viral-ngs: v1.10.0

New:

• assembly.order_and_orient() now supports --outAlternateContigs to output sequences (in FASTA format) from alternative contigs that mapped, but were not chosen for the final scaffold output
• read_utils.align_and_fix() now supports aligning with bwa as an alternative to Novoalign
• It is now possible to say: easy-deploy-viral-ngs.sh upgrade to upgrade the installed version of viral-ngs
• The easy-deploy-viral-ngs.sh script now exits if the current machine has an insufficient amount of RAM to run viral-ngs (currently allows >1GB). This serves as a helpful reminder to run viral-ngs on compute cluster nodes, and not on login nodes.
• The parameters assembly_replace_length and min_pct_contig_aligned are now exposed in the config.yaml file of the snakemake pipeline

Fixed:

• factor correction for min_pct_id used during scaffolding
• sample name matching is now greedy, so for example, a file including [sample]-1-2 yields [sample]-1 as the sample name
• The reports.rules snakemake rules now uses conda-installed FastQC
• easy-deploy-viral-ngs.sh now pins conda version to avoid upstream instabilities with conda
• Read trimming, de-duplicatation, and subsampling prior to de novo assembly has been reworked to favor paired-end reads, and to add in singletons when the paired read count is insufficient to reach the count desired for de novo assembly. It performs subsampling of reads in bam space via Picard.
• scaffolded sequences are now given IDs that differ from the reference IDs (avoids collision which can be a problem for aligners downstream)
• blank lines are now permitted in the tabular section of SampleSheet.csv files

Updated:

• krona updated to 2.7
• fastqc now included