Releases: DennisSchmitz/Jovian_archive
1.2.07
1.2.06
The various workflows now request certain memory resources. This is mostly required for grid-mode.
This should not interfere with local mode as snakemake automatically dials resources down to availability on a local system.
We currently use fixed memory resource requirements. However, we're still exploring the option of flexible memory resources based on input files.
1.2.05
- Fixes a bowtie2 dependency error which was introduced a few days ago (version pinning of
tbb
package) - Introduces the improved method of removing primers from Illumina reads (reference based workflow)
1.2.04
This is a minor (hot)fix for outputting reads in the RemovePrimers script.
This change causes the scripts to not output empty reads as this can cause problems in downstream processing steps
1.2.03
This is a minor fix for the nanopore-reference workflow.
During read-cleaning and primer removal it (was) possible that one or more empty reads ended up in the fastq fastq files causing problems with downstream processes.
This release removes empty reads from the fastq files.
Release 1.2.02
This release introduces a new method of removing primers from sequencing reads.
This new method is able to accurately, and relatively quickly, match reads to its corresponding primers which were used and remove said primers from Nanopore reads where necessary.
We consider this method stable for the Nanopore reference workflow.
Release 1.2.01
This update contains minor improvements to the nanopore reference workflow.
The nanopore reference workflow is now able to handle genome insertions in the consensus sequence.
Also included is an update that aims to improve the installation speed of the Jovian_master
conda environment.
This is especially useful when jovian is running completely automated and the Jovian_master
environment has te be built many times.
By using mamba instead of conda (for installing the main dependencies) the installation time is brought back significantly
Release of Jovian 1.2.0
This update includes exclusively changes to the Nanopore reference alignment workflow.
Some notable changes are as follows:
-
Significant speed improvements. The Jovian 1.2.0 nanopore workflow brings the average runtime down to about 20 minutes per sample whereas the previous version required an average 3 hours per sample
-
Improved cleaning of adapters. Jovian 1.2.0 no longer relies on PoreChop in order to remove adapters from the reads.
-
Improved removal of primer sequences. The processes of removing adapters and removing primer sequences are more finetuned to clean reads more precisely and more consistently.
-
Inclusion of "safety nets" in order to properly parse the primer sequences given by the user
-
A new, custom, consensus caller that is able to correct for sequencing and alignment errors up to 1 codon within an ORF.
Release 1.1.0
Release of Jovian version 1.1.0
Jovian 1.0.x
to Jovian 1.1.0
then please execute bash jovian --update
twice
This release includes the following changes and improvements:
General improvements:
- The Jovian workflow(s) have had a significant make-over, improving speed, structure, and code-readability.
As an added bonus: all workflow(s) now follow the same code style and structure for improved clarity - All python code now follows the same formatting
- "Redesigned" various functions within Jovian for improved speed and user-feedback
- Various stability changes to the "Jovian_master" and "Jovian_helper" environments
- Fixed a bug in the Jupyter notebooks configuration
- Fixed several bugs in the database installer
- Improvements have been made to the database installation process
- The Notebook Report should now display all your run information automatically.
Introducing multiple workflows in Jovian
Jovian used to be exclusively a viromics/metagenomics workflow. However, Jovian now supports multiple workflows that the user can choose to execute.
You're able to choose from the following workflows:
- Illumina Metagenomics
- Illumina Reference alignment (new 🎉)
- Nanopore Reference alignment (new 🎉)
Illumina reference alignment:
- Improved consensus calling, added indel calling.
- Visualized minority SNPs in
IGVjs
for manual inspection. - Fixed a bug with the rendering of the GC-content track in
IGVjs
. - Fixed a bug, human data is now correctly removed.
- When new majority SNPs are found after the consensus sequence is generated, a warning is thrown. Manually assess the alignment in
IGVjs
and determine if a another round of alignment is required. This is a work in progress. - Code cleanup, work in progress.
- Importantly, please double check the consensus genomes generated with coverage low coverage (<10x) manually in the provided IGVjs viewer to assess errors: mistakes are observed at these coverages.
Nanopore reference alignment:
A new workflow designed for processing and aligning Nanopore (multiplex pcr sequencing) data to a reference.
- Removes adapter sequences
- Removes overrepresented primer sequences
- Removes background host information (human genome)
- Performs Quality Control specific for nanopore data
- Aligns the data to a user-given fasta
- Creates a consensus genome with various depth-of-coverage thresholds [1x, 5x, 10x, 30x, 100x]
- Produces information about the "Breadth-of-coverage" (percentage of the reference which could be aligned) for each coverage threshold
- Detects majority SNPs and displays this information in an easy-to-read table
Documentation of Jovian can now be found on jovian.rivm-bioinformatics.com
Release 1.0.01
This release includes the following changes:
- Some minor textual changes
- A permissions fix for some of the supporting components of Jovian
- Addition of the GNU BC (basic calculator) package to the Jovian_master environment.
- A bugfix for the
--rebuild-archive
flag - Changes to the Jovian updater (
--update
)
No changes have been made to the core analysis workflow(s)
When updating from 1.0.0
or lower to 1.0.01
or higher, please execute bash jovian --update
twice