Humann

Latest version: v3.9

Safety actively analyzes 682416 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 7 of 9

0.7.1

New Features

* A new option, "--search-mode <uniref50/uniref90>", was added. The default is set based on the translated database selected. This option sets the percent identity, chocophlan annotation index, and the translated search parameters.
* A new tool, humann2_gene_families_genus_level, and tutorial were added to create genus level gene families and pathways files.
* A new tool was added, humann2_split_stratified_table, to split output files into stratified and unstratified tables.

0.7.0

New Features

* A new translated search filtering feature has been added. This by default filters alignments with less than 90% query coverage. This option is named "--translated-query-coverage-threshold <90.0>". The previous coverage option "--coverage-threshold <40.0>" has been renamed to "--translated-subject-coverage-threshold <50.0>" for clarity with the default increased from 40% to 50%.
* New options have been added to the humann2_rename_table script. An option to switch between community and levelwise normalization was added with the default set to community normalization. Special features like UNMAPPED categories can now be excluded when normalizing. Also the flag "--norm" has been renamed "--units" to clarify this option only changes the units reported not the style of normalization.
* Four new translated search diamond formatted databases are now available. Two of the databases contain the full set of protein sequences (one for UniRef50 and one for UniRef90). The other databases are filtered to only include those proteins with ECs of level 4 or that are included in a MetaCyc pathway. The full databases allow users to identify unclassified proteins in their data set while the filtered databases require less memory and run time.

0.6.2

New Features

* HUMAnN is now pip installable. The documentation has been updated to reflect this new feature.
* New options were added, to add names to kegg pathways and modules, to the humann2_rename_table script.
* A new tool was added to build a custom diamond database with taxonomic limitation. Adding this tool expands the Kegg legacy flow tutorial to start with quality controlled metagenome files and to include MetaPhlAn2 output.

0.6.1

New Features

* The UniRef50_unknown and UniRef90_unknown values are now included in the gene families file.
* For information on these values, please see the documentation section on the gene families file.
* More functional tests have been added
* 12 functional end-to-end humann2 tests were added which run in about 19 minutes
* 21 functional tool tests were added
* The tool humann2_join_tables now has the option to search sub-directories of the input folder for files to join.

Bug Fixes

* If "--remove-stratified-output" is selected, the stratified output for the UNINTEGRATED pathways values are no longer printed
* The name of the UniRef50 to RXN mapping file, for script humann2_regroup_table, was updated
* This resolves the error seen when running humann2_regroup_table with option "--groups uniref50_rxn"
* The step to build the fasta file for the custom ChocoPhlAn database has been split into multiple steps
* This resolves the max arguments error seen when running on a MacOS with "--bypass-prescreen"

0.6.0

New Features

* To reduce the influence of spurious hits in the translated search step, raw translated search results are now subjected to an initial protein coverage filter (default >50%). A read's weight is only divided over proteins that meet this coverage threshold, which (in addition to removing spurious hits) has the added benefit of substantially reducing the size of the genefamilies.tsv output file. To change the default coverage threshold, use the option "--coverage-threshold <50.0>".
* Unmapped reads are now included in the gene families abundance output file as a new "UNMAPPED" feature. This value represents the total unmapped reads after both searches, nucleotide and translated. For more information on this computation, see the documentation section about the gene families abundance file.
* Unmapped read abundance is similarly carried through to the pathway abundance file, along with a new stratified feature, "UNINTEGRATED," which reflects the total abundance of genes that did not contribute to a metabolic pathway. These features are included in the pathway coverage file to maintain pathway abundance/coverage concordance (see below). For information on how these values are calculated, see the documentation section for each file.

Other Changes

* The included mapping from level-4 EC groups to UniRef50 clusters has been expanded based on annotations from TrEMBL. ~10x more UniRef50s can now be mapped to a level-4 EC group.
* The regroup_table script has been modified to include an "UNGROUPED" group, which captures the abundance of features that failed to map to another, non-trivial group. The "UNMAPPED" feature (see above) will always carry through to a regrouped table.
* The format of the pathway abundance and coverage output files has been updated so both files include the same order of pathways and species. The pathways are ordered by decreasing abundance. Pathways with zero abundance are not included.

0.5.0

New Features

* Added default read length alignment normalization
* This replaces the optional read length normalization using "--average-read-length L"
* Added new script humann2_merge_abundance_tables
* This script merges the tables to stratify gene families with respect to pathways

Other Changes

* Updated gap fill method to fill single lowest score
* Join script updated to allow for input files containing multiple samples
* Required Diamond version updated to v0.7.9+

Page 7 of 9

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.