#CHANGE LOG for ALEXA-Seq Analysis Code 

#ALEXA-SEQ VERSION 0.1
- Beta release for external collaborators

#ALEXA-SEQ VERSION 1.1
- First public release

#ALEXA-SEQ VERSION 1.2
- Numerous bug fixes and updates
- Performed complete code test for both the annotation and analysis arms
- Tested the annotation arm by generating an annotation DB for Zebrafish: 'dr_57_8c' (now available online at www.AlexaPlatform.org)
- Tested the analysis arm by generating analysis for the project: 'LnCAP_AR_KnockIn' (now available online at www.AlexaPlatform.org)

#ALEXA-SEQ VERSION 1.3
- Numerous minor bug fixes
- Added Xapian-Omega template files to the source code repository (as examples)
- Added additional project and annotation configuration file examples

#ALEXA-SEQ VERSION 1.4
- Fixed missing directories bug in fresh build of analysis dirs
- Added extra instructions for automated installation of BioPerl
- Fixed bug with jobs that determine the number of quality read counts
- Simplified summary of quality read counts and average read length
- Modified calculation of library complexity jobs to avoid collisions between multiple projects being processed at the same time
- Fixed bug caused by unusual duplication of a few ensembl exon record in EnsEMBL 57
- Fixed minor bug in naming of junction and boundary parsing log files
- Removed accidental hard coding of server name in some expression calculation jobs
- Added code to help check completion status of parsing jobs

#ALEXA-SEQ VERSION 1.5
- Updated code to allow better control of stringency of hits to exon junctions/boundaries (particularly important for short reads such as 36-mers)
- Updated examples parameter settings for different read lengths 
- Removed all mapping and assignment parameters from the project configs files.  These parameters are now determined automatically and without input from the user.
- Modified use of mapping and assignment parameters to accomodate libraries made up of lanes with different read lengths
- Added a utility to assist in the resetting all read assignments
- Added additional memory usage monitoring to read assignment steps
- Fixed bug in creation of ALEXA-seq annotation databases that resulted in missing Entrez gene name information
- Reduced memory footprint of all read assignment jobs
- Improved tracking of creation of ALEXA-Seq viewer data for each gene
- Added code to assist in packaging of results files for distribution to collaborators

#ALEXA-SEQ VERSION 1.6
- Added sanity check for non-unique flowcell-lane combinations
- Added sanity checks for lane, library and comparison records in user provided config files
- Removed hard coding of navigation links for ALEXA-Seq gene records, etc.
- Modified expression calculations so that values are normalized for library size differences according to mapped bases instead of mapped reads
- Eliminated hard coding of Google Analytics ID and made this a global parameter specified by the user
- Updated example configuration files and re-organized these
- Improved efficiency of File I/O usage for all expression calculation steps and added QC checks to ensure these steps completed successfully
- Updated mapping parameter values to accomodate reads up to 100 bases in length
- Added lane-by-lane quality analysis for each library to the ALEXA-Seq viewer 
- Added lane-by-lane summary of mapping statistics to the ALEXA-Seq viewer
- Added position bias analysis for each library to the ALEXA-Seq viewer
- Added percent gene coverage analysis for each library to the ALEXA-Seq viewer
- Added tag redundancy (library complexity) analysis and plots to the ALEXA-Seq viewer 
- Added basic library info to the ALEXA-Seq viewer
- Added highlighting of Significant AE events to the ALEXA-Seq viewer

#ALEXA-SEQ VERSION 1.7
- Various minor bug fixes
- Various performance improvements
- Added automatic generation of candidate gene lists for AE+DE genes, AE genes only, gains only, and losses only
- Updated main project summary records in data viewer to include more basic info on the project and the analysis
- Updated gene records in data viewer to use smarter color schemes in line plots
- Added gene model diagrams to gene records in data viewer
- Added automatic generation of candidate peptides for design of antibodies that are potentially specific to each condition

#ALEXA-SEQ VERSION 1.8
- Various minor bug fixes
- Various display tweaks of the data viewer
- Implemented use of BWA aligner (the user can select to use BLAST only or BWA+BLAST)
- Added second method of estimating library complexity - using mapping positions instead of sequence identity
- Added beta code to modify data viewer for larger datasets (i.e. projects involving many libraries)
- Added beta code to perform statistics on comparisons of library groups in addition to pairwise library comparisons
- Added code to automate installation of ALEXA-Seq annotation databases and associated mysql databases 

#ALEXA-SEQ VERSION 1.9
- Reduced memory usage of estimate library complexity jobs
- Added BWA only option
- Various minor bug fixes related to implementation of BWA alignments

#ALEXA-SEQ VERSION 1.10
- Various minor bug fixes in code
- Various bug fixes and performance improvement within the Virtual Machine
- Added beta code that optimizes the data viewer for projects with larger numbers of libraries 
- Added beta code that performs differential and alternative expression analysis using GROUP comparisons instead of pairwise
- Added beta code that used ANCOVA to identify interesting AE events between arbitrary numbers of sample groups 

#ALEXA-SEQ VERSION 1.11
- Fixed bug that broke the demo in Virtual Machines
- Added step to partition the ALEXA-Seq annotation database by genome region (allows for memory and fileI/O performance improvements in various downstream steps)
- Improved file I/O performance of gene record generation for the data viewer
- Improved file I/O performance for all expression calculation steps
- Reduced memory usage during identification of candidate genes
- Reduced memory usage during identification of candidate peptide sequences

#ALEXA-SEQ VERSION 1.12
- Various minor bug fixes
- Various performance tweaks for the Virtual Machine Demo
- Added code to identify the protein coding frame associated with all junction sequences
- Improved file I/O performance of gene record generation for the data
- Reduced memory usage during creation of Matrix data files
- Reduced memory usage during Differential and Alternative Expression calculations
- Reduced memory usage during identification of candidate genes (again)
- Improved file I/O efficiency during creation of gene records in data viewer

#ALEXA-SEQ VERSION 1.13
- Improved file I/O efficiency during creation of gene records in data viewer (again)

#ALEXA-SEQ VERSION 1.14
- Added script to clean up temp files at the end of an ALEXA-seq analysis
- Fixed regex in Web.pm to allow lib names with underscore
- Minor fixes to handling of fastq files
- Added new script to create dummy R2 qseq files for SE data
- Various other minor tweaks

#ALEXA-SEQ VERSION 1.15
- Minor formatting updates to create analysis commands steps
- Added scripts to assist in the creation of multiple junction/boundary sequence lengths
- Added information to manual and commands scripts explaining which jobs can be run concurrently

#ALEXA-SEQ VERSION 1.16
- Various minor bug fixes
- Improved performance of fastq data import
- Fixed problem with BWA alignment parsing that allowed some alignments with gaps to be scored too highly
- Added helper script to demonstate splitting large single lanes into smaller 'sub-lanes'

#ALEXA-SEQ VERSION 1.17
- Dramatically reduce memory usage of generate expression values (Gene/Exon) when there are many redundant alignments (e.g. high coverage of MT genes)
- Added library skew statistics and summary to the library summary output
- Added support for recent versions of Ensembl (tested on v65)
- Added support for null passwords in annotation database creation code
- Fixed various places in the website generation code where the junction/boundary length was being hard-coded instead of retrieved from global config file