Recent Releases of rosella
rosella - v0.5.0
What's Changed
- v0.5.0-prerelease by @rhysnewell in https://github.com/rhysnewell/rosella/pull/43
- v0.5.0 by @rhysnewell in https://github.com/rhysnewell/rosella/pull/44
- feat(dev): update coverm engine and docs creations add workflows by @rhysnewell in https://github.com/rhysnewell/rosella/pull/45
Full Changelog: https://github.com/rhysnewell/rosella/compare/v0.4.2...v0.5.0
- Rust
Published by github-actions[bot] over 2 years ago
rosella - v0.4.1
Commits
- update version to try and bump bioconda build #20 (Rhys Newell)
- Rust
Published by rhysnewell over 4 years ago
rosella - v0.3.0
Major version update, 0.3.0
Iterative binning now implemented and bugs are fixed. - Clusters are now internally checked for quality. Bins that failed are then iteratively re-embedded using UMAP and re-clustered with HDBSCAN - Plotting added for iterative binning. Watch your contigs jump from cluster to cluster. Fun but useless. - Removed correlation contig reassignment as it just creates noise
- Rust
Published by rhysnewell over 5 years ago
rosella - v0.2.0
This release brings a variety bug fixes, changes in default parameters, and better utilization of the UMAP algorithm.
CHANGES: - Rosella now handles situations with 2 or less samples differently. Opting to perform a single UMAP projection - When 3 or more sample are present then an intersection of three UMAP projections is used. These projections include a metabat distance projection on the contig coverage and variance, TNF correlation projection, and CLR coverage correlation projection. Discussions regarding combining and intersecting UMAP projections can be found here: https://github.com/lmcinnes/umap/issues/206 - MetaBat 1 and 2 provide an awesome method for calculating the distance between normal distributions which has been included into rosella as distance metric. This method seems to work much better than using kullbacl-leibler divergence - Only canonical tetranucleotide frequencies are used now. - Should be faster in most cases due to less reliance on custom metrics for TNF UMAP and correlation UMAP
- Rust
Published by rhysnewell over 5 years ago
rosella - v0.1.0
Rosella
Rosella is a metagenome assembled genome binning algorithm leveraging the amazing UMAP and HDBSCAN algorithms. Rosella is easy to use with a front end developed in rust. It can handle all of the mapping and coverage calculations if there aren't already coverage results available.
Compared to other binners, rosella performs about as well as metabat2 but in some situations can outperform ensemble binning algorithms like DASTool. See the below figure for the results of running rosella on the CAMI High complexity challenge dataset. DASTool was run using every listed binning algorithm EXCEPT for rosella. DASTool would only improve with the results of rosella included. Ensemble binning is still the recommended approach to generating a finalized set of high quality MAGs.
- Rust
Published by rhysnewell over 5 years ago
rosella - v0.0.1
This is the first stable release of Rosella, an experimental metagenomic binning algorithm that utlizes tetranucleotide frequencies and coverage values passed to UMAP and HDBSCAN to recover metagenome assembled genomes.
Currently, Rosella performs fairly well when provided multiple samples. There seems to be some difficulty with very low complexity samples (n < 2) and highly complex metagenomes with low numbers of samples.
- Rust
Published by rhysnewell over 5 years ago