Version 1.5.0 - Hashing, Antimicrobial Resistance Screening, MST Clustering and more

Overview

• Hashed cgMLST
• Clustering functionality and interface for minimum-spanning-trees
• Antimicrobial Resistance Screening with NCBI/AMRFinder
• Robustified allelic typing
• Improvements in user interface
• Option to save assembly files to local database
• Various bugfixes
• and more ...

To install this update follow the instructions given in README - 2 Installation.

Release 1.5.0 contains substantial improvements, novelties, fixes and other changes. Most notable is the introduction of hashed cgMLST. Allele sequences are now hashed by default using SHA-256. Upon download of a cgMLST scheme, all alleles respectively belonging to the loci included in the scheme are hashed, with the resulting unique 64-bit words substituting numeric sequence indexes. This change allows efficient and robust comparison of results from collaborating working groups. PhyloTrace is aiming to become universally compatible - a goal only achievable using hashing technology. This fundament is now integrated and will be enriched with more utility in upcoming releases.

Another milestone is the introduction of Antimicrobial Resistance Screening with AMRFinder/NCBI. The new "Gene Screening" interface allows to query the isolates present in the local database for resistance, virulence, stress genes and more. For now, the results are saved to the database and can be inspected and exported. In future users will be able to visualize this information.

Minimum-spanning tree (MST) network graphs were supplemented with a cluster feature - an ability crucial for isolate comparison. Users can choose between "Area" and "Skeleton" cluster types, both visualizing isolates clustering with each other according to a customizable threshold. Next to these major changes, version 1.5.0 contains numerous improvements regarding user experience, interface and robustness.

Note, that due to these significant changes, PhyloTrace 1.5.0 is not backwards-compatible with previous versions. Local databases need to be completely reconstructed.

Version 1.4.1 - Locus Sequence Browser, Report Functionality, Windows Compatibility and more

The minor release 1.4.1 contains various UI and UX improvements, changes and fixes. Users can now install PhyloTrace on the Microsoft Windows Subsystem for Linux (WSL). The instructions were added to the README. Besides that, loci can now be viewed in a separate tab allowing inspection of individual variant sequences and locus metadata, e.g. respective gene products. Moreover the report generation interface has been revised. Overall this update adds robustness and refinements augmenting the experience.

• PhyloTrace installable via WSL on Microsoft Windows
• UI improvements (Reactivity, Unifications)
• Fixes for known minor bugs
• Source file renaming
• Local logging of user actions, warnings and issues
• Report creation interface optimized
• Dependencies updated
• Locus metadata and sequence viewer in new tab

Version 1.4.0 - Faster Typing & MST Extension

Besides various improvements, changes and fixes, the release 1.4.0 introduces parallelized algorithms speeding up the typing process by a factor of 4 to 5. Moreover, minimum-spanning trees now feature the option to map variables to the node color. This way, epidemiologic metadata such as country/city of sample acquisition, isolation date or any other custom variable can enrich the analysis with valuable information. In case identical isolates carry different values for the selected variable, the nodes transition to pie charts showing the respective share of several values for the identical group (see attached image). A new scheme for Klebsiella oxytoca sensu lato was added to the cgMLST.org database and made available in this version of PhyloTrace.

• Accelerated typing through parallelization
• Minimum-spanning tree functionality extended by variable mapping feature (pie charts as nodes)
• Minor UI changes
• Miscellaneous bugfixes
• New scheme for Klebsiella oxytoca sensu lato

Version 1.3.0 - Detection of New Allele Variants

Besides various improvements and changes, the minor release 1.3.0 introduces the crucial ability to detect and add new allele variants to the local scheme. The variant calling is now facilitated by the BLAT algorithm (BLAST-like Alignmet Tool) for whole genome assemblies. The previously used KMA (k-mer Alignment) algorithm will be deferred and reintroduced soon to allow allelic typing of raw reads. If none of the variants from the local cgMLST scheme is present in the bacterial isolate a potential new gene variant is evaluated by applying a robust variant validation program.

• Deferred KMA algorithm
• Introduced BLAT algorithm
• Revised allelic typing process
• Improved user feedback on typing results
• Introduced variant detection and validation logic
• Minor UI changes

Version 1.2.0 - Intelligent Tree Visualization

For the new minor release 1.2.0 the visualization section was enhanced in functionality and graphical interface. It is difficult to average suitable plot parameters for a diversity of isolates having different qualities and quantities. Mapping epidemiological meta data as variables, e.g. in the form of colors or a heatmap, adds another layer of complexity. Hence the user was required to adjust the elements of their individual plot to receive a balanced appearance. In the new release, the complexity of the visualization control panels was drastically reduced while the previously necessary manual adjustments are minimized due to intelligent and automated creation of plots.

• Simplified and beautified control interface for NJ and UPGMA trees
• Variable mapping system revolutionized
• Automatic adaption of plot elements and control inputs according to quality and quantity of isolates included in the graph
• Several new functions, e.g. selection of customized color scales for mapped variables

Version 1.1.1

Several changes and corrections have been made in patch v1.1.1. In addition, the loop that checks and appends the results of the KMA algorithm has been benchmarked and accelerated.

• Revision of multi typing status & feedback functionality to enable more robust and less error-prone typing
• Changes to code formatting and variable naming for better readability
• Fixed new bugs related to the custom database selection function
• Accelerated function which evaluates KMA algorithm results (automatic_typing.R & single_typing.R)
• Minor UI changes
• Few less relevant changes

Version 1.1.0 - Custom local database path selection

In version 1.0.0 the local database was statically located inside the PhyloTrace directory. In version 1.1.0 the user can now select any custom path leading to a local database. This allows for more flexibility and ease of use and also improves interoperability.

• Selected database must be compatible with PhyloTrace (otherwise error thrown)
• Create new PhyloTrace-compatible database from scratch in a custom directory
• Load last used database as time-saving shortcut
• Pending multi typing is recognized and allows to load last database with respective scheme only