Skip to content
This repository has been archived by the owner on Jan 13, 2022. It is now read-only.

Latest commit

 

History

History
99 lines (86 loc) · 5.07 KB

RELEASES.md

File metadata and controls

99 lines (86 loc) · 5.07 KB
Raw

Scrappie

Scrappie is a technology demonstrator for the Oxford Nanopore Research Algorithms group

This software is subject to the terms of the Mozilla Public License, v. 2.0. If a copy of the MPL was not distributed with this file, You can obtain one at https://mozilla.org/MPL/2.0/.

(c) 2017--2019 Oxford Nanopore Technologies Ltd.

The vectorised math functions src/sse_mathfun.h are from https://gruntthepeon.free.fr/ssemath/ and the original version of this file is under the 'zlib' licence. See the top of src/sse_mathfun.h for details.

The Fasta / Fastq input library 'src/kseq.h' is from https://github.com/attractivechaos/klib/blob/master/kseq.h and the original version of the file is under the 'MIT' licence. See the top of src/kseq.h for details.

This project began life as a proof (bet) that a base caller could be written from scratch in a low level language in under 8 hours. Some of the poor and just plain odd design decisions, along with the lack of documentation, are a result of its inception. In keeping with ONT's fish naming policy, the project was originally called Crappie (genus Pomoxis).

Scrappie's purpose is to demonstrate the next generation of base calling and, as such, may change drastically between releases and breaks backwards compatibility. A new version may not support older features.

Release history

The intention is that behaviour will be stable within a series, with only bug fixes or minor improvements being applied. An improvement or change in behaviour that is not a major shift in the algorithm will be a new series with a bump of the minor version number. Any major changes in the algorithm will be a new series with the major number bumped.

  • 1.4 series: Basecalls for better consensus
    • release-1.4.2
      • RNA squiggle model
    • release-1.4.1
      • Temperature corrected R10 model (rgrgr_r10)
      • New R9.4 squiggle model with improved dwell correction
      • Allow read translocation rate to vary in signal-to-squiggle mapping
    • release-1.4.0
      • Direct correction for homopolymers
      • Temperature adjustment for improved accuracy (rgrgr_r94 and rgrgr_r941 models only).
      • Module to detect event boundaries without calling.
      • Remove some old models
  • 1.3 series: Recurrent Neural Network - Random Fields, Python, and Alignment
    • release-1.3.3
      • Alignment of signals to sequences via RNN label scores.
    • release-1.3.2
      • Support older platforms that don't support C11 memalign functions (e.g. Mac)
    • release-1.3.1
      • Scrappie-mappy, map signal to predicted squiggle.
      • Python interface for mapping and basecalling.
    • release-1.3.0
      • Initial support for RNN-RF models for raw data calling (R9.4 only).
  • 1.2 series: Sequence-to-squiggle
    • release-1.2.0
      • Ability to predict expected squiggle from sequence.
  • 1.1 series: Improved basecalling from raw signal
    • release-1.1.1
      • Improved rgrgr models.
      • Ability to write output to named file.
      • Rename argument '--outformat' to '--format'.
    • release-1.1.0
      • Implementation of raw data 'pirate' networks (rGr and rgrgr).
      • Stand-alone event detection.
      • Stay penalty to allow ratio of insertions to deletions to be changed.
      • Local (Smith-Waterman like) penalty on edges of basecall to remove noisy bases.
      • Improvements to error reporting.
  • 1.0 series: Public release
    • release-1.0.1
      • Integration with Travis-ci.
      • Rudimentaty unittests.
    • release-1.0.0
      • Change licence to MPL.
    • Increased number for assertions.
    • A few stability and bug fixes.
  • 0.3 series: Basecalling from raw signal
    • release-0.3.2 Minor fixes to support use of OpenBLAS and HDF5 libraries in non-standard locations.
    • release-0.3.1 Expose options for segmenting raw signal.
    • release-0.3.0 Initial release implementing calling from raw signal.
  • 0.2 series: Post-hoc correction of homopolymer lengths from dwells.
    • release-0.2.8
      • Updated model.
    • Use Kahan summation during normalisation.
    • Build scripts don't require presence of C++ compiler.
    • release-0.2.7
      • Allow segmentation to be taken from different analysis to event detection.
      • Basic support for recalling files produced by Albacore.
    • release-0.2.6 Support for building on OSX
    • release-0.2.5 Allow output to be in simple SAM format
    • release-0.2.4 Add document describing release history
    • release-0.2.3 Bugfix: Segfault when there is insufficent steps to calibrate dwell scaling factor.
    • release-0.2.2 Remove redundant Python scripts for dwell correct.
    • release-0.2.1 Make dwell correction the default
    • release-0.2.0 Initial release
  • 0.1 series: Refactoring
    • release-0.1.1 Allow location of segmentation to be specified on commandline.
    • release-0.1.0 Reorganisation + move to CMake.
  • 0.0 series: Transducers
    • release-0.0.3 Output more information about each read + internal changes to memory management.
    • release-0.0.2 Initial release of Scrappie.