v1.2.0

This release of KOunt contains the following new features/changes:

• Removal of Barrnap and tRNAscan-SE from the pipeline, all data that would've been annotated by them is run through the rest of the pipeline to save time, flowchart amended to reflect this
• Inclusion of the new quality filtering step, evenness of protein coverage. All of the coverage and kofamscan_results rules have been updated to give users the option of using/no using this
• Replacement of the bedtools_samtools_KOunt.yaml with coverage_KOunt.yaml
• Fixing bug in kallisto KOunt calculation
• Amending bed file creation in the nohit_fastq rules
• Added commands to remove large unwanted output files
• Changed the coverage rules to convert bam files to bed files which reduces memory needed
• Updated the template config.yaml
• Added a file with the average memory and time used for the samples in our manuscript

v1.1.0

This release of KOunt contains updated conda environmental yamls to allow installation with mamba; this is recommended as it is much faster. Links to subsampled rumen metagenome reads (ERR2027889) are also included to enable quick testing of whether installation has been successful. The expected location of the input to KOunt has been changed, now it expects each sample to be in a folder with the sample name; it will now use all samples within the raw_reads directory specified in config.yaml.

The following bug fixes have also been made:

• Edited the recommended snakemake command to include the number of cores
• Removed profiles.tar.gz once gunzipped
• Removed the all_kos_found file once complete
• Removed v0.24.tar.gz once gunzipped
• Updated rule kofamscan to have the output of rule kegg_db as input
• Added commands to shorten the trimmed fastq read ids and add /1 or /2 to the end of read ids that don't already end with that
• Added fastq_utils to rule trim to check read names are unique - user has the option to check this or not
• Updated envs/fastp_utils_KOunt.yaml to include the software fastq_utils
• Added option to config.yaml to check read names are unique
• Amended rule kallisto_unmapped to use mkdir -p in case a barrnap directory hadn't been made
• Amended scripts/rrna_trna_cov.sh to avoid segmentation fault if no RNAs are found
• Fixed errors in rule rna_abundance_noclustering not finding output files
• To make rule kofamscan_results_no_clustering independant of rule kegg_fastas the following new rules were created:
  • barrnap_noclustering
  • diamond_search_noclustering
  • kallisto_noclustering
  • nohit_annotate_reads_noclustering
  • nohit_fastas_noclustering
  • nohit_fastq_noclustering
  • rna_abundance_noclustering
  • trnascan_noclustering
• Replaced params.miss with output.miss in rule rna_abundance and rule rna_abundance_no_clustering
• Removed -c from the kallisto commands
• Amended rule kallisto, rule kallisto_unmapped and rule kallisto_noclustering to filter out KOs with zero abundance
• Added LC_ALL=C before the sort commands in rule coverage to prevent a bedtools error with the input not being sorted lexicographically
• Amended scripts/rrna_trna_cov.sh to remove unneeded output files
• Amended scripts/unmapped_reads_ko_100.sh to remove unneeded output files
• Added commands to delete raw kallisto output and other unneeded output files
• Changed output of files going into mmseqs to MMSeq_Outputs
• Removed the nohit file output from the rule kofamscan_results_noref