CHANGELOG.md

Daily Change Log:

• [2026.7.16] - Added parallelization support for pathway coverage using profile_pathway_coverage from kegg_pathway_profiler>=2026.7.16 in leviathan-profile-pathway.py. The --n_jobs flag now applies to both salmon quant and pathway coverage computation.
• [2026.7.16] - Added --deterministic flag to leviathan-profile-pathway.py which passes --deterministic to salmon quant for byte-identical results across runs and thread counts.
• [2026.7.8] - Added --update_salmon_index to leviathan-index.py and remove -u alias on --update_with_genomes
• [2026.7.8] - BREAKING CHANGE: Salmon is reimplemented with Rust from C++ so any Leviathan databases using Salmon 1.x will need to be updated. Leviathan now requires Salmon ≥ 2.x.
• [2026.3.11] - Added support for -r/--single_reads in both profile-taxonomy.py and profile-pathway.py. However, the profile-pathway.py needs oarfish to use long reads but there are no synthetic long-read metatranscriptomics to benchmark. issue/#24
• [2026.3.10] - Added --veba_major_version to compile-manifest-from-veba.py since cluster output directory is changing
• [2026.3.3] - Added step_coverage output files issue/#22
• [2025.12.17] - Added --table_format parquet|tsv to leviathan-merge.py and merge functions in leviathan-profile-pathway.py/leviathan-profile-taxonomy.py Issue #20
• [2025.12.2] - Updated xarray concatenation to prepare for usage change in future versions issue/#17
• [2025.12.2] - Fixed issue/#15 where parquet files weren't being created when user is not using genome clusters (i.e., 4 column input). Related to issue/#18 where contains_genome_cluster_mapping was set to True even when genome clusters were not provided.
• [2025.12.1] - Fixed issue/#16 where feature_prevalence and feature_prevalence-binary output the same values. Needed to provide df_gene_abundance instead of df_feature_abundance
• [2025.7.10] - Updated MIT to Apache 2.0
• [2025.7.3] - Remove genomes that do not have features in leviathan-preprocess.py and write excluded genomes to genomes_excluded.list. Also checks in leviathan-index.py to make sure genomes in feature_mapping.tsv.gz matches those in genomes.tsv.gz
• [2025.7.1] - Adding gzipping as an option to maintain consistent runtime benchmarking and removing aux_info/unmapped_names.txt from Salmon output
• [2025.6.30] - Gzipped quant.sf from Salmon Issue #12
• [2025.6.27] - Updated --sylph_profile_minimum_number_kmers 20 to --sylph_profile_minimum_number_kmers 50
• [2025.5.18] - Added --salmon_include_mappings to profile-pathway.py and run_salmon_quant
• [2025.5.18] - Added taxonomic_abundance and sequence_abundance levels to taxonomic_abundances.*.nc files from leviathan-merge.py
• [2025.5.9] - Added sequence_abundance.genomes.parquet and sequence_abundance.genome_clusters.parquet to leviathan-profile-taxonomy.py
• [2025.3.13] - Added -f/--output_format to profile-taxonomy.py and profile-pathway.py modules with parquet as default but continued support for .tsv
• [2025.3.13] - Dropped gene_abundance merging because files are too big, sparse, and probably not that useful...
• [2025.3.12] - Added h5netcdf to requirements.txt
• [2025.3.12] - Removed option for output formats in leviathan-merge.py. Taxonomic profiles are pandas stored as parquet and pathway profiles are xarray stored as netcdf files.
• [2025.3.12] - Changed merge_taxonomic_profiling_tables to merge_taxonomic_profiling_tables_as_pandas
• [2025.3.12] - Changed merge_pathway_profiling_tables to merge_pathway_profiling_tables_as_pandas and created merge_pathway_profiling_tables_as_xarray
• [2025.3.12] - Moved merge_pathway_profiling_tables from utils to profile_pathway
• [2025.3.12] - Moved merge_taxonomic_profiling_tables from utils to profile_taxonomy
• [2025.3.11] - Added default to transpose tables in leviathan-merge.py but added --no_transpose_taxonomic_profiles and --no_transpose_pathway_profiles
• [2025.3.6] - Added default --output_format for leviathan-merge.py to be parquet (which will be transposed relative to tsv and pickle)
• [2025.1.21] - Added leviathan-merge module along with adding merge_taxonomic_profiling_tables and merge_pathway_profiling_tables to utils
• [2025.1.18] - Added support for pykofamsearch-reformatted and pyhmmsearch-reformatted in leviathan-preprocess.py and read_annotations
• [2025.1.17] - Port functions from pyexeggutor
• [2025.1.17] - Added compile-manifest-from-veba.py script
• [2024.9.27] - Removed dependency for every genome to require a hit in the feauture mapping table
• [2024.9.23] - Fixed bug where feature_abundances.genomes.tsv.gz had multiple rows per feature.
• [2024.9.23] - Replaced BioPython with the faster and more lightweight PyFastx
• [2024.9.23] - Added pathway support for leviathan index and leviath profile-pathway. The latter, outputs pathway abundance and coverage values if a pathway database is available.
• [2024.9.23] - Added KEGG Pathway Profiler as dependency which is used for building pathway databases if the input features are KEGG orthologs
• [2024.7.31] - Added leviathan profile-pathway but still needs pathway abundance and coverage support.
• [2024.7.30] - Added leviathan profile-taxonomy
• [2024.7.26] - Unversioned pre-alpha release on GitHub which only includes utils and index modules.

Bugs:

• Memory profiler is incorrect