Changes in version 0.2.6                        

  - Resolve instrument model metadata locally instead of querying remote
    services
  - Make fileMetaData() extraction offline-safe and simplify timestamp /
    sample ID handling
  - Remove stale package metadata from DESCRIPTION
  - Update README links and wording
  - Refresh tests for the current detectPeaks() API

                        Changes in version 0.2.5                        

  - Remove mzR functions for data handling and replace with MSnbase
  - Streamline parallel processing using furrr and future_map

                        Changes in version 0.2.4                        

  - Add removeSample method to remove unwanted samples form SRM objects

                        Changes in version 0.2.3                        

  - Add parallel support via future_map to openSRM and detectPeaks
  - Add keepTransitions method to remove redundant data from SRM objects

                        Changes in version 0.2.2                        

  - Use the CentWave algorithm for peak detection
  - Remove explicit slot for transformed chromatograms
  - Standardise transition naming
  - Add peakAsymmetry method
  - Add method for creating peak groups
  - Export results using peak group IDs
  - Add backend as an argument in openSRM. Select either mzR (Default)
    or q3ML (this is hopefully a short term solution to mzML files which
    have been converted using newer versions of pwiz, and are currently
    not supported by mzR.

                        Changes in version 0.2.1                        

  - Support for Shimadzu QqQ-MS

                        Changes in version 0.2.0                        

Breaking Changes

  - Multiple file input. SRM objects can be used for single mzML files
    or batch mode for multiple files
  - Basic support for peak smoothing (Savitzky-Golay implemented at
    present)
  - Support for chromatogram feature detection (pracma & xcms
    matchedFilter)
  - Improved plotting methods
  - Complete re-write of openSRM function and all underlying methods
  - All previous methods form v0.1.4 and lower; now deprecated

                        Changes in version 0.1.4                        

  - The main change is that extraction and decoding of chromatogram time
    and intensity arrays is now performed by mzR.
  - xml2 is still used for extraction of header data which is currently
    unsupported by mzR
  - A lot of the internal workings have been updated / replaced with
    more efficent approaches; mainly utilising dplyr, tibble and purrr
  - New method export_to_skyline exports SRM transitions in a format
    that can saved as .csv and uploaded to Skyline
  - The peaks slot in SRM-object is now a list of tibbles
  - index slot in SRM-object incorporates polarity (- or +)
  - SRM-object headeris now a tibble
  - Bug fix in plotAll which caused multiple outputs during layering of
    plots
  - Bug fix in plotMulti which caused switching of colours between
    geom_lines and labels
  - Bug fix in plotAll which was causing for a new index order to be
    created
  - Additional labelling options (default or custom) for plotMulti
  - Added vignette

                        Changes in version 0.1.3                        

  - Add plotMulti for easy graphical representation of complex SRM-MS
    chromatograms
  - Cleaned up filter naming throughout
  - Fixed CI build environment, so there is no longer a list of
    unnecessary Imports in DESCRIPTION
  - openSRMfile wasn't checking for file compression and assuming that
    file were converted with no compression. openSRMfile now checks for
    zlib compression and if found passes the gzip compression type arg
    to sRm::decodePeaks
  - Instrument, Instrument Model and Instrument Serial are all separate
    fields in the meta method
  - Internal helper function add (scanPolarity) to extract scan polarity
    from cvParam and insert into object@header
  - Parent (Q1) and product (Q3) masses are extracted from cvParams
    using getQMZs and are no longer string processed from filter names
  - filter slot added to the SRM class. This contains a raw copy of
    idRefs
  - object@header now includes scan polarity; -1 = negative, 1 =
    positive
  - totalIonCount has become totIonCount to improve readability
  - combineTransitions now does a "best effort" at checking for
    transitions which have been acquired using dynamic MRM mode
      - In these circumstances; a transition set needs to consider the
        retention window aswell as the parent m/z
      - Unique filter ID strings are created from parsed idRefs
      - Where Rt lengths are unequal within unique transitions set;
        these are (for now) removed from the transitions class with a
        on-screen message stating which idRefs have been dropped

                        Changes in version 0.1.2                        

  - Fix no visible binding' NOTE for 'ggplot2
      - Update manual
      - Increase test coverage

                        Changes in version 0.1.1                        

  - Adapt combineTransitions to catch SRM's of different lengths
      - Result of combineTransitions is now S3 class
      - Plot method for combineTransitions class ;(plotAll)
      - Plot method for single transition events; (plotSRM)
      - Bug fixed for when time and intensity arrays are encoded with
        different precisions
      - grid and gridExtra are listed as Dependencies

                        Changes in version 0.1.0                        

  - First stable release
      - Reading and parsing of Thermo TSQ Vantage SRM data
      - Single function for reading and decoding peaks; stored to S4
        object
      - S4 method for combining all transitions from unique parent (Q1)
        m/z value
      - S4 methods for viewing object: show, meta, transitions
      - Basic tests (testthat + codecov)
      - Travis and AppVeyor Continuous Integration
      - Passes R CMD check with no ERRORS or WARNINGS