Today, I released a full Tutorial on using ShortStack, complete with step-by-step instructions and a full set of sample data.  The tutorial is at http://axtelldata.bio.psu.edu/data/ShortStack_TestData/.  In addition, a new ShortStack version was released (0.4.2), which is simply an update to the README documentation to reflect the new tutorial data.  The tutorial should make it easier for users to begin using ShortStack productively in their work. Finally, I'm happy to report that a manuscript describing ShortStack and its performance in seven species (four plants, three animals), has been accepted for publication in the journal RNA.

As always, comments and bug reports welcome. -- Mike Axtell

ShortStack 0.4.0 has been released.  This is a major update which offers substantial speed improvements relative to previous versions.  The speed improvements are due primarily to a new filtering method that reduces the number of RNALfold calls.  Previously, all small RNA clusters were subjected to RNALfolding (except those that were too long -- see documentation).  In this new release, highly repetitive clusters, and those that are not dominated by small RNAs in the acceptable size range (in other words, have DicerCall of "N") are not eligible for secondary structure analysis.

Posted update to 0.2.3 today, which fixes an error in the formatting of the BED files ... in all older versions the blockStarts field (column 12 of the bed line) was incorrectly reported in 1-based coordinates.  With this fix, blockStarts are now correctly listed in zero-based coordinates per the BED specification.

Major update in this release.  Most significant change is a major improvement in the accuracy of calling hairpin-derived, but not miRNA, loci.  Many fewer false positives / borderline cases are called by this new release, while annotation of longer hairpins (e.g. loci like Arabidopsis IR71) is much improved.  See the ShortStack page.

New release is here.  Changes:

1) Prep_bam.pl helper script updated
    to deal with .sam lines corresponding to unmapped reads without
    breaking. This also enables "one-step" mapping and .bam file preparation
    using bowtie (see 'Quick Start' section below).

2) ShortStack.pl also
    updated to prevent breaking upon encountering .sam lines corresponding
    to unmapped reads.

3) Coloring of .bed files now dynamically adjusts to
    make a nice 'rainbow' (from red through green to blue) regardless of the
    size of the dicer range.

4) Finally, an additional filter was added to
    prevent excessive splitting of initial de-novo clusters into
    hairpin-associated clusters ... initial clusters are now prevented from
    splitting if they would spawn more than --maxsplit child hairpins
    (default, 3). This prevents inappropriate splitting of very large
    clusters that might harbor a number of sub-optimal hairpins.