v3.1.9 - Improved results reproducibility

• Fixed impact of contig ordering impacting the final results reproducibility (issue #160)
• Fixed biopython issue with gc_fraction available from version 1.80 onwards (issue #163)

v3.1.8 - Improved biomarkers report consistency

This release includes small but important convenience fixes such as fix to #152 issue where in cases when there are absolutely no plasmid biomarkers (i.e. empty biomarkers report) mob_typer would fail. Additionally, this release improves marker reporting consistency between biomarkers and the main plasmid typing reports generated by mob_recon and mob_typer. The biomarkers report introduced since v3.1.7 lists the main plasmid biomarkers and their coordinates. Finally, plasmid databases were cleaned, synced and redundant files removed. A new databases archive url was updated and now points to https://zenodo.org/records/10304948/files/data.tar.gz?download=1

• Fixed permanently issue #152 and added a unit test test_mob_recon_no_plasmid_biomarkers()
• Updated the core database archive data.tar.gz: a) removed redundant host_range_ncbirefseq_plasmidDB.txt and host_range_ncbirefseq_plasmidDB.csv; b) updated clusters.txt with new mob_typer biomarker results for consistency (the cluster codes and number of records did not change)
• Improved biomarkers report consistency with other reports by making sure that all markers listed in the mob_typer report are present in the biomarkers report. Implemented a new unit tests listed in test_databases_consistency.py

Addition of filtered biomarker blast report

This release includes a fix to #150 where uninitialized databases to mob_typer would result in confusing error messages. Additionally, this release includes the addition of a new biomarkers blast report file which will be automatically generated by mob_recon and optionally by mob_typer if the --biomarker_report_file parameter is included. This file contains the filtered biomarker blast results for oriT, replicon, relaxase, and mate-pair-formation as a single tab-delimited file of blast results with an included column labelling which biomarker type the sequence belongs to.

MOB-cluster bug fixes

This update includes fixes to issues #149 with thanks to @rpetit3 additionally, addresses the generation of mge reports with empty results when called through mob-typer

MOB-cluster bug fixes

Largely resolving issues with MOB-cluster not running due to changes in ETE3 DB parameters

Indentation fix and version bump

Fixed an issue with ETE3 database not being set with default database

MOB-suite v 3.1.2

Fixed numpy dependency due to breaking changes in later versions

MOB-suite: Addition of MGE detection and minor bug fixes v3.1.0

MOB-suite v3.1.0 adds in a reporting of mobile genetic elements (MGE) based on the repetitive masking file comprised of IS/TN elements. It will report the blast HSP results combined with the molecule_type and primary MOB-cluster. This MGE report file will be created by default with MOB-recon and is optional with MOB-typer by specifying a report file.

This release also addresses several different issues:

MOB-recon and MOB-typer will now report out the original contig_id supplied in fasta files instead of converting spaces to "_" #99, #87 #85
MOB-recon has re-implemented the maximum contig size and plasmid size restrictions #101
MOB-recon has removed the -t flag as typer is always invoked #100
MOB-recon and MOB-typer will no longer restrict threads to 1 for tblastn searches #96 #104
Removed ETE3 lock file mechanism #93
Default argument values are now visible when using --help #91

Make MOB-Suite read-only file system friendly

• Fixes issue #82 and issue #63
• Fixes file permissions for plasmid database files by not copying original permissions from data.tar.gz
• Fixes mob_init start on systems with python binary defined other than python thanks to use of sys.executable variable in verify_init()

Fixed broken database mirror

This release is purely to address broken download links for the database