Releases: ThomasDOtto/ILRA
Releases · ThomasDOtto/ILRA
v1.5.2
- Updated some steps of installation (e.g., conda environments now within external_software directory, and conda cleaning of cache to avoid excessive number of files)
- iCORN2 now compressing the intermediate bam files (for now level 9, WIP in the future to make this customizable by the user)
- iCORN2 with a new mode "low_space" (for now the uncompressing of raw reads can occur in /dev/shm, WIP to keep there more intermediate files)
v1.5.1
v1.5.0
v1.4.2
- When processing large assemblies/contigs, MegaBLAST execution in step 2 may be too computationally expensive. Made the parallel processing more robust and efficient (so the largest contigs are processed first and would fail first but not impede the processing of the rest).
- New parameter to let the user decide whether to parallelize and how the processing of the contigs that failed
- Captured the situation and exit ILRA when contigs are just too large for the current RAM available
v1.4.1
- Hotfix in 'path_to_source' so the correct R version is used when required (i.e. including ggplot2)
- Minor fixes and typos in the description of arguments by 'ILRA.sh -h'. Added default values, tidied and reorganized some
- Added a counter so the users could have time to abort an ILRA run and avoid removal when the provided output folder contains previous results
v1.4.0
- Bug fix in 'ILRA.removesmalls.pl' when filtering contigs based on length when the length of a contig was exactly the chosen threshold
- Updates in the main script to use the new script 'ILRA.removesmalls.R' to process the results of the filtering based on a threshold of contig length (i.e. new figures to show the distribution of the lengths pre- and post- filtering and aligning back the discarded sequences to the filtered assembly to assess or confirm their redundancy)
v1.3.1
v1.3.0
- New parameter for specifying block size for megaBLAST parallel processing
- New parameters to allow input by the user for the thresholds to control contig merging based on overlapping, identity (megaBLAST), coverage of Illumina short reads...
- New default threshold for range of coverage of Illumina short reads at overlapping of contigs (0.5+0.1)
- Made more robust the computation of mean Illumina coverage in the script 'ILRA.findoverlaps_ver3.pl'
- Added log2 visualization of the distribution of contig lengths for extreme cases in the script 'ILRA.removesmalls.pl'