v1.1.4c

Known issues

• set_diagnose.pl runs into errors with the snp matrix header. However, the pipeline has finished running by this point, and the results are still valid.

v1.2b

Version 1.2b

v1.1.8b

• Fixed coordinates problem when multithreading phage masking
• Note: this version has been superseded by version v1.1.8c

v1.2

• Stampy.py fix from @chrisgulvik
• Makefile is refactored; installation is much better
• Merging VCFs has been made more stable

v1.1.8

Changes:

• Stampy mapper added, but smalt is still the default mapper.
• Better diagnose script
• Better parallelization in makeRegions.pl; smarter ways to get regions in makeRegions.pl
• Fixes for clade distances script
• Fixed parallelization issue with mergeVcf

v1.1.4b

• Fixes a racing condition found in v1.1.4
• NOTE: v1.1.6 is a newer version
• NOTE: a more stable version of 1.1.4 has been released as v1.1.4c

v1.1.6

• BayesHammer should now be working correctly
• Some fixes on cladeDistances
• Fixed known bugs on race conditions for varscan
• Outputs a SNP-only file
• NOTE: this release is superseded by a more stable release v1.1.6b

v1.1.4 - unstable

• Edit, 2015-08-28 A bug was found where multiple contigs in the reference genome will cause race conditions. Therefore, please instead download v1.1.4b or a newer version.
• SNP annotation with SnpEff (still kind of experimental; thanks to @teishi for the main script)
• Multithreaded mpileup and VCF merging, ie, faster SNP calling
• To multithread and find equally sized regions, there is now the helper script makeRegions.pl.
• SNPs-only VCF is created for viewing purposes. However it is not used for downstream analysis and so SNP calls have not been changed.
• Fixed a problem where /tmp/ is used at one point for a really large file; it is not used anymore and so you aren’t limited by space anymore
• Perl libraries are now discoverable; no need to change PERL5LIB in your environment. Only PATH.
• Clade distances script - find distances between output clades. Still kind of experimental; thanks to @andrewdhuang
• Can now use shuffleSplitReads.pl to shuffle a whole set of paired-end reads; makes it easier to start off a Lyve-SET project.
• Main output files are put into the main Lyve-SET directory as symlinks.
• set_diagnose.pl: helps you start investigating if anything went wrong in your Lyve-SET run. I think there is a lot of potential to put more things in this script; suggestions welcome.
• Employs samtools v1.2 instead of v1.1
• Easier-to-look at log folder. Timestamps are now shown for start/stop in the log. All options that were used are now shown in the log.
• Many, many bug fixes

v1.1.2 stable

• Basis for SnpEff. However it is not installed by default yet.
• Now there is a definite and streamlined process going from pooled vcf -> matrix -> alignment -> trees. Good paper trail.
• Misc bug fixes
• New utility script to shuffle a whole bunch of split reads. Perfect for getting something like a MiSeq run into the interleaved format that Lyve-SET requires.

Version 1.1.0 unstable

note: I found a bug in this and am marking it as unstable

• Informative alignment creation is more transparent
• Config files under /config; presents under /config
• sampling sites for --fast is more accurate and correct, for finding the best outgroup
• Masking for cliffs in bam files for individual genomes
• Separated Lmono test dataset to clade1 and clade2
• New Salmonella dataset (thanks Ruth!)
• Updated installation for Perl modules under make install
• Simplified names in the MSA directory
• Streamlining more in the LyveSET.pm perl module, e.g., logmsg() and file extensions
• moving from using _processMsa to _processVcf script
• Some better documentation
• Cleaning mechanism using either CGP or BayesHammer
• Lots of bug fixing