Dot-bracket (DB): RNA secondary structures ------------------------------------------------------------------------ DB file: Content format ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ In a DB file, each RNA structure record comprises two parts: - A header that contains the name of the structure. - A body that contains the structure on one line (and for only the first structure, the RNA sequence on the line before). See `DB format`_ for more information. DB header line """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" The first line of each structure record is the name of the structure. DB body lines """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" For the first structure, the line after the header is the RNA sequence, which can contain A, C, G, U, and N. The structure is encoded on one line as a series of dots (``.``) and brackets (``(``/``)``, ``[``/``]``, ``{``/``}``, ``<``/``>``). A dot indicates that the base does not pair with another base. A bracket indicates that the base pairs with the matching bracket. DB file: Path format ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ DB file extensions """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" SEISMIC-RNA accepts the following extensions for dot-bracket files: - ``.db`` (default) - ``.dbn`` - ``.dot`` DB path parsing """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" Every output DB file is named after its mutational profile. DB file: Uses ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ DB as output file """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" - The ``fold`` command outputs a DB file for each predicted structure. .. _DB format: https://rna.urmc.rochester.edu/Text/File_Formats.html