************************************************** * * * 1.0 SERIES NEWS * * * ************************************************** CHANGES IN GENOMINATOR VERSION 1.1 NEW FEATURES o Included an example on working with priming weights to the vignette as well as a section on gene models. o Added computePrimingWeights and addPrimingWeights; this is a method recentley developed by the authors to try to make RNA-Seq reads more uniformly distributed across expressed transcripts. o makeGeneRepresentation has been added and made public. This function computes Union, Union-intersetion, Background and Regions of constant expression (ROCEs) based on annotation. o validAnnotation is now public. o Initial rewrite of the "Working with the ShortRead package" vignette which will contain a fully workable analysis of data from the yeastRNASeq package. o The NEWS files is now parsable by the R function call news(package = "Genominator"). o Changed importFromAlignedReads to have only one argument 'x' instead of 'alignedReads' and 'filenames'. o Added functionality to importFromAlignedReads so that it is possible to immediately collapse several files together, at the cost of having them all in memory at the same time (this later point might be changed). o Added groupBy to summarizeByAnnotation. o Added to Bioconductor. o Improved User Guide substantially o Added CITATION file. BUG FIXES o Fixed a bug related to the deprecation of IRanges::matchMatrix. o Fixed two bugs discovered by Elizabeth Purdom regarding wrong indexing in .writeRegionsTable and importToExpData. o Changed some internals of the plotting functionality to make it work with a GenomeGraphs update. o The new version of DBI (0.2-5) fixes the name mangling of the SQL reserved keyword "end". Workarounds for this mangling has been removed and Genominator now depends on DBI >= 0.2-5. o Added a check to importToExpData so that only rows with non-missing values of (chr, location, strand) gets written to the database. o Fixed the connection pool to be outside of individual ExpData objects. This elucidated a hitherto unknown bug where a user could inadvertently write to a read-only database. o (Internal) Changed a few tryCatch statements to use the SQL "... IF EXISTS ..." instead. o Changed the db and filename to be in sync: there is now only a dbFIlename slot in ExpData and all functions accept dbFilename. CHANGES IN GENOMINATOR VERSION 1.0 NEW FEATURES o Changed underlying database model where now the database exists in two modes, either a read or write mode. This should be somewhat transparent to the end user. o Added importFromAlignedRead which allows the importing of objects read using readAligned from ShortRead package. o Added support for arbitrary index columns, this support allows for the inclusion of more complex types of reads, e.g. reads of different lengths or paired end reads. This support is preliminary, but should allow for a mix of representations. o Added '$' and '[' methods to expData objects to directly query the database. o Fixed vignette to more correctly describe splitByAnnotation. o Added ability to filter the reads when importing an Eland export file. o Added whereClause to summarizeExpData. o Added sample = FALSE argument to plot function for goodness of fit plots. This allows one to plot fewer than all of the points for smaller pdfs. o Added regionGoodnessOfFit.df to work with matrices or data.frames. o Added in-memory ordering to importToExpData; this speeds things up considerably when indexing. BUG FIXES o Fixed bug where colnames are trashed when preserveColnames = T and bindAnno = T o Modified fix below when we include "what" argument. o Fixed bug in splitByAnnotation when expand = TRUE and addOverStrands = TRUE and only one strand was present in the sequencing reads. o Fixed bug in splitByAnnotation when expand = TRUE where reads were lost. %%%% Local Variables: %%%% mode: change-log %%%% coding: utf-8 %%%% End: