -url argument (as opposed to overloading -file). removed LWP::Simple requirement, if it is not present failover to use Bio::Root::HTTPget.

Three minor fixes: - switch off $VERBOSE by default - require LWP::Simple instead of use - return the return value of print from _print method

More intelligent _pushback and _readline

Bio::Root::Root

Rich's fix for untainting

Bio::Root::RootI

Unify not_implemented messsage

Correct a fib, caused by my miscounting for caller method

Bio::Root::Storable

Will Spooner's fixes

Bio::Root::Version

Increased the version number to be differenet from the release

Bio::Search::BlastStatistics

I forgot to commit this yesterday.

Bio::Search::GenericStatistics

Created wrrappers for the statics and parameters in a generic result. I geared this specifically for Blast in the name of Java interoperability. The reason the interoperability was spoiled because of the use of a has to represent statistics and parameters. This efficiently and elegantly fixes the problem.

Bio::Search::HSP::GenericHSP

* new(): Modified the frac_identical/frac_conserved setting calls to use the logical length of the aligned region where appropriate via SearchUtils::logical_length(). Also, the percent_identity() set call relies on frac_identical() (calculate something once, why calculate it again? ;-).

Migrating revision 1.60.2.1 to the main trunk. Fixes Issue #1597

Store Group field

Argument order and data initialized was flip-flopped

Bio::Search::HSP::HSPI

Don't try and calculate match overlap when there is no seq string stored (like PSL)

Bio::Search::Hit::GenericHit

Added method _warn_about_no_hsps() which is now called by all methods requiring HSP data (most of which also call tile_hsps()).

* frac_identical(), frac_conserved(): Added comments describing the when the "logical" length of the aligned portion of query/hit sequences are used. * length_aln(): Moved the setter code before the attempt to tile hsps. * Also migrating mods from revision 1.30 to branch-1-4.

Refactored logical_length() to delegate to Bio::Search::SearchUtils::logical_length()

Migrating revisions 1.29.2.1 and 1.29.2.2 to the main trunk.

Fixed check for undef omissions. Also, "$acc.$version" doesn't look very good if $version is undef or an empty string (fixed now).

* Moved the _warn_about_no_hsps() method from GenericHit.pm into SearchUtils.pm. Refactored SearchUtils::tile_hsps() to use it. * Added more descriptive text to the warning message in _warn_about_no_hsps() explaining the most likely cause of it.

Migrating revision 1.29.2.3 to the main trunk.

Issue #1714 - there was inconsistent expectation of the state of _hsps arrayref

Bio::Search::Iteration::IterationI

Issue #1611 - add_hit to a BlastResult object

Bio::Search::Result::BlastResult

Created wrrappers for the statics and parameters in a generic result. I geared this specifically for Blast in the name of Java interoperability. The reason the interoperability was spoiled because of the use of a has to represent statistics and parameters. This efficiently and elegantly fixes the problem.

I forgot to commit this yesterday.

Issue #1611 - add_hit to a BlastResult object

Bio::Search::Result::GenericResult

Created wrrappers for the statics and parameters in a generic result. I geared this specifically for Blast in the name of Java interoperability. The reason the interoperability was spoiled because of the use of a has to represent statistics and parameters. This efficiently and elegantly fixes the problem.

Bio::Search::Result::ResultI

Added sort_hits() method -- allows sorting of hits by user-supplied coderef. Added _default_sort_hits method -- default codref for sort_hits(), sorts by descending score

Bio::Search::SearchUtils

Added warn call to tile_hsps() in the event of no HSPs in the supplied hit object. tile_hsps() also returns a boolean value: true if it was able to tile, false otherwise. Not currently used.

* Migrating 1.11 modification to branch-1.4. * Added logical_length() to compute the length of the aligned sequence based on the algorithm type (e.g., blastx) and sequence type (e.g., query vs hit). This refactors coded used by GenericHit and GenericHSP. * Fixes Issue #1597

Migrating revision 1.10.2.1 to the main trunk.

* Moved the _warn_about_no_hsps() method from GenericHit.pm into SearchUtils.pm. Refactored SearchUtils::tile_hsps() to use it. * Added more descriptive text to the warning message in _warn_about_no_hsps() explaining the most likely cause of it.

Migrating revision 1.10.2.2 to the main trunk.

No FASTA algorithm is as abominable as TBLASTX; similarly, there's no TFASTN (and no one should ever use TFASTA, so we don't support it)

Bio::Search::StatisticsI

Created wrrappers for the statics and parameters in a generic result. I geared this specifically for Blast in the name of Java interoperability. The reason the interoperability was spoiled because of the use of a has to represent statistics and parameters. This efficiently and elegantly fixes the problem.

Bio::SearchIO::IteratedSearchResultEventBuilder

Issue #1653 get rank back for hits

Issue #1653 get rank back for hits

Bio::SearchIO::SearchResultEventBuilder

Handle cases when query_length or hit_length should be derived from the parsed data rather than counted from the alignment bases

Issue #1653 get rank back for hits

Issue #1653 get rank back for hits

Bio::SearchIO::Writer::GbrowseGFF

Squashed a bug in the GFF3 formatter. The GbrowseGFF formatter is now updated to be GFF3 compliant. It also uses GFF3 by default now.

Resolved a GFF3/2.5 formatting issue

Fixed some syntax errors(??)

Bio::SearchIO::Writer::HTMLResultWriter

Deal with HMMER output okay

Igor Dolgalev reported unbalanced </tr> in statistics table; also make HMMPFAM writing quieter by reusing bit/score value when one is missing

Bio::SearchIO::Writer::TextResultWriter

Deal with HMMER output okay

Marc Logghe's commit

Bio::SearchIO::blast

* Fixed Issue #1126 (504 letters) * Commented out some debug calls

Fixed syntax error created by previous commit.

Fixed check for undef omissions. Also, "$acc.$version" doesn't look very good if $version is undef or an empty string (fixed now).

Handle NCBI weirdness/bug on OS X blast 2.2.8 where numbers aren't really numbers...

Deal with part of Issue #1598, parsing on OSX

Merge necessary main trunk fixes to the branch in anticipation of a 1.4.1 release one day

Treat empty and undef strings as empty

Handle extra val in the Expect Evalue # (only present in megablast output I think) -from bl2seq -m T output at any rate)

Parse -noseqs wublast output

Parse Group field

Handle extra space introduced by HTML blast or HTML::Strip

Move the test for whitespace only string later in the characters subroutine. Also fix hsp_group parsing to properly capture the data (missing the ',' in the parse). Initialization parameter is -hsp_group now not -group

Silence uninit warnings

Issue #1703 fixes

Bio::SearchIO::blasttable

Apply Steve's small caching speedup to blastable

Bio::SearchIO::fasta

More fixes for fasta parsing <sigh>

Match changes in new FASTA ungapped has become similar match both; slight code indenting

Set positive count properly too

Bio::SearchIO::hmmer

Better regexp to distinguish between RF lines and those with the characters RF in the alignment

Change HMMER parser to report a single HSP per Hit so domains with multiple alignments get separate Hits (more FASTA like) since they aren't really HSPs

Bio::SearchIO::sim4

Major fix: 'HSPs' that were preceded by another one for which the definition finished at the very end of an alignment block did not get initialized correctly.

Add comment and escape comment in regexp

Old changes some cleaning up

Bio::Seq

Fixing docs to indicate that these objects have get/set on the alphabet call

Correction

Allow unsetting of the primary ID.

Made primary_id() to delegate. Yes, always. Also, added a Seq test to make sure glaring inconsistencies don't happen again.

Bio::Seq::LargeLocatableSeq

Started adding sequence alignments that store sequences in a file. coding by Albert Vilella

Minor doc and cosmetic changes

Bio::Seq::LargePrimarySeq

Started adding sequence alignments that store sequences in a file. coding by Albert Vilella

Bio::Seq::LargeSeq

Started adding sequence alignments that store sequences in a file. coding by Albert Vilella

Bio::Seq::LargeSeqI

Started adding sequence alignments that store sequences in a file. coding by Albert Vilella

Remove unnecessary length() calls, more docs

Bio::Seq::PrimaryQual

Fixed a problem detected by Luke McCarthy in which leading spaces in quality lines threw off the parser.

Added the ability to specify a header when calling write_seq on a qual

Bio::Seq::QualI

Phillip's patch

Bio::Seq::RichSeqI

Fixed transcript glyph so that nonstranded features are not automatically treated as + strand

Bio::Seq::SeqFastaSpeedFactory

Allow alphabet to be passed to create() so we dont have to guess it if its known

Bio::Seq::SeqWithQuality

Continuing to make Singlets work with the Assembly framework.

Bio::Seq::SequenceTrace

Need to pass arguments when delegating

Fixed case bug + commented out apparently unused variables

Bio::SeqFeature::Annotated

This class tries to be clean of *_tag_* calls, opting to use Bio::AnnotationCollectionI as a replacement. we should be able to map most of the *_tag_* calls to Bio::AnnotationCollectionI calls, but i haven't done it yet. currently throwing an error on *_tag_* calls.

Added a source() method to Bio::SeqFeature::Annotated. updated warning messages to state that tag* methods are deprecated

Added a shortcut get_Annotations() method to feature to bypass ->annotation-> middleman call. also added nice list->scalar autoconverstion

Two things: * adding SOFA as an available ontology to DocumentRegistry.pm * modifying FeatureIO::gff to use SOFA to validate, and to parse Ontology_term

Trying to add Target handling to FeatureIO::gff--it doesn't work yet, but is no more broken than before. Also added a phase method to Bio::SeqFeatureAnnotated and fixed confusion over frame vs phase.

Methods for targets

'source' method now is really based on the Annotation::Collection. FeatureIO::gff profits from this.

Bringing SeqFeature::Annotated up to be Bio::SeqFeatureI compliant. added add_SeqFeature() and remove_SeqFeature() methods to Bio::FeatureHolderI

Fixed syntax errors

Dropped duplicated methods.

Cleanup to use annotation shortcut methods

Updated Bio::AnnotationCollection to implement *_tag_* methods with deprecation warning. these were taken from Bio::SeqFeatureI and Bio::SeqFeature::Generic. *_tag_* methods in Bio::SeqFeature::Annotated are now implemented by explicit pasthru to the conttained Bio::Annotation::Collection instance.

Changes to work with scott's test gff. target processing seems borken.

Removed the 'my $self = shift;' lines in 'has_tag', 'add_tag_value', 'get_tag_values', 'get_all_tags' and 'remove_tag' because they are not needed here.

. Bio::SeqFeatureI inherits Bio::AnnotatableI NOT Bio::AnnotationCollectionI . *_tag_* methods are in Bio::AnnotatableI, and internally defer to Bio::AnnotatableI->annotation->some_analagous_mapped_function() . deprecation warnings commented until 1.6 . Bio::AnnotatableI now keeps a tag->annotation_type registry to allow new tags to be created (see Bio::SeqFeature::AnnotationAdaptor). . Bio::SeqFeature::AnnotationAdaptor is now not very useful, as *_tag_* methods map directly onto Bio::AnnotationI's Bio::AnnotationCollectionI instance. . Unflattener and Unflattener2 tests pass with no changes.

Added support for the 'EXPAND' option in 'add_SeqFeature'.

'source' now gives back its value instead of a Bio::Annotation::SimpleValue object. (This is consisitent with 'type').

'source' now gives back a Bio::Annotation::SimpleValue object, but we make sure that we really always get something, so that we can call $feat->source->value without bothering about the existence of a source annotation. I think this behaviour should consistently be implemented for all parts of the annotation.

Bio::SeqFeature::Annotated: Unified the implementation of attribute accessor methods. 'seqid', 'type', 'source', 'score', 'frame' and 'phase' now all use the AnnotationCollection directly and give back Bio::AnnotationI implementing objects. Scalar values as well as appropriate Bio::AnnotationI implementing objects can be used on set. Default values ('.') are returned when called without previous setting. 'seqid' and 'source' uri_escape their values on set with scalars. 'seqid' itself is new and should replace 'id' for better compatibility with gff2 specs, but I left 'id' untouched. Bio::FeatureIO::gff: Adapted to the changes in Bio::SeqFeature::Annotated. Especially switched back to the use of 'seqid' instead of 'id'.

If we are setting undef scores to '.', we have to allow that as a valid value!

I'm trying to 'fix' the constructor of Annotated.pm and the write_feature method of gff.pm so that I can create an arbitrary Annotated sequence that I can use to generate a line of gff. I'm sure I did several things along the way.

Consolidated seq_id/seqid/id to seq_id.

FTHelper uses SeqFeatureI calls rather than SeqFeature::Generic-specific calls.

Bio::SeqFeature::AnnotationAdaptor

. Bio::SeqFeatureI inherits Bio::AnnotatableI NOT Bio::AnnotationCollectionI . *_tag_* methods are in Bio::AnnotatableI, and internally defer to Bio::AnnotatableI->annotation->some_analagous_mapped_function() . deprecation warnings commented until 1.6 . Bio::AnnotatableI now keeps a tag->annotation_type registry to allow new tags to be created (see Bio::SeqFeature::AnnotationAdaptor). . Bio::SeqFeature::AnnotationAdaptor is now not very useful, as *_tag_* methods map directly onto Bio::AnnotationI's Bio::AnnotationCollectionI instance. . Unflattener and Unflattener2 tests pass with no changes.

Bio::SeqFeature::Collection

Insure that feature is-a Generic feature before getting FTstring in debug

Bio::SeqFeature::Gene::Intron

Acceptor and donor splice_site methods now accept 0 as a legitimate value for splice_site

Bio::SeqFeature::Generic

Protect in event on calls add_SeqFeature with an empty value

See bp maillist for full desc the script seq/unflatten_seq will now generate GFF3 - the unflattener module is used to build the 'feature graph' connecting genes, transcripts, exons and CDSs together. This means we can have GFF3 for anything in genbank! As far as I'm aware, the only other sensible output formats to use here (ie formats that support feature graphs/containment hierarchies) are: chado, chaos, and the write-only asciitree. This feature graph is written out in the GFF3 using the ID and Parent tags. To do this there is an extra intermediate step - the bioperl FeatureHolderI hierarchy is traversed and ID/Parent tags are generated. Here is a description of the changes I have made: [unless you're a bioperl hacker you don't really need to read the rest of this] You can get the context of what I'm on about from this thread: http://bioperl.org/pipermail/bioperl-l/2003-December/014150.html Two new public methods: FeatureHolderI->set_ParentIDs_from_hierarchy sets both ID and ParentID from FeatureHolder hierarchy SeqFeatureI->generate_unique_persistent_id this is required by the above method Lincoln wanted this to be private, but I think it has to be called from outside FeatureHolderI->create_hierarchy_from_ParentIDs the inverse of set_ParentIDs_from_hierarchy

Support specifying multiple values per tag (already supported by add_tag_value)

. Bio::SeqFeatureI inherits Bio::AnnotatableI NOT Bio::AnnotationCollectionI . *_tag_* methods are in Bio::AnnotatableI, and internally defer to Bio::AnnotatableI->annotation->some_analagous_mapped_function() . deprecation warnings commented until 1.6 . Bio::AnnotatableI now keeps a tag->annotation_type registry to allow new tags to be created (see Bio::SeqFeature::AnnotationAdaptor). . Bio::SeqFeature::AnnotationAdaptor is now not very useful, as *_tag_* methods map directly onto Bio::AnnotationI's Bio::AnnotationCollectionI instance. . Unflattener and Unflattener2 tests pass with no changes.

Bio::SeqFeature::Primer

Updated Tm calculation to reflect suggestions by Barry Moore. See bioperl-l for discussion of why.

Bio::SeqFeature::Similarity

Will's fix

Bio::SeqFeature::Tools::FeatureNamer

New module for granting names to features

Bio::SeqFeature::Tools::IDHandler

Moved ID methods to new class IDHandler see thread 'new GFF3 support methods' http://bioperl.org/pipermail/bioperl-l/2004-March/thread.html

Tidied up ID handling

Steffen Grossman's patch; test case upcoming

In ID generation, transcript_id or protein_id from genbank file are used if available

Bio::SeqFeature::Tools::TypeMapper

Added get_relationship_type_by_parent_child

Fixed docs

Modified TypeMapper.pm to translate 'Protein' to 'protein' (perhaps there ought to be a general 'lower' and then deal with exceptions like CDS). Also modified genbank2gff3.PLS to produce a full reference sequence line in addition to the ##sequence-region directive.

Bio::SeqFeature::Tools::Unflattener

Malcolm cook's patch to populate exon accessors

See bp maillist for full desc the script seq/unflatten_seq will now generate GFF3 - the unflattener module is used to build the 'feature graph' connecting genes, transcripts, exons and CDSs together. This means we can have GFF3 for anything in genbank! As far as I'm aware, the only other sensible output formats to use here (ie formats that support feature graphs/containment hierarchies) are: chado, chaos, and the write-only asciitree. This feature graph is written out in the GFF3 using the ID and Parent tags. To do this there is an extra intermediate step - the bioperl FeatureHolderI hierarchy is traversed and ID/Parent tags are generated. Here is a description of the changes I have made: [unless you're a bioperl hacker you don't really need to read the rest of this] You can get the context of what I'm on about from this thread: http://bioperl.org/pipermail/bioperl-l/2003-December/014150.html Two new public methods: FeatureHolderI->set_ParentIDs_from_hierarchy sets both ID and ParentID from FeatureHolder hierarchy SeqFeatureI->generate_unique_persistent_id this is required by the above method Lincoln wanted this to be private, but I think it has to be called from outside FeatureHolderI->create_hierarchy_from_ParentIDs the inverse of set_ParentIDs_from_hierarchy

Allowed subfeatures to EXPAND the container feature; this is necessary to deal with how genbank represents dicistronic genes; see ZFP91 and CNTF in build 34.3 this is not the ideal way of dealing with dicistronics but it should suffice for now expansion causes a level=1 problem to be thrown

Now deals with pseudogenes

Rationalised ID generation code in chaosxml write adapter changed verbose reporting in unflattener to use stderr

Turn verbosity down

Turn verbosity down

Verbose option now consistently writes to STDERR more liberal in paranoid checking of subfeature ordering; now handles unusual ensembl style genbank revcomp splitlocs

Be quiet when verbosity is <=0

Make it clean when verbosity < 0

Bio::SeqFeatureI

See bp maillist for full desc the script seq/unflatten_seq will now generate GFF3 - the unflattener module is used to build the 'feature graph' connecting genes, transcripts, exons and CDSs together. This means we can have GFF3 for anything in genbank! As far as I'm aware, the only other sensible output formats to use here (ie formats that support feature graphs/containment hierarchies) are: chado, chaos, and the write-only asciitree. This feature graph is written out in the GFF3 using the ID and Parent tags. To do this there is an extra intermediate step - the bioperl FeatureHolderI hierarchy is traversed and ID/Parent tags are generated. Here is a description of the changes I have made: [unless you're a bioperl hacker you don't really need to read the rest of this] You can get the context of what I'm on about from this thread: http://bioperl.org/pipermail/bioperl-l/2003-December/014150.html Two new public methods: FeatureHolderI->set_ParentIDs_from_hierarchy sets both ID and ParentID from FeatureHolder hierarchy SeqFeatureI->generate_unique_persistent_id this is required by the above method Lincoln wanted this to be private, but I think it has to be called from outside FeatureHolderI->create_hierarchy_from_ParentIDs the inverse of set_ParentIDs_from_hierarchy

Source is part of unique key

Moved ID methods to new class IDHandler see thread 'new GFF3 support methods' http://bioperl.org/pipermail/bioperl-l/2004-March/thread.html

Bug fixes for James Thompson's reported bug in spliced_seq, changed the API so it supports an optional 2nd argument which if true, will not sort the sub-locations in a split-location before splicing together the pieces. Also fixed an error in how revcomplemented pieces were stitched together

Updated Bio::AnnotationCollection to implement *_tag_* methods with deprecation warning. these were taken from Bio::SeqFeatureI and Bio::SeqFeature::Generic. *_tag_* methods in Bio::SeqFeature::Annotated are now implemented by explicit pasthru to the conttained Bio::Annotation::Collection instance.

Changes to Bio::AnnotationCollectionI and Bio::SeqFeatureI. * Bio::SeqFeatureI now ISA Bio::AnnotationCollectionI * All Bio::SeqFeatureI *_tag_* methods have been moved to Bio::AnnotationCollectionI, marked as deprecated, and mapped to their analogous and mostly pre-existing Bio::AnnotationCollectionI methods. Methods which were not in Bio::AnnotationCollectionI, but were i Bio::Annotation::Collection and were necessary for *_tag_* method remapping were created in Bio::AnnotationCollecitonI. * Bio::RangeI and Bio::AnnotationCollectionI method documentation removed from Bio::SeqFeatureI, and replaced with a link to the interface class inherited from. This reduces documentation maintenance overhead.

. Bio::SeqFeatureI inherits Bio::AnnotatableI NOT Bio::AnnotationCollectionI . *_tag_* methods are in Bio::AnnotatableI, and internally defer to Bio::AnnotatableI->annotation->some_analagous_mapped_function() . deprecation warnings commented until 1.6 . Bio::AnnotatableI now keeps a tag->annotation_type registry to allow new tags to be created (see Bio::SeqFeature::AnnotationAdaptor). . Bio::SeqFeature::AnnotationAdaptor is now not very useful, as *_tag_* methods map directly onto Bio::AnnotationI's Bio::AnnotationCollectionI instance. . Unflattener and Unflattener2 tests pass with no changes.

Only do the prepending when nosort is specified

Bio::SeqIO

Add mention of new formats

Add tinyseq

Fix documentation since the default doesn't work anymore without -fh => \*ARGV

Fix documentation since the default doesn't work anymore without -fh => \*ARGV

Doc fix

Check to see if -file or -fh arguments have been passed in with undefined values. If that's the case, don't fall back to $ARGV[0], rather throw an exception.

List of formats in 1 file, not 3

Bio::SeqIO::FTHelper

Seqid needs to be set for the feature object as well

Update doc

FTHelper uses SeqFeatureI calls rather than SeqFeature::Generic-specific calls.

Bio::SeqIO::agave

Initial version of AGAVE XML parser

Bio::SeqIO::bsml_sax

BSML parsing via SAX parser

Add Id line

Bio::SeqIO::chadoxml

See bp maillist for full desc the script seq/unflatten_seq will now generate GFF3 - the unflattener module is used to build the 'feature graph' connecting genes, transcripts, exons and CDSs together. This means we can have GFF3 for anything in genbank! As far as I'm aware, the only other sensible output formats to use here (ie formats that support feature graphs/containment hierarchies) are: chado, chaos, and the write-only asciitree. This feature graph is written out in the GFF3 using the ID and Parent tags. To do this there is an extra intermediate step - the bioperl FeatureHolderI hierarchy is traversed and ID/Parent tags are generated. Here is a description of the changes I have made: [unless you're a bioperl hacker you don't really need to read the rest of this] You can get the context of what I'm on about from this thread: http://bioperl.org/pipermail/bioperl-l/2003-December/014150.html Two new public methods: FeatureHolderI->set_ParentIDs_from_hierarchy sets both ID and ParentID from FeatureHolder hierarchy SeqFeatureI->generate_unique_persistent_id this is required by the above method Lincoln wanted this to be private, but I think it has to be called from outside FeatureHolderI->create_hierarchy_from_ParentIDs the inverse of set_ParentIDs_from_hierarchy

Moved ID methods to new class IDHandler see thread 'new GFF3 support methods' http://bioperl.org/pipermail/bioperl-l/2004-March/thread.html

Add following input to write_seq method: is_analysis flag: for setting the is_analysis flag in chado feature table; data_source: 'GenBank' or 'GFF' to handle GenBank and GFF data differently. change pub.miniref to pub.uniquename to conform to chado schema change.

Edits

Bio::SeqIO::chaos

Modules for producing chaos-xml

Tidied up ID handling

Rationalised ID generation code in chaosxml write adapter changed verbose reporting in unflattener to use stderr

Bio::SeqIO::chaosxml

Modules for producing chaos-xml

Bio::SeqIO::embl

Handle Feature Table-less records and set unknown id to something without spaces

Remove debugging

Parse sequences with no FT

Added extraction of NCBI taxon ID to the embl parser.

James Abbott's patch

Merged J.Abbot's patch and the recognition of the NCBI taxon ID from the main trunk.

All Bio::DB sequence retrieval modules only warn not throw on missing seq

All Bio::DB sequence retrieval modules only warn not throw on missing seq

Space required in ID line between accession and 'standard'

Indent so that table looks good

Consolidate this into a single test

Consolidate this into a single test

Issue #1662. copied code from genbank to embl to avoid truncation of long feature values; word wrapping added to warning

Previous fix on Issue #1662 highlighted a spurious warning in code. fixed

Added return values to all writing statements, so that write_seq() now returns 1 on success and undef on failure. (The documentation said that it did already, but the docs were incorrect)

Apply Simon's patch

Nathan Haigh's suggestions to call return instead of last when short circuiting, also tightened up the code a little, need to protect calls to regexps

Resove Issue #1618 for long, over 10 character, ids

Bio::SeqIO::fasta

Quash errors when display_id is undef

Protect for empty value possibility (to get rid of warnings)

Avoid undef warnings when no id for a fasta sequence (which is legal)

Allow alphabet to be set for SeqIO::fasta, thus determining the alphabet of all it's sequences - doesn't then rely on _guess_alphabet, but limited to objects that contain one one type of alphabet if this is to be used

Bio::SeqIO::game::featHandler

Made changes to fix broken protein ID handling, various cosmetic changes

Various patches applied to get round-tripping with the Generic Genome Browser working

Bio::SeqIO::game::gameHandler

Various patches applied to get round-tripping with the Generic Genome Browser working

Bio::SeqIO::game::gameSubs

Made changes to fix broken protein ID handling, various cosmetic changes

Various patches applied to get round-tripping with the Generic Genome Browser working

Bio::SeqIO::game::gameWriter

Made changes to fix broken protein ID handling, various cosmetic changes

Various patches applied to get round-tripping with the Generic Genome Browser working

Bio::SeqIO::game::seqHandler

Made changes to fix broken protein ID handling, various cosmetic changes

Various patches applied to get round-tripping with the Generic Genome Browser working

Bio::SeqIO::genbank

Fix missing 6 trailing spaces after ORIGIN when writing GenBank

Merge fix for trailing space onto branch

Documentation fixes.

Issue #1588 -- Fix parsing problem when there is PUBMED id but not MEDLINE id in the REFERENCE block

Added extraction of NCBI taxon ID to the embl parser.

Now accepts unconventional Organism names. Have given examples in head2. Will sort out embl.pm and others that need doing as well as updating tests soon.

I seem to have altered the contributors list, @ to at, unintentionally

Updated old version previously. this one is correct.

Accepts another instance of strange Organism naming.

Some valid, if unconventional, species labels had been split into subspecies. Not anymore.

Exit the loop after the taxon was found. Purely for efficiency.

Bug fix: stray unmatched parenthesis in SOURCE line, messed up a regex.

Tenative fix for Issue #1650 - comment lines truncated in genbank writing

Indent so that table looks good

Issue #1662. copied code from genbank to embl to avoid truncation of long feature values; word wrapping added to warning

Previous fix on Issue #1662 highlighted a spurious warning in code. fixed

Fix Issue #1673 -definition line parsing not quite proper before heading to REFERENCE parsing

Skip writing ORIGIN and BASE count when dealing with a CONTIG record. Issue #1314 should be fixed now.

Parse WGS records

Bio::SeqIO::interpro

Adding support for parsing IPRscan output XML files.

Added documentation to Bio::SeqIO::interpro.pm

Dos2unix

Made SYNOPSIS section to compile, helped by "cd maintenance; ./modules.pl --synopsis"

Bio::SeqIO::kegg

Keggification -- parse pathway name now

Update docs wrt pathway

Cosmetics -- Albert Vilella

Entry_id now can cope with .N (versioning) - changed some parameters to be optional, as it seems some kegg files, like ftp://ftp.genome.ad.jp/pub/kegg/genomes/genes/L.major.ent, dont have, for example, NAME or other fields -- Albert Vilella

Bio::SeqIO::metafasta

Edits

Bio::SeqIO::qual

Added the ability to specify a header when calling write_seq on a qual

Write_seq can be passed either of 2 different objects, but only 1 of these has a header() method. Let's check to see what we have before we call header().

I like this syntax better...

Rearrange HEADER, call header() before id()

Bio::SeqIO::scf

Because of a 'feature' discovered by Anthony Underwood (Hi Anthony!) I made a change to scf.pm. The original specification for scf as found here: (http://staden.sourceforge.net/manual/formats_unix_8.html) allows the program writing scf to place base information before sample information in the file. I did not detect that behavior in any of my programs so I read the samples and then the bases sequentially. I changed scf.pm to 'seek' around the file looking for information based on what is contained in the header rather then assuming a given program behaves logically. I wasn't able to find a 'seek' in and of the bioperl IO modules so I used a perl call. I hope that doesn't break any platform-specific seek issues.

Remove warnings - make them debug stmts

Make obj dump a debug stmt

Fixed version issue in writer subroutine

Fixed missing last base bug.

Bio::SeqIO::swiss

Fixed Issue #1584

Fixed Issue #1584 on branch

All Bio::DB sequence retrieval modules only warn not throw on missing seq

All Bio::DB sequence retrieval modules only warn not throw on missing seq

Now allows unconventional OS labels.

Indent so that table looks good

Major overhaul - SeqIO code needs an audit! Parsing of multi-lined RP lines and better usage of the _pushback function in Bio::Root::IO. Create all the references at once as well

Added capability to parse new gene name (GN) line format.

Added recognition and capability of dealing with RG line in swissprot format.

Fixed reference parsing if the RA lacking reference is not the last one. Fixed species parsing to not generate warning if genus gets undefined for environmental sample seqs.

Fix problem when species name contains regexp special characters

Fixed parsing of references.

Partial fix for Issue #1734

Bio::SeqIO::tab

Check for tabs in id before writing out

Check for tabs in id before writing out

Bio::SeqIO::tigr

Fixed a bug in the KEYWORD regex Added the _process_tiling_path function

Fixed my e-mail address.

Fix for newly used IS_PRIMARY

Fixed GO regex

Bio::SeqIO::tigrxml

Maybe not perfect version of tigrxml - I might move this somewhere else later

Grab the description from TIGRXML

Cleanup slightly

Get the ASMBL ID properly, although not really necessary since we grab accession later on

Get the fname properly

Bio::SeqIO::tinyseq

Initial check-in of modules for parsing NCBI TinySeq xml sequences

Added method to Bio::SeqIO::tinyseq to write taxid/organism info if available Added method to Bio::SeqIO::tinyseq::tinyseqHandler.pm to reject docs w/o tinyseq DTD

Bio::SeqIO::tinyseq::tinyseqHandler

Initial check-in of modules for parsing NCBI TinySeq xml sequences

Added method to Bio::SeqIO::tinyseq to write taxid/organism info if available Added method to Bio::SeqIO::tinyseq::tinyseqHandler.pm to reject docs w/o tinyseq DTD

Bio::SimpleAlign

Started adding sequence alignments that store sequences in a file. coding by Albert Vilella

Applying Dmitry Samborsky's patch

Dmitry's patch

Fixing indentation

Dmitri's patch

Accept score as an init option

Handle a couple of odd cases I invented, use the deprecated method

Let 0 be a valid state, so look for undefined chars

Brad F's suggestion to propigate strand down to newly created seqs in a slice. Also a third option to the slice() function allows slice to return columns in the slice which are all gaps

Bio::SimpleAnalysisI

Fix Revision string problems

Bio::Species

Wes' fix

Bio::Structure::IO::pdb

Dave's patch

Bio::Structure::SecStr::DSSP::Res

Updated DSSP parser to successfully deal with DSSP output generated from ATOM-line only input files.

Bio::Taxonomy::Node

Taxonomy::Node can now masquerade as a Bio::Species

Add API for supporting retrieval of child nodes

Bio::Tools::Alignment::Consed

Fixed a fault when a singlets file does not exist.

Bio::Tools::Analysis::Protein::GOR4

Debug for 1.5 release

Bio::Tools::Analysis::Protein::Scansite

Improved docs and argument processing

to handle server errors

Bio::Tools::BPbl2seq

Attempt to cleanup memory cycles- Issue #1581

Bio::Tools::BPlite

Localize $_, make sure /o some regexps for potential speed, and apply patches for Issue #1668 reported by Michael Cariaso parsing support for MPIblast output

Bio::Tools::BPlite::Sbjct

Applied Frederic Pecqueur's fix for subset of databases

Localize $_, make sure /o some regexps for potential speed, and apply patches for Issue #1668 reported by Michael Cariaso parsing support for MPIblast output

Bio::Tools::Blat

Neil's patch

Neil's patch

Bio::Tools::CodonTable

Test translate gaps - condense 3gaps to a single gap for translate to work instead of 'X'

Translate with triplet gaps are made into a single gap in the resultant protein

New method, reverse_translate_all() which reverse translates the whole aa sequence to a IUPAC nucleotide string

Enhanced reverse_translate_all() now takes a codon usage table as an argument

Bio::Tools::EPCR

Apply Malcolm Cook's patch to make the module more flexible. Also capitalize the 'Note' field and add supporting get/set methods for source,primary,groupclass

Last bit of Malcom Cook's changes I forgot

Bio::Tools::Fgenesh

Added Christopher Dwan's Fgenesh.pm

Parse with or without leading whitespace

Dos2unix

Bio::Tools::GFF

Make parse header not display warnings when opening a writer stream as it always calls _parse_header - not a good way to prevent this so turn of the warnings

Risking another stint in purgatory, I bravely change the GFF formatter. The changes are: 1) new format, GFF2.5, is available. The only difference is in the way that it formats Target tag/values in column 9 using the tstart and tend tags. 2) when calling gff_string on a SeqFeature::FeaturePair it will now use the 'hit' sequence as its Reference (column 1), rather than the 'query' sequence; this is the behaviour that Gbrowse expects, and it just makes more sense that way IMO. GFF3 format now forces lowercase on the first character of non-reserved words, since the spec reserves all ucfirst tags for reserved words in the future. At the moment the list of reserved words is hard-coded, so this will need to be looked at in the future. GFF3 format now produces the Target tag with proper format, and uses information from the 'query' sequence to get its information, rather than the 'hit' sequence. If this breaks anyone, I'm sure I will hear the screams.

Squashed a bug in the GFF3 formatter. The GbrowseGFF formatter is now updated to be GFF3 compliant. It also uses GFF3 by default now.

Oops. GFF3 is super-sensitive to spaces in column 9. All tests pass now.

See bp maillist for full desc the script seq/unflatten_seq will now generate GFF3 - the unflattener module is used to build the 'feature graph' connecting genes, transcripts, exons and CDSs together. This means we can have GFF3 for anything in genbank! As far as I'm aware, the only other sensible output formats to use here (ie formats that support feature graphs/containment hierarchies) are: chado, chaos, and the write-only asciitree. This feature graph is written out in the GFF3 using the ID and Parent tags. To do this there is an extra intermediate step - the bioperl FeatureHolderI hierarchy is traversed and ID/Parent tags are generated. Here is a description of the changes I have made: [unless you're a bioperl hacker you don't really need to read the rest of this] You can get the context of what I'm on about from this thread: http://bioperl.org/pipermail/bioperl-l/2003-December/014150.html Two new public methods: FeatureHolderI->set_ParentIDs_from_hierarchy sets both ID and ParentID from FeatureHolder hierarchy SeqFeatureI->generate_unique_persistent_id this is required by the above method Lincoln wanted this to be private, but I think it has to be called from outside FeatureHolderI->create_hierarchy_from_ParentIDs the inverse of set_ParentIDs_from_hierarchy

Protect for empty value possibility (to get rid of warnings)

Added ability to parse sequence data in GFF3 - see NOTES section & email to bioperl list for details

Adopted Aarons suggestions for method names

Altering _from_string for GFF3 to specify tab delimited and thus allow unescaped spaces--the GFF3 spec changed in this regard

Fix Issue #1690 and do some code formatting fix

Non-existent 'source_tag' entries should give a '.' in the gff string. Fixed this.

Disallowing double quotes in GFF3

Bio::Tools::Genewise

One step these pattern matches

Bio::Tools::IUPAC

New hash added to map nucleotide combinations -> IUPAC code

Prevent warning of GT deprecation

Requested feature

Bio::Tools::Phylo::PAML

Parse kappa properly for codeml reports

Merge changes for parsing kappa to branch

Add kappa info to the docs

Add kappa info to the docs

Parse omega when it is fixed as well as when it is estimated - undo problem introduced when applying kappa fix

Added why PAML short circuited - need to add check to warn specifically in the case of stop codons

Strip whitespace so nodenames are proper

BASEML parsing (partial - pariwise only for now)

Parse tree data too not just pairwise

Grab rate matrix

Bio::Tools::Phylo::PAML::Result

Avoid falling over when there are no nssite results

Merge fix for not failing when no NSSites were run

Update doc - it is a Bio::Matrix::PhylipDist object now

BASEML parsing (partial - pariwise only for now)

Bio::Tools::Phylo::Phylip::ProtDist

Shorten code

Bio::Tools::Primer::AssessorI

Fix a bug

Bio::Tools::Primer3

Michael's patch applied

Bio::Tools::Run::GenericParameters

Created wrrappers for the statics and parameters in a generic result. I geared this specifically for Blast in the name of Java interoperability. The reason the interoperability was spoiled because of the use of a has to represent statistics and parameters. This efficiently and elegantly fixes the problem.

Bio::Tools::Run::ParametersI

Created wrrappers for the statics and parameters in a generic result. I geared this specifically for Blast in the name of Java interoperability. The reason the interoperability was spoiled because of the use of a has to represent statistics and parameters. This efficiently and elegantly fixes the problem.

Bio::Tools::Run::RemoteBlast

Modified RemoteBlast.pm to understand the entire QBlast PUT/GET API.

Made independent of tempfiles - results held in memory

Explicit file handle close stops tempfile errors

Add RETRIEVALHEADER example

Bio::Tools::Run::StandAloneBlast

Modified _setparams so queries against multiple databases can use dbs in default directory ($BLASTDATADIR) or be specified by full path.

Issue #1599

Issue #1599

Correct some documentation

New versions of bl2seq requires a program to provided, will not default to blastp anymore

Old doc commits

Documentation fix

Quiet and q parameter were getting stored in the same slot... Another reason I HATE AUTOLOAD

Added -R option for PSI-TBLASTN per Issue #1137

Back out duplicate

Handle files which have been segmented

Update allowable blastpgp params

Bio::Tools::Run::WrapperBase

Argument get/set for passing commandline parameters to exec command

Docu update

Bio::Tools::SeqStats

0.01 dalton adjustments to amino acid molweights to match those in SWISS-PROT.

0.01 dalton adjustments to amino acid molweights to match those in SWISS-PROT.

Add 1 significant digit to molecular weight result

Bio::Tools::SeqWords

Fix some documentation

Bio::Tools::SiRNA

Modified to allow oligos in 3prime UTR as an option Fixed calls to $target->start / $target->end. Modified to use any Bio::SeqI compliant object as a target.

Moved Don's fixes from head

Extensive revision to support multiple rulesets as Bio::Tools::SiRNA::ruleset subclasses

Bio::Tools::SiRNA::Ruleset::saigo

Initial check-in of bioperl object for designing siRNA reagents using the ruleset published by Ui-Tei et al.

Bio::Tools::SiRNA::Ruleset::tuschl

Initial check in of bioperl object for designing siRNA reagents using rules developed by the Tuschl group (note - these used to be part of Bio::Tools::SiRNA).

Bio::Tools::Sim4::Results

Issue #1644 fixed - distinguish between EOF and no exons found

I'm paranoid - let's not really force the line to start with this

Issue #1644 fixed on branch - distinguish between EOF and 'no-exons-found'

Bio::Tools::Spidey::Exon

Add Ryan's modules

Bio::Tools::Spidey::Results

Add Ryan's modules

Bio::Tools::WebBlat

Added a module that runs a blat at ucsc using the standard web form cgi

Doc update

Bio::Tools::dpAlign

Implemented global alignment using dynamic programming.

Now supports custom subsitution matrix for protein

Fixed the write_pretty_str_align call

Fixed line 376 with call to Bio::Ext

Fully qualify the 'Align' package

Bio::Tools::tRNAscanSE

TRNAscan-SE parser of tRNA location predictions in genome; not parsing structure at this time

Bio::Tree::DistanceFactory

Distance Factory for generating phylogenetic trees from pairwise distance matricies - UPGMA implemented

Neighbor-Joining implemented, WooHoo

Removed left-over that perl would complain about.

Bio::Tree::Draw::Cladogram

A Cladogram drawing module

A Cladogram and Tanglegram drawing module

A Cladogram and Tanglegram drawing module

Bio::Tree::Node

Suppport non-numeric values for bl for PAML labels - although they should probably go on bootstraps

Support new method for auto-quoting ids which have normally unallowed values in node ids and labels

Leaf nodes have height 0

Bio::Tree::NodeI

Support new method for auto-quoting ids which have normally unallowed values in node ids and labels

Depth function to get 'how far we are from the root'. Height is 'how far are we from the bottom (0)'. Small change to get_all_Descedents so that DFS should work properly now

Bio::Tree::Statistics

A simplistic (and probably still flawed) boostrap counting method - need to switch to consensus tree building instead I expect

Bio::Tree::TreeFunctionsI

Document the function

Doc inconsistent with method name

Doc fix

Correction to how reroot works when requested for a 'leaf' node and how internal 'fake' root nodes are removed during the re-root process. Thanks to John Calley for illustrating the bug

Bio::TreeIO::TreeEventBuilder

Support labeling root nodes

Bio::TreeIO::cluster

Guillaume Rousse's code for turning Algorithm::Cluster::treecluster into a tree

Bio::TreeIO::lintree

Fix Issue #1614 - was returning empty trees instead of undef when got to end of file

Fix Issue #1614 - was returning empty trees instead of undef when got to end of file

Valentin's fix for long taxaids; Issue #1625

Valentin's fix; Issue #1616

Merge on branch Valentin's fix; Issue #1616

Bio::TreeIO::newick

Print out a tree count at the beginning of the tree output which is needed for things like protml and codeml

Special case for PAML branch labels

Rename doc to 'new' to avoid confusing

Support labeled root label, fix problem with single node clades and their ancestors getting written as sister nodes. Added new function id_output which auto quotes ids which contain spaces, parens,;,commas

Bio::TreeIO::nexus

Support MrBayes produced trees

Some cleanup- wasn't matching all nexus formats

Handle nexus trees a little better I hope - more testing needed

Issue #1656

Back out attempted MrBayes parsing patch - shouldn't be needed; #1619

Support writing NEXUS format trees - still need some more documentation

Bio::TreeIO::pag

Added pagel format output

Documentation added

Based on Mark/Daniel's code more closely - still needs work to fully support the phylip2pag module

Print trait count based on the number of traits seen (we assume that trait have been consistently assigned for all tip nodes - better code to check this in the future

Bio::TreeIO::svggraph

Parameterize values in SVG writing - from Guilaume Rousse

Bio::WebAgent

Fix Revision string problems

FAQ

Added a FAQ about the project design and links to mailing list

Merge FAQ changes onto branch

Where is Factory::EMBOSS?

Oops - Brian accidently committed html not txt here

Regen faq with updates

Info about bioperl-ext

Remove 7

Add back

Added citation to MBE paper using Bio::PopGen modules

Makefile.PL

Added SVG::Graph dep

Add module names that use SVG or SVG::Graph

Add Clone and AutoClass

Mention XML::SAX dependancy

biodatabases.pod

Paragraph about OBDA

Add over/back

Correct bioperl-db module names, some editing

bioperl.pod

Sentence on HOWTOs

Edits. Remove repetitive, rearrange

bioscripts.pod

Extra line

Add mention of nexus2nh.pl

Removed script

Add mention of contig_draw.PLS

Add mention of search2table.PLS

Added a script to turn fasta -m9 into NCBI-like m9 output

Added pagel format output

Add mention of bp_embl2picture.PLS

Remove this mention, Tools::RestrictionEnzyme is no longer supported

Describe genbank2gff3.PLS

bptutorial.pl

Change version

Add maf to AlignIO list

Add tinyseq

A little more detail

Add mention of dpAlign

Fix links, add links

Fix Issue #1598

Old link

Add mention of AlignIO po format

Add KEGG to list of formats

Require XML::Writer so that tutorial.t fails more gracefully

Add Jurgen's example

Add AGAVE

Add mention of interpro

List of formats in 1 file, not 3

Edits

S/desc/description/; supress empty string warnings on print

doc/makedoc.PL

Dos2unix

examples/biblio/biblio_examples.pl

Useful example

Show ultra-abbreviated version as well

Vocabulary, title

Note that there are 3 repositories whose contents aren't necessarily the same!

Dos2unix

examples/contributed/rebase2list.pl

Dos2unix

examples/generate_random_seq.pl

Dos2unix

examples/longorf.pl

Dos2unix

examples/revcom_dir.pl

Dos2unix

examples/searchio/blast_example.pl

Format

examples/searchio/custom_writer.pl

Added -fh=>\*ARGV to SearchIO input constructor calls to enable STDIN or @ARGV reading, given change in Root::IO::_readline() as of release 1.303.

examples/searchio/hitwriter.pl

* Added -fh => \*ARGV to the input SearchIO constructor call. * Added note about which cols require HSP alignment data. * Ouptputting query length and number of hits for each report.

Exercising the interfaces a little more.

Migrating revision 1.2.2.1 to the main trunk.

examples/searchio/hspwriter.pl

Added -fh=>\*ARGV to SearchIO input constructor calls to enable STDIN or @ARGV reading, given change in Root::IO::_readline() as of release 1.303.

examples/searchio/htmlwriter.pl

Added -fh=>\*ARGV to SearchIO input constructor calls to enable STDIN or @ARGV reading, given change in Root::IO::_readline() as of release 1.303.

examples/searchio/psiblast_features.pl

Added -fh=>\*ARGV to SearchIO input constructor calls to enable STDIN or @ARGV reading, given change in Root::IO::_readline() as of release 1.303.

examples/searchio/psiblast_iterations.pl

Added -fh=>\*ARGV to SearchIO input constructor calls to enable STDIN or @ARGV reading, given change in Root::IO::_readline() as of release 1.303.

examples/searchio/rawwriter.pl

Added -fh=>\*ARGV to SearchIO input constructor calls to enable STDIN or @ARGV reading, given change in Root::IO::_readline() as of release 1.303.

examples/searchio/resultwriter.pl

Added -fh=>\*ARGV to SearchIO input constructor calls to enable STDIN or @ARGV reading, given change in Root::IO::_readline() as of release 1.303.

examples/searchio/waba2gff.pl

Added -fh=>\*ARGV to SearchIO input constructor calls to enable STDIN or @ARGV reading, given change in Root::IO::_readline() as of release 1.303.

examples/searchio/waba2gff3.pl

Unroll WABA states into suitable GFF3

examples/structure/structure-io.pl

Strucure example

Dos2unix

examples/tools/restriction.pl

Remove this, Tools::RestrictionEnzyme is no longer supported

examples/tools/run_genscan.pl

Better, more Bioperl-ish

Correction

examples/tools/standaloneblast.pl

Cleanup per Issue #1598 due to migration to SearchIO by default instead of BPlite

Small api change fix as well

Fix per Issue #1598 on the branch

Dos2unix

maintenance/authors.pl

Add Id line

maintenance/modules.pl

Added definitions of class categories as written down by Albert

Option ---dir vasnot documented

Add Id line

maintenance/pod.pl

Add Id line

maintenance/version.pl

Add the version declaration for each module

Add Id line

models/README

Add mention that these are version 1.0 diagrams

scripts/Bio-DB-GFF/bp_genbank2gff.PLS

Updating docs

scripts/Bio-DB-GFF/bulk_load_gff.PLS

Added a workable mechanism to change the grouping behavior in the ninth column of GFF2

Added some timing diagnostics to the gff load scripts; you will need Time::HiRes to see this feature

Quenched a bug that prevented Bio::DB::GFF from loading attributes whose values are zero.

Removed debugging code inadvertently left in the bulk loader

-Changed streaming behavior to allow multiple embedded sequences -Added option to process very large numbers of GFF files (> kernel limit for command line arguments)

Added warnings regarding the maxbin value

Fixed a mess of errors introduced when I tried to add bin overflow checking to the GFF loaders

Give a hint as to what the size so we can more easily adjust MAX_BIN

-brought fast_load_gff into sync with bulk_load_gff wrt 1) support for multiple embedded sequences 2) support for large numbers of files > kernel limit for command line args -removed STDIN dependency in both scripts for standalone --fasta loading

-merged Lincoln's maxfeature changes back into fast_load_gff -fixed regex that was not sorting compressed fasta and gff files in @ARGV properly

scripts/Bio-DB-GFF/fast_load_gff.PLS

Added a workable mechanism to change the grouping behavior in the ninth column of GFF2

Added some timing diagnostics to the gff load scripts; you will need Time::HiRes to see this feature

Added warnings regarding the maxbin value

Fixed a mess of errors introduced when I tried to add bin overflow checking to the GFF loaders

-brought fast_load_gff into sync with bulk_load_gff wrt 1) support for multiple embedded sequences 2) support for large numbers of files > kernel limit for command line args -removed STDIN dependency in both scripts for standalone --fasta loading

-merged Lincoln's maxfeature changes back into fast_load_gff -fixed regex that was not sorting compressed fasta and gff files in @ARGV properly

scripts/Bio-DB-GFF/genbank2gff3.PLS

A script to convert Genbank flatfiles into GFF3 suitable for gbrowse on a Bio::DB:GFF backend. It uses Chris Mungall's Unflattener libraries and borrows elements from Scott Cain's GenBank unflattener. This script handles gene-related feature IDs differently to avoid loss of alternative splice variants that have common three prime and five prime ends. It has been tested on the refseq genbank-format files for the mouse and human genome builds as well as a few 3rd party annotated genbank accessions.

Modified TypeMapper.pm to translate 'Protein' to 'protein' (perhaps there ought to be a general 'lower' and then deal with exceptions like CDS). Also modified genbank2gff3.PLS to produce a full reference sequence line in addition to the ##sequence-region directive.

Fixing a bug where '0' would print for a gff line if problems happened

Added --ethresh option for unflattener; setting this high raises the threshold at which errors are showstoppers

scripts/Bio-DB-GFF/load_gff.PLS

Added a workable mechanism to change the grouping behavior in the ninth column of GFF2

Added warnings regarding the maxbin value

Fixed a mess of errors introduced when I tried to add bin overflow checking to the GFF loaders

scripts/Bio-DB-GFF/meta_gff.PLS

Added simple Bio::DB::GFF meta-data getter/setter

scripts/Bio-DB-GFF/pg_bulk_load_gff.PLS

Mirrorring the group preference stuff lincoln added to the other loaders

scripts/Bio-DB-GFF/process_ncbi_human.PLS

Script out of date and being removed until a replacement script can be produced

scripts/DB/bioflat_index.PLS

Fixed transcript glyph so that nonstranded features are not automatically treated as + strand

scripts/DB/biogetseq.PLS

Fixed Bio::DB::Failover to properly pass get_seq_by_version() method, and fixed Bio::DB::Flat::BDB to properly implement it

scripts/biographics/bp_embl2picture.PLS

Added the EMBL/Genbank entry renderer to the scripts directory, since it is probably of general interest

scripts/biographics/bp_glyphs2-demo.PLS

Added new glyphs

scripts/graphics/contig_draw.PLS

This is the start of another demo that draws contigs./

scripts/graphics/frend.PLS

Fixed handling of reverse strands in the super-short version of the featureFile renderer

scripts/index/bp_fetch.PLS

Allow ":" in the sequence id name

scripts/index/bp_index.PLS

Case insensitive search for the module to use

scripts/searchio/README

Added a script to turn fasta -m9 into NCBI-like m9 output

scripts/searchio/fastam9_to_table.PLS

Added a script to turn fasta -m9 into NCBI-like m9 output

Oops - fix so that we match lines where number is > 999

Deal with possibility of regexp in the sequence name

scripts/searchio/search2table.PLS

This looks like a *PLS script to me!

scripts/searchio/search2table.pl

Searchio reports into a tabular format like NCBI's -m 9 format

This looks like a *PLS script to me!

scripts/seq/seqconvert.PLS

Fixed it so that it recognises chadoxml

scripts/seq/split_seq.PLS

Support cmdline and STDIN file with transparent ARGV handle

scripts/seq/translate_seq.PLS

Lincoln's changes to initialize_io mean that magic <> will not pick up cmdline stuff - either pass in on cmdline

Default to fasta format

scripts/seq/unflatten_seq.PLS

See bp maillist for full desc the script seq/unflatten_seq will now generate GFF3 - the unflattener module is used to build the 'feature graph' connecting genes, transcripts, exons and CDSs together. This means we can have GFF3 for anything in genbank! As far as I'm aware, the only other sensible output formats to use here (ie formats that support feature graphs/containment hierarchies) are: chado, chaos, and the write-only asciitree. This feature graph is written out in the GFF3 using the ID and Parent tags. To do this there is an extra intermediate step - the bioperl FeatureHolderI hierarchy is traversed and ID/Parent tags are generated. Here is a description of the changes I have made: [unless you're a bioperl hacker you don't really need to read the rest of this] You can get the context of what I'm on about from this thread: http://bioperl.org/pipermail/bioperl-l/2003-December/014150.html Two new public methods: FeatureHolderI->set_ParentIDs_from_hierarchy sets both ID and ParentID from FeatureHolder hierarchy SeqFeatureI->generate_unique_persistent_id this is required by the above method Lincoln wanted this to be private, but I think it has to be called from outside FeatureHolderI->create_hierarchy_from_ParentIDs the inverse of set_ParentIDs_from_hierarchy

Moved ID methods to new class IDHandler see thread 'new GFF3 support methods' http://bioperl.org/pipermail/bioperl-l/2004-March/thread.html

Rationalised ID generation code in chaosxml write adapter changed verbose reporting in unflattener to use stderr

scripts/taxa/local_taxonomydb_query.PLS

Some slightly more interesting examples

scripts/tree/TAG

Added pagel format output

scripts/tree/nexus2nh.PLS

Convert to proper PLS

scripts/tree/nexus2nh.pl

Convert nexus trees to newhampshire (but maintain long taxon names)

Convert to proper PLS

scripts/tree/tree2pag.PLS

Added pagel format output

scripts/utilities/bp_sreformat.PLS

Autodetect MSA formats, also die with different message when file cannot be opened

Support same default option of grabbing file from cmdline

Accept cmd-line stuff again

Allow cmdline setting of displayname to flat

scripts/utilities/dbsplit.PLS

Some defaults

scripts/utilities/mutate.PLS

Little bug

Dos2unix

scripts/utilities/pairwise_kaks.PLS

More verbose warnings

Use uc for uppercase

scripts/utilities/search2gff.PLS

Search2gff does the right thing now - now strand is done right properly

More cleanup for GFF3 supported

Simplify code - get min/max for match from HSP data rather than computing from tile overlap - get match container score from hit->bits

Quiet if wanted

scripts/utilities/search2tribe.PLS

Support old behavior