-
Couldn't load subscription status.
- Fork 4
DETONATE: DE novo TranscriptOme rNa-seq Assembly with or without the Truth Evaluation
Couldn't load subscription status.
deweylab/detonate
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
*** This file is autogenerated. Don't edit it directly. ***
DETONATE: DE novo TranscriptOme rNa-seq Assembly with or without the Truth
Evaluation
Overview
DETONATE (DE novo TranscriptOme rNa-seq Assembly with or without the
Truth Evaluation) consists of two component packages, RSEM-EVAL and
REF-EVAL. Both packages are mainly intended to be used to evaluate de
novo transcriptome assemblies, although REF-EVAL can be used to
compare sets of any kinds of genomic sequences.
RSEM-EVAL is a reference-free evaluation method based on a novel
probabilistic model that depends only on an assembly and the RNA-Seq
reads used for its construction. Unlike N50, RSEM-EVAL combines
multiple factors, including the compactness of an assembly and the
support of the assembly from the RNA-Seq data, into a single,
statistically-principled evaluation score. This score can be used to
select a best assembler, optimize an assembler's parameters, and
guide new assembler design as an objective function. In addition, for
each contig within an assembly, RSEM-EVAL provides a score that
assesses how well that contig is supported by the RNA-Seq data and
can be used to filter unnecessary contigs.
REF-EVAL is a toolkit of reference-based measures, including contig,
nucleotide, and pair precision, recall, and F1 scores, a novel kmer
compression score, and several scores that compare induced kmer
distributions between the assembly and the reference. REF-EVAL also
includes a program to estimate the "true" assembly of a set of
RNA-Seq reads, relative to a collection of full-length reference
transcripts. See [1]here, or ref-eval/README in the distribution, for
detailed information.
DETONATE is motivated and described in detail in the following paper:
Bo Li*, Nathanael Fillmore*, Yongsheng Bai, Mike Collins, James A.
Thomson, Ron Stewart, and Colin N. Dewey. Evaluation of de novo
transcriptome assemblies from RNA-Seq data. Genome Biology 2014,
15:553. [2]link
* = equal contributions
Downloading
The current version of DETONATE (1.11) is available here:
* [3]detonate-1.11.tar.gz
As a convenience, we also provide a precompiled version, for Linux.
This precompiled version may or may not work for you; if it doesn't,
please try to compile the version above, instead.
* [4]detonate-1.11-precompiled.tar.gz
Building/installation
To build RSEM-EVAL and REF-EVAL, simply type "make" in the top-level
detonate-1.11 directory, after you've upacked the tarball (tar xvf
detonate-1.11.tar.gz). If you have trouble with any of this, please
[5]contact us.
Vignette
A vignette with an extended example of the basic usage of DETONATE is
available [6]here, or in VIGNETTE in the distribution.
Usage
For information about using RSEM-EVAL, see [7]here, or
rsem-eval/README.md in the distribution.
For information about using REF-EVAL, see [8]here, or ref-eval/README
in the distribution.
For information about using REF-EVAL-ESTIMATE-TRUE-ASSEMBLY, see
[9]here, or ref-eval/README.REF-EVAL-ESTIMATE-TRUE-ASSEMBLY in the
distribution.
Authors
RSEM-EVAL was coded by Bo Li, and REF-EVAL was coded by Nathanael
Fillmore. Bo, Nate, and Colin Dewey jointly developed the RSEM-EVAL
and REF-EVAL methodology, with feedback on the methodology from the
other coauthors.
Mailing list
Subscribe to DETONATE Users
Email: [10]_____________________ [11][ Subscribe ]
[12]Visit this group
Development
A repository for the DETONATE source code is available at Github,
[13]here.
Axolotl assembly
The RSEM-EVAL-guided Axolotl assembly described in the DETONATE paper
(see above) is available [14]here.
Changelog
In [15]DETONATE 1.11 (the current version), released on Jun 22, 2016,
the following changes were made:
* Reduced memory usage in RSEM-EVAL.
* Changed build process to use the systemwide version of CMake,
unless DETONATE is compiled with "NEED_CMAKE=yes make".
* Restored missing "examples" directory in released version.
In [16]DETONATE 1.10, released on Sep 1, 2015, the following changes
were made:
* Added an option to store single-precision numbers in the hash
table used for computing KC and kmer scores, so as to use less
memory.
In [17]DETONATE 1.9, released on Mar 25, 2015, the following changes
were made:
* Canceled f function values from "Prior_score_on_contig_lengths"
and "Correction_term".
* Changed short to int to handle longer inferred insert sizes.
In [18]DETONATE 1.8.1, released on Oct 2, 2014, the following changes
were made:
* Updated the toy example to match that in the paper.
In [19]DETONATE 1.8, released on Sep 26, 2014, and comprised of
RSEM-EVAL 1.8 and REF-EVAL 1.8, the following changes were made:
* Added support for paired-end data.
* Improved the interface.
In [20]DETONATE 20140123, comprised of RSEM-EVAL 1.6 and REF-EVAL
20140123, the following changes were made:
* Fixed UI bug in REF-EVAL related to --alignment-policy.
* Incorporated changes from RSEM 1.2.9 into RSEM-EVAL.
[21]DETONATE 20140108, comprised of RSEM-EVAL 1.5 and REF-EVAL
20140108 was the initial release.
References
Visible links
1. http://deweylab.biostat.wisc.edu/detonate/ref-eval.html
2. http://genomebiology.com/2014/15/12/553
3. http://deweylab.biostat.wisc.edu/detonate/detonate-1.11.tar.gz
4. http://deweylab.biostat.wisc.edu/detonate/detonate-1.11-precompiled.tar.gz
5. https://groups.google.com/d/forum/detonate-users
6. http://deweylab.biostat.wisc.edu/detonate/vignette.html
7. http://deweylab.biostat.wisc.edu/detonate/rsem-eval.html
8. http://deweylab.biostat.wisc.edu/detonate/ref-eval.html
9. http://deweylab.biostat.wisc.edu/detonate/ref-eval-estimate-true-assembly.html
12. http://groups.google.com/group/detonate-users
13. https://github.com/deweylab/detonate
14. http://deweylab.biostat.wisc.edu/detonate/AxoJuveBlastema_RSEM-EVAL-1.4_Trinity.fasta.gz
15. http://deweylab.biostat.wisc.edu/detonate/detonate-1.11.tar.gz
16. http://deweylab.biostat.wisc.edu/detonate/detonate-1.10.tar.gz
17. http://deweylab.biostat.wisc.edu/detonate/detonate-1.9.tar.gz
18. http://deweylab.biostat.wisc.edu/detonate/detonate-1.8.1.tar.gz
19. http://deweylab.biostat.wisc.edu/detonate/detonate-1.8.tar.gz
20. http://deweylab.biostat.wisc.edu/detonate/detonate-20140123.tar.gz
21. http://deweylab.biostat.wisc.edu/detonate/detonate-20140108.tar.gz
About
DETONATE: DE novo TranscriptOme rNa-seq Assembly with or without the Truth Evaluation
Resources
Stars
Watchers
Forks
Packages 0
No packages published