Transposome version 0.12.1

Summary of changes in this version:

• Fix for #40 with reads not being found in cluster merging step. This was due to how different
  Casava versions are handled. The original routine was correct but an issue lingered due to
  this not always being invoked.
• The above fixes an secondary issue that was causing the annotated repeat fraction to be reported
  as being lower than in reality, though this has been resolved.
• Add YES/NO boolean for setting in-memory analysis in configuration. This should be more intuitive
  than the 1/0 form (which will still work).
• Remove the 'each' hash loop syntax from all classes. This is known to cause some issues, especially
  when adding/removing keys so we changed all methods to use 'keys' instead.
• Label repeat summaries in log as either theoretical or biological if they are from just clustering or
  clustering + annotated repeats, respectively.

See the "Changes" file for more information.

Transposome version 0.12.0

Summary of changes in this version:

• Refactor the complete analysis procedure to run individual analysis steps sequentially.
• Add Transposome::Analysis::Pipeline class for controlling analysis steps rather than having the code in one script.
• Add Transposome::Log class for getting/setting logging properties. This was previously all rolled into one script and now the methods and code are shared across numerous classes.
• Add execution scripts for running analysis steps instead of controlling the program flow from a single command script. This greatly reduces the memory usage of the full analysis, it improves the efficiency for large data sets, and delivers the same biological results.
• Remove redundant test script ('15-transposome_app.t') for running the full pipeline. The individual steps were already tested thoroughly so this set of tests became completely redundant. The full pipeline is also tested in the '14-analysis_steps.t' script.
• Fix bug with Illumina sequence format not being parsed correctly when the command line analysis was done.

See the "Changes" file for more information.

Transposome version 0.11.4

Summary of changes in this version:

• Update typemap to include 'Unknown_TIR' type.

See the "Changes" file for more information.

Transposome version 0.11.3

Summary of changes in this version:

• Change family mapping procedure in Transposome::Annotation::Mapping class to be a single abstract
  regex insted of trying to conditionally match names for each transposon type. This method captures
  far more annotated families, so it is more reflective of the TE diversity in a sample.
• Add 'TRIM' and 'LARD' LTR retrotransposon types to type map in Transposome::Annotation::Typemap
  class.
• Reverse sort annotation summary file to show largest families (by genome abundance) first.
  This was the old behaviour but it appears to have been reverted to an ascending sort.

See the "Changes" file for more information.

Transposome version 0.11.2

Summary of changes in this version:

• Bug fix for transforming sequence IDs that appeared to be Illumina but are not (fixes #37).
• Refactor SeqIO superclass to have ID setting method inherited by subclasses instead of repeating
  the code in each subclass.
• Refactor SeqFactory class to take a hash of options/arguments instead of repeating code to set
  class attributes based on the sequence format.
• Add 'seqtype' class attribute so that Illumina data can be safely transformed into a
  single ID with no spaces to work better with Transposome and external tools like BLAST.

See the "Changes" file for more information.

Transposome version 0.11.1

Summary of changes in this version:

• Add method to return the common superfamily name based on the commonly used 3-letter
  codes for transposons.
• Add tests for annotation methods that do mapping of family and superfamily
  names from IDs.

See the "Changes" file for more information.

Transposome version 0.10.1

Summary of changes in this version:

• Bug fix for directory structure of archived cluster FASTA files and
  annotations.

See the "Changes" file for more information.

Transposome version 0.10.0

Summary of changes in this version:

• Reduce fields in summary annotation file, and expand column headdings,
  to simply interpretation of the results.

See the "Changes" file for more information.

Transposome version 0.09.7

Summary of changes in this version:

• Fix type constraint for allowing numbers to be passed in with underscores.

See the "Changes" file for more information.

Transposome version 0.09.6

Summary of changes in this version:

• Report cluster number after clustering process so it is easy to run the standalone
  annotation step.
• Fix empty clusters (i.e, singletons) being written to the cluster file.
• Remove use of DBM-Deep, which was contributing to two bugs (a memory leak and an issue
  with creating large DBM files).
• Rework methods in PairFinder and Cluster classes to be more memory
  efficient. This includes using a new indexing method with SQLite. A number of methods now use
  autovivication to reduce the number of hash lookups and simplify the code.
• Fix type constaints on the format of numbers accepted by the Transposome::Run::Blast class.
• Rework tests for clustering methods to better handle behavior of indexing vs in-memory analysis.

See the "Changes" file for more information.