RepeatModeler 2.0.7

• Added the RepeatAfterMe auto-extension step to Refiner. This is applied to both
  RepeatScout and RECON generated family models.
• RepeatClassifier now checks for a complete Dfam FamDB database.
• A new redundancy filter is applied to RepeatScout families.
• Lowered masking score threshold to reduce rediscovery.
• Fixed a bug with the recoverDir round detection logic.

RepeatModeler 2.0.6

• Updated to RepeatScout 1.0.7 which has the capability of reporting the
  extended coordinates of the seeds. This reduces the complexity searching
  for examplars using the consensus sequence.
• Reduced the minimum family instances for RepeatScout families.
• Fixed a bug that caused RepeatClassifier to run with only one thread in the
  last release.
• Fixed a bug that caused some RepeatScout derived exemplars to be reported
  on the wrong strand.
• Isolate long sequence identifiers from LTR_retriever which will alter
  names longer than 13 characters.
• Fixes for LTR_retriever changes in TRF parameter.
• MultAln - new API for calculating Kimura divergence
  - new API for slicing operations on MSA
  - switch to Smitten V2 for default stk output
  - improvements to the Linup reporting format
• Refiner reorganization for future extension algorithm

RepeatModeler 2.0.5

• Fixed a bug that caused failures in "absolute" reproducibility. Prior
  to this release use of the "-srand" would not gaurantee that the ouputs
  consensi.fa.classified and families-classified.stk were exactly the same
  in sequence and sequence order. It did gaurantee that the same samples
  were drawn from the genome, and that equivalent scoring families were
  derived at at each step. In this release secondary sorts were added
  to gaurantee a fixed sort order among equally scoring results, generating
  exactly the same output files each time the random number generator seed
  is used. NOTE: This change only applies to results generated with
  this version and future releases.
• Added "-long" option to faToTwoBit to support larger genomes.

RepeatModeler 2.0.4

• Improved the gathering of RepeatScout examplars for building
  seed alignments.
• Parallelized and improved the masking between rounds for faster
  runs and fewer redicoveries.
• The 'pa' (parallel batches) option has been replaced with a new
  'threads' option which maps directly to the maximum number of
  threads the program will attempt to use.
• Takes advantage of RepeatMasker 4.1.4 and RMBlast 2.13.0 parallel
  query feature.
• Larger default sample sizes are now possible with speed improvements.
  The original sampling strategy can be selected with the new 'quick'
  option.
• ABBlast is no longer supported.
• Fixed a few visual artifacts in the viewMSA.pl html
  visualization.

RepeatModeler 2.0.3

• The program now generates a logfile in the working directory named
  -rmod.log. This file contains the random seed number used
  and some high level stats on the run for use with reporting problems
  with the program.
• Fixed a problem with the orientation/coordinates provided in the
  Stockholm output format that affected a subset of the sequences.
• Fixed a bug affecting the trim functions of the Linup tool.

RepeatModeler 2.0.2

• First release of a set of manual curation tools for use with de-novo generated TE libraries.
• Added generateSeedAlignments.pl to generate Dfam compatible seed alignments given a consensus based TE library and RepeatMasker output.
• Fixed bug in N50 calculation.
• Fixed Ruzzo-Tompa maximal scoring subsequences implementation.
• Several minor bugfixes.

RepeatModeler 2.0.1

Bugfix release

• Work around a bug introduced in the new NCBI Blast 2.10.0 with version 5 databases.

• RepeatScout in some rare cases will generate models for very long-period satellites. This can cause Refiner to go crazy creating tons of off-diagonal alignments. This version filters out these rare cases.

• This version now prints out the version of each dependency in the log.

RepeatModeler 2.0

RepeatModeler 2.0 now supports LTR structural search using a combination of LTR_harvest and LTR_retriever.

RepeatModeler open-1.0.11

Bugfix release. Avoid bailing if Refiner could not produce a consensus from the family sequence set.

RepeatModeler open-1.0.10

• Bugfix release. Fixes a bug that will cause RepeatModeler to exit if no new repeats are detected after round-2. In typical runs repeats will continue to be discovered even up to the last round and this bug will not be exercised.