S2S 2.0 Release Candidate 3
================================

Bugs fixed
----------

* the RNAVIEW algorithms was not able to annotate efficiently a tertiary structure with, at least, twice the same molecule

Improvements
------------
* S2S uses now only RNA algorithms exposed as Web Services (for details see http://serialized-thoughts.blogspot.com/2009/07/forget-installation-of-rna-algorithms.html)
* for the 2D prediction, the Contrafold algorithm has been added (http://contra.stanford.edu/contrafold/)
* for the 2D drawing, the VARNA tool has been added (http://varna.lri.fr/)
* when an helix is removed in the 2D panel, the secondary interactions can be kept as tertiary ones. For details see http://serialized-thoughts.blogspot.com/2009/07/new-features-to-edit-rna-secondary.html
* a new helix created in the 2D panel can overlap existing ones. For details see http://serialized-thoughts.blogspot.com/2009/07/new-features-to-edit-rna-secondary.html

S2S 2.0 Release Candidate 2
===========================

Bugs fixed
----------

* the user was not able to browse directories when he wanted to save or export some data
* the 2D generated with the "Infer 2D from Alignment" or "Infer Assemble Model from Alignment" was bugged
* 2D edition: if an helix deleted is at the 5'-end of an RNA, the full molecule became a single-strand

Improvements
------------

* when an RNA algorithm has no solution, a dedicated message is displayed (and not an error)
* if a sequence contains an unknown modified residue, if the user doesn't know the unmodified version (A, U, G or C), he can choose "?". But to be used carefully
* a plugin system is now available (check http://www.bioinformatics.org/s2s/plugins.html for details)
* it is now possible to use the embedded paradise platform to run user-defined groovy scripts (by typing "java -jar paradise.jar" in the lib directory)
* to infer an Assemble model, only one button is now available in the alignment toolbar. When the button is clicked, a 2D is inferred for the sequence selected and stored in a file as an Assemble model using the binary format.
  If the structural alignment has been constructed from a 3D structure, the 3D coordinates will also be exported.
  If the structural alignment has been constructed from a secondary structure, the Assemble model will contains no 3D coordinates.
  Exporting the 2D structure as an Assemble model allows to export non-canonical interactions inferred from the alignment. This is not possible with a CT export (as done by the previous release)
* when a new sequence is imported into the alignment (using the File->Import menu choice), S2S gives the choice to the user for a precomputed alignment
* to load Stockholm files, the suffix of the files has to be ".stk" or ".txt" (and not ".stockholm" anymore)
* to load FASTA files, the suffix of the files can be ".fasta", ".fna", ".fas",".fa" or ".txt"
* the algorithms_installation.sh script has been improved to warn the user if it is not launched from its own directory
* the Stockholm format parser can manage empty lines (for example between the "# STOCKHOLM 1.0" declaration and the first sequence). It can now parse files whose gaps are described using the '.' or '-' characters
* when S2S starts, it checks if the user has started it from its own directory. This is necessary to manage some algoritms like the SVG export of RNAplot (at least, until we find another workaround)
* the 2D model can be exported in an SVG file
* S2S can load an RNA secondary Structure stored in a FASTA file and described with the bracket notation (for details see http://serialized-thoughts.blogspot.com/2009/06/load-secondary-structure-described-with.html)