Release Notes 2019-2021

View descriptions of the updates and fixes in each release from 2021 or before

Releases from 2022 onwards will be grouped by year as Gitbook cannot handle long pages in their new editor.

v12.0.4

21st December 2021

Changed

• The collection view AMR tables now display the antibiotic name and not the code in the column headers.

v12.0.3

13th December 2021

New

• Collections can be created from a set of genomes of any species, though they do still have to be the same species. Trees of the collected genomes are still only generated for the same restricted set of organisms.

• Two new filters have been added to the Genomes List page:

  • The list of species and number of genomes with cgMLST assignments.

  • The list of species with collections and the number of collections.

• A new "fair share" algorithm has been implemented for managing task priorities. This should guarantee quick access to our resources for users who are uploading small numbers of genomes.

v11.1.0

3rd December 2021

Updated

• Pangolin updated to PangoLEARN 2021-1125

v11.0.0

30th November 2021

This release marks the addition of enhanced support for Vibrio cholerae. We expect to significantly expand the number of public genomes in the near future, and invite feedback on all aspects.

Added (for V. cholerae)

• SNP trees and collections

• AMR predictions

• Virulence and pathogenicity islands (using our in-house tool Vista)
• MLST, cgMLS and cgMLST-based clusters.

v10.2.0

30th November 2021

Updated

• Pangolin updated to PangoLEARN 2021-11-18

v12.0.3

13th December 2021

New

• Collections can be created from a set of genomes of any species, though they do still have to be the same species. Trees of the collected genomes are still only generated for the same restricted set of organisms.

• Two new filters have been added to the Genomes List page:

  • The list of species and number of genomes with cgMLST assignments.

  • The list of species with collections and the number of collections.

• A new "fair share" algorithm has been implemented for managing task priorities. This should guarantee quick access to our resources for users who are uploading small numbers of genomes.

v11.1.0

3rd December 2021

Updated

• Pangolin updated to PangoLEARN 2021-1125

v11.0.0

30th November 2021

This release marks the addition of enhanced support for Vibrio cholerae. We expect to significantly expand the number of public genomes in the near future, and invite feedback on all aspects.

Added (for V. cholerae)

• SNP trees and collections

• AMR predictions

• Virulence and pathogenicity islands (using our in-house tool Vista)
• MLST, cgMLS and cgMLST-based clusters.

v10.2.0

30th November 2021

Updated

• Pangolin updated to PangoLEARN 2021-11-18

v10.1.0

17th November 2021

Updated

• Pangolin task updated to PangoLEARN 2021-11-04

v10.0.23

11th November 2021

New

• Under the cluster panel in the genome report (for species with a cgMLST scheme) there is a new "List Genomes" button. Clicking this will open the Genome List view and display the set of genomes making up the current cluster. This provides a method of generating a genomic context for a query genome, which can then be used to build a collection and tree for supported species, or extract genome analysis detailed downloads.
• Documentation describing the use of the Clustering panel in the Genome Reports.

v10.0.21

11th November 2021

Fixed

• The Strain download (based on PopPUNK 2) for Streptococcus pneumoniae is working again.

v10.0.18

4th November 2021

Fixed

• The queries underlying the genome list view have been substantially sped up, fixing most issues with scrolling and sorting, especially when logged in.

• The genome statistics charts are again sorted according to the list tab.

• We believe that the slow queries from the genome list view were the primary contributor to most, if not all, of the outages the website has experienced since we migrated it to the new infrastructure. As a consequence of the above improvements, the website should be much more stable from now on.

v10.0.17

3rd November 2021

Updated

• Pangolin updated to v3.1.16 and PangoLEARN 2021-10-18

v10.0.16

1st November 2021

Fixed

• Service stability Improvements.

v10.0.15

25th October 2021

Updated

• Poppunk has been updated to PopPunk2

v10.0.14

15th October 2021

Updated

• Pangolin has been updated to version v3.1.14 and PANGOLEARN 2021-09-28. All SARS-CoV-2 genomes have been updated with the latest assignments.

v10.0.13

13th October 2021

Updated

• The Klebsiella pneumoniae-complex cgMLST scheme has been replaced with the new scheme, "scgMLST629_S", hosted by the Pasteur Institute. This affects clustering for the following organisms:

  • Klebsiella pneumoniae (taxon id: 573)

  • Klebsiella quasipneumoniae (taxon id: 1463165)

  • Klebsiella variicola (taxon id: 244366)

  • Klebsiella africana (taxon id: 2489010)

v10.0.12

12th October 2021

Updated

• Internal task management tweaks should overall increase robustness

Fixed

• Errors when building trees are no longer reported prematurely. This led to failures being reported in the collection view and then the tree being available upon refresh some point later.

v10.0.10-update

11th October 2021

Fixed

• All now collections appear in the collection filter for each "supported organism" in the Genome View. This was due to some missing links in the database, lost during the cloud migration, that have been recreated. No code was been changed.

v10.0.10

7th October 2021

This is our first update release since we migrated to the cloud. This time we were able to use a new queue priority system, which allowed us to update all the relevant genomes via Pathogenwatch itself without affecting user uploads.

Updated

• Pangolin has been update to version v.2.4.2, along with PangoLEARN to version 2021-04-28.
• The NG-STAR database was recompiled on 15th September 2021.
• The MLST and cgMLST databases were recompiled on 15th September 2021.

We've updated the cgMLST assignments for nearly 400,000 unique genomes in this release of Pathogenwatch, requiring over 15,000 hours of CPU time. Being in the cloud allows us to easily expand the system to handle big jobs.

v10.0.9

6th October 2021

Fixed (-ish)

• The statistics panel in the genomes page tended not to load for larger datasets and put a significant load on the database. While we were able to resolve most cases, to fix it for logged in users appeared to require a substantial change in the underlying code with potential knock-on effects on the performance of other parts of the system. So for now we have disabled sorting of the results in the stats view and the stats are just presented in their "natural" order - approximately the order the genomes were uploaded to Pathogenwatch. We apologise to anyone who use this feature, and we are looking to reinstate it in the future

v10.0.8

4th October 2021

New

• Tree building and cluster searches have been given priority on the task queues to improve the website interactivity. This emulates the past behaviour of Pathogenwatch (prior to our recent migration), where we ran separate servers for the tree building and normal tasks.

Fixed

• After uploading genomes, the correct queue length is now shown when waiting for tasks to run. Previously the tasks were shown as running even if there was a long wait.

• Task failures are now correctly reported for uploads. Previously they were only visible when refreshing the upload page. They are also only reported once a task has fully failed and not while it is being retried.

v10.0.6

30th September 2021

We are aware of ongoing issues with website stability and queue length reporting since release 10.0.0. We are working on updates to the code, system configuration and metric/log insights and are looking to improve the situation as fast as possible. Please bear with us.

v10.0.1

15th September 2021

Updated

• Internal service stability improvements. If a worker service freezes it will automatically be restarted after a period of time.

v10.0.0

8th September 2021

Updated

• Pathogenwatch has been migrated to new servers hosted by DigitalOcean. This development is key to our ability to support Pathogenwatch into the future. If you have any concerns about your data please do contact us. And many thanks to our previous hosts at the Wellcome Sanger Institute.
• The maximum size of a collection for all supported species is now 2000, except SARS-COV-2 which remains at 1000 due to the higher computational requirements for tree building.
• The size of links in the download menu has been normalised, so that downloads are reliably initiated when the menu item is clicked.

• The assembly service has been brought into the main processing system, and so should be more stable.

v9.6.3 Data Update

17th June 2021

Added

*The NIHR Global Health Research Unit (GHRU), an international network of labs building genomic surveillance capabilities, have contributed assemblies and metadata for 1635 Klebsiella genomes. This includes 1451 Klebsiella pneumoniae, which are aggregated into a single collection: https://pathogen.watch/collection/f9tijvy5mn5f-ghru-k-pneumoniae. All the genomes and analysis of them are discussed further in an upcoming paper.

v9.6.3

16 June 2021

Updated

• Pangolin was updated to v3.1.1 and built with PangoLEARN 2021-06-15. All SARS-CoV-2 genomes have been updated to the latest version.

v9.6.1 Data Update

25 May 2021

Added

• 520 Salmonella Typhi genomes grouped into ten new collections each linked to 10 publications. Several are reports about azithromycin-resistant isolates, including the acrB R717L mutation included in release v9.6.0 (e.g. Sajob et al, 2021), and other extremely drug-resistant isolates. We also included genomes from the previously unrepresented Chile and the Republic of Korea (travel-associated).

Updated

• Collection sub-trees have been updated to include the new genomes.

• The complete public reference collection has been updated.

v9.6.1

19 May 2021

Updated

• The Genotyphi typhing scheme for Salmonella Typhi has been updated with the latest groups from genotyphi v1.9.1. The current Typhi genomes have been updated with these assignments.

v9.6.0

13 May 2021

Added

• A search tool specific to the porA gene for Neisseria gonorrhoeae was added. The output is shown in the Typing table in Collection and Cluster Views and in individual Genome Reports.

Updated

• The antimicrobial resistance panels (AMR) are now visible in the Cluster Views for all species that Pathogenwatch provides AMR assignments for. The network nodes can be coloured by selected antibiotics and genotypes. Significantly, this allows Streptococcus pneumoniae AMR results to be compared visually with each for the first time - see the example below.
• Resistance-suppressing mutations, such as disrupted resistance genes, are now displayed in the Variant genotype panel and Genome Reports. Suppressing mutations are teal, while "inducible" is orange. These were included in the predictions before but not properly displayed within the website.
• The AZM resistance mutation acrB R717L was added to the Typhi resistance library.

• The mtr_mosaic* were renamed to mtrR_promoter_mosaic* in the N. gonorrhoeae AMR library to clarify what they represent.

• Pangolin was updated to the current release (v2.4.2).

Cluster view of four S. pneumoniae genomes, coloured by the resistance-contributing variant folA I100L

Fixed

• The PANGOLearn library version is now correctly displayed in the Pangolin section of the Genome Reports.

v9.5.3

4 May 2021

Updated

• Pangolin was updated to version v2.4.0 and all SARS-CoV-2 assemblies have been rerun.

v9.5.2

19 April 2021

Updated

• Pangolin was updated to version v2.3.8 and all SARS-CoV-2 assemblies have been rerun.

v9.5.1

13 April 2021

Fixed

• Population subtrees were failing to build for single genome collections in supported species:

  • Salmonella Typhi

  • Staphylococcus aureus

  • Renibacterium salmoninarum

• Affected collections are rerunning and the trees should shortly be present. Please contact us if you are still affected.

v9.5.0

1 April 2021

New

• Pathogenwatch now provides support for SARS-CoV-2 analysis.

  • Genomes are annotated with lineages from Pangolin and notable mutations from the Type Variants Database.
  • Genomes are also aligned to the Wuhan Hu 1 reference and dendrograms of up to 1000 samples can be built from the alignments by creating collections.
• Viral genomes can be uploaded in batches as a single FASTA.

Updated

• The 404 page now provides clearer information and support links.

v9.4.2

22 March 2021

Updated

• Kleborate was updated to v2.0.4 . This brings several improvements and fixes to the typing assignments and allele calling.

v9.4.0

3 March 2021

Added

• Inctyper plasmid type assignments are available in the Cluster view for all bacterial species.
• Kleborate typing columns and AMR tables are now available in the cluster view for all Klebsiella and Raoultella.
• Genotyphi is now visible in the Cluster view for Typhi.
• A new button on each collection description card in the Collection Browser that takes you to the Genome Browser with those genomes displayed. This is useful if you want to build a new collection that includes the genomes from one or more other collections.

v9.3.3

22 February 2021

Changed

• The "MLST" column in the Genome List view has been replaced with "Type" and "Type source" to allow the display of multiple typing schemes. Genotyphi is now also displayed alongside MLST for Typhi genomes.

• Inctyper results are now displayed in the Typing table for all bacterial species collections.

v3.13.17

25 January 2021

Updated

• Kleborate updated to the v2.0.1 release. This includes a small bug fix that we believe did not affect any genomes in Pathogenwatch.
• The Kleborate download CSV available from the Genomes View now includes the complete Kleborate output.

v3.13.16

19 January 2021

Added

• Core statistics summary added to the Genome View downloads for organisms that have a Pathogenwatch core. This new file includes the core coverage, number of complete alleles and Pathogenwatch reference.

v3.13.15

11 January 2021

Fixed

• The Cluster View now correctly reduces the number of genomes in the metadata tables to just those shown in the network panel when the threshold is reduced using the threshold slider in the network panel.

v3.13.14

11 January 2021

Added

• The Kleborate AMR annotation tables are now visible in the Cluster View and not just in the Collection View.
• Nodes in the cluster view can now be coloured by the presence/absence of Kleborate AMR-associated genes and variants or by Kleborate-assigned AMR phenotype.

Fixed

• "OmpK35/36_truncated" matches are now correctly identified in the Kleborate AMR Genotypes table in the Cluster/Collection Views.

v3.13.11

10 December 2020

Fixed

• cgMLST downloads from the Genomes tab now work again.

v3.13.9

26 November 2020

Updated

• Kleborate has been updated to Kleborate version 2.0 and run over all ~16,000 public Klebsiella and Raoultella assemblies, along with a similar number of private assemblies.

v3.13.1

11 November 2020

Updated

• Our new email address is [e-mail]. The links have been updated in the website and our documentation. This is also the best way to report a problem or make a suggestion as now described in "Report an Issue". It will directly create a ticket on our internal issue tracker without you needing to make a GitLab account, and you can simply send further information by replying in the same thread.

v3.13.0

10 November 2020

Added

• The "AMR Genotypes" table has been add to the K. pneumoniae collection view. It displays AMR resistance genes and variants identified by Kleborate and grouped by resistance phenotype in a similar way to the AMR Genes table in the other species' collection views.

• Downloads for the "Antibiotics" and "AMR Genotypes" tables in the K. pneumoniae collection view.

Updated

• Changed the table name of "Kleborate AMR" to "Antibiotics" for consistency.

• Enabled colouring of the tree and map panels by Antibiotic and AMR Genotype table columns as was already possible for AMR panels in other species.

Fixed

• Fixed the mouse-over text in the K. pneumoniae antibiotics panel.

v3.12.14

7 November 2020

Updated

• A new version of Inctyper was built using the 5/11/2020 version of the plasmidfinder database.
• All ~300,000 genomes in Pathogenwatch, including the 70,000 public genomes, have been updated.

v3.12.9

23 October 2020

Fixed

• Advanced searches now work in the collection view. Previously clicking the '+' button produced a blank browser window. Now it produces search options.

• Mousing over resistance markers in the Kleborate AMR tab now shows the resistance genes and variants in the tooltip instead of {Object: object}.

v3.12.7

13 October 2020

Updated

• Replaced / with -- when representing contigs with multiple Inc locus matches in the Typing table of the Collection View. Some Inc types include the / making it difficult to programmatically parse when downloaded.

v3.12.6

30 September 2020

New

• The Klebsiella O Locus types can now be filtered on in the Genome View and type members can be accessed from individual Genome Reports.

Fixed

• The K locus filter and genome report links have been fixed, including for Raoultella species.

v3.12.4

11 September 2020

Fixed

• Core variance downloads for N. gonorrhoeae were failing with a network error. This is now fixed.

v3.12.0

4 September 2020

New

• SARS-CoV-2 genomes are now identified by the Speciator tool and can be selected for view using the subspecies filter in the genome page.

Updated

• The Speciator tool has been updated to provide more accurate (and faster) species assignments, especially for Klebsiella and other Enterobacteriaceae. We will be updating assignments within the database for both public and private genomes over the next week. Please report any issues to us. The new version was developed by Corin Yeats & Anthony Underwood with many thanks to Kat Holt for the Kleborate library and Bacsort and the rest of the CGPS team for testing.

v3.11.21

1 September 2020

Fixed

• The "Variance" download is now available again from the collection view. However, we note it is still failing on some larger or more diverse collections. The resolution of this can be tracked at https://gitlab.com/cgps/pathogenwatch/roadmap/-/issues/58.

v3.11.16

1 July 2020

Updated

• Updates to the AMR libraries for Neisseria gonorrhoeae and Salmonella Typhi.

Intermediate penicillin resistance determinant penA A501T was added to the N. gonorrhoeae library, while the qnr gene and QRDR mutation phenotypes were corrected according to evidence from other Salmonella. Thanks to Leo Sánchez-Busó, Silvia Argimon, Corin Yeats and our collaborators in the N. gonorrhoeae Pathogenwatch Scientific Steering Group and Kat Holt.

v3.11.15

18 June 2020

Updated

• Kleborate now run for the Raoultella genus species.

The Raoultella genus is in the process of being merged back into Klebsiella as genome sequence evidence does not support them being a separate genus. Kleborate provides accurate speciation and analysis for them either way.

v3.11.13

30 April 2020

Updated

• AMR for Salmonella Typhi.

Added the dfrA17 gene as a marker for trimethoprim resistance and restricted the ESBL matches to exact alleles only after we identified some false positive hits.

v3.11.12

21 April 2020

Updated

• AMR for Neisseria Gonorrhoeae and Salmonella Typhi.

This release fixes a bug affecting the very small number of assemblies (1 in 10,000) with non-ATCGN characters within AMR-associated genes. It also refines the porA analysis for N. gonorrhoeae to ensure that disrupted (premature stop codon or frameshift) genes are not reported.

v3.11.11

15 April 2020

Updated

• MLST and NG-STAR

Novel alleles are called using Blast. Most of these are called with a minimum percentage identity of 80% and a minimum coverage of 80%. This release fixes a bug which, in rare cases, permitted lower quality matches below this threshold and caused some novel STs to be called incorrectly.

Most of these were found in Klebsiella. This release enforces a minimum identity of 80% for all novel alleles. We're reran MLST and NG-STAR on all samples; you may notice that some novel STs have now been called as known.

v3.11.10

14 April 2020

Updated

• Major update to the Neisseria gonorrhoeae AMR library and phenotypes.

Every gene, variant and combination thereof was verified against an in-house set of ~5000 assemblies with MIC values (manuscript in preparation).

Developed by Leonor Sánchez-Busó (CGPS) & Corin Yeats (CGPS) with thanks to the CGPS team and our collaborators: Michelle Cole (Public Health England, UK), Yonatan H. Grad (Harvard TH Chan School of Public Health, Boston, USA), Irene Martin (Public Health Agency of Canada), William M. Shafer (Emory Antibiotic Resistance Center, Atlanta, USA), Gianfranco Spiteri (European Centre for Disease Prevention and Control, Sweden), Katy Town (Centre for Disease Prevention and Control, Atlanta, USA), Magnus Unemo (WHO Collaborating Centre for Gonorrhoea and other STIs, Örebro University, Sweden), Teodora Wi (World Health Organization, Geneva)

v3.11.9

7 April 2020

Updated

• Salmonella Typhi AMR assignments based on feedback and literature review by Silvia Argimon (CGPS).
• Genotyphi now correctly detects the 3.2 clade.

v3.11.8

6 April 2020

Updated

Databases have been updated for:

• MLST

• cgMLST

• NG-STAR

We have also changed the algorithm for assigning novel STs for MLST, cgMLST and NG-STAR. We assign STs by hashing the list of alleles of each locus. We now remove duplicate alleles for each gene before calculating or looking up the ST.

This approach gives better results for some schemes such as NG-STAR. Our experience shows that some loci are duplicated many times in long read (or reference data) but only once in assemblies from short reads. Before this change, long and short read data would give different results.

All genomes have been re-run with this new algorithm. You should expect better ST assignment and fewer novel results for all species and all of these tools.

v3.11.0

24 February 2020

Updated:

• Clicking on an assembly name when browsing a collection or viewing a cluster now pops up the detailed genome report for that assembly.

v3.10.0

29 January 2020

Updated:

• S. pneumoniae AMR library updated with help from Ben Metcalf (CDC) and Stephanie Lo.

v3.9.1

17 January 2020

Updated:

• Kleborate updated to v1.0.0. This update includes improvements to the species assignment, updated MLST, a more comprehensive antibiotic resistance profile and more. For more details about Kleborate in Pathogenwatch please see the documentation.

v3.9.0

16 January 2020

Added

• Streptococcus pneumoniae AMR predictions from PBP sequence analysis using the ML method from Li et al (2016). Provided by Ben James Metcalf at the Centres for Disease Control and Prevention (CDC) and with help from Stephanie Lo.
• For more details see the documentation here.

v3.8.0

14 January 2020

Added

• Timeline in collection and clustering views

• SVG download for trees

• PNG and SVG download for networks

Changed

• Tree views upgraded to Phylocanvas 3

• Improved collection history

• Updated social media login branding

v3.7.5

4 December 2019

Added

• Dual MLST schemes for A. baumannii (Oxford/Pasteur) and E. coli (EnteroBase/Pasteur)

• NG-STAR for N. gonorrhoeae (ngstar.canada.ca)
• Filter by collection for supported organisms

• Extra typing filters:

  • NG-MAST for N. gonorrhoeae

  • Genotype for S. Typhi

  • K locus for Klebsiella

• Text search for filter values

Changed

• MLST schemes updated as of October 2019

• Automatically apply filters that only contain one value

• Use asterisk for novel STs, use (novel) and (missing) in MLST profiles

v3.6.0

17 October 2019

Added

• Identification of plasmid Inc types for bacterial species as defined by the PlasmidFinder database and the in-house search tool "Inctyper". Results are displayed in the Assembly Reports with integrated AMR gene annotations from the Pathogenwatch AMR module where available and in the Typing table of the collections.

v3.5.0

7 October 2019

Added

• Campylobacter ciprofloxacin resistance prediction.

v3.4.0

2 October 2019

Added

• Klebsiella pneumoniae sensu stricto collection trees are now available using an in-house core genome developed by Sophia David & Corin Yeats.

v3.3.9

2 October 2019

Changed

• Updated the Colistin resistance library gene names to be consistent with the CPE & ESBL libraries.

v3.2.2

23 July 2019

Added

• Private metadata can now be added to collections and clustering views

• A notification dot will appear next to "Previous Uploads" to prompt completion of an incomplete upload session.

Changed

• File extensions are now trimmed from genome names if a name is not provided in metadata.

v3.1.3

1 July 2019

Added

• Kleborate (https://github.com/katholt/Kleborate) is now available to all.
• Genome reports can now be printed.

v3.0.10

21 June 2019

Added

• Updated AMR library for Typhi to include an improved version of the ESBL library with cleaned up naming.

• Improved integration of the Kleborate output with the genome report for Klebsiella.

v3.0.7

5 June 2019

Added

• Automated assembly of short reads is now available as a trial service.

v2.8.9

10 April 2019

Added

• AMR results in the genome reports now link to the public AMR library repository at https://gitlab.com/cgps/pathogenwatch/amr-libraries/tree/master. For more information consult the README.md and the Pathogenwatch documentation.
• Salmonella Typhi AMR was updated to show AMP resistance when CEP resistance is found.

v2.8.2

21 March 2019

Added

• Salmonella Typhi AMR library has been extended to include a comprehensive set of acquired resistance genes for two new antibiotics, Meropenem (MEM) and Colistin (CST), and extending coverage of Broad Spectrum Cephalosporins (CEP). The dataset was assembled by Sandra Reuters, Sophia David, and Silvia Argimon, with sequences sourced from ResFinder and CARD.

• SISTR serotyping for Salmonella.

v2.6.11

10 January 2019

Added

• Kleborate - a Klebsiella typing and resistance analysis tool from Kat Holt et al.

Fixed

• Better stability of the task runner engine.

v2.4.3

23 September 2018

Fixed

• Fix highlight resetting the map zoom

• Refine click behaviour in network panel

• Fix performance regression when fetching edges for network

v2.4.0

20 September 2018

Changed

• Add threshold slider to cluster view

• Complete intersecting filters for collections and clusters

Fixed

• Fix crashes in cluster view

• Fix network layout freezing when moving from report to viewer

• Amend click behaviour in network panel

v2.3.0

13 September 2018

Changed

• Added network panel to cluster view

• Begin to intersect filters in collection and cluster views

• Add links to genomes for cgMLST species on homepage

Fixed

• Tidy selection in map on Genomes page

v2.2.1

6 September 2018

Fixed

• Fix GFF downloads

• Remove GFF downloads from cluster views

• Improve page title for cluster views

v2.2.0

5 September 2018

Changed

• Add Streptococcus equi core and collections

Fixed

• Regex searching in a collection or cluster view

v2.1.1

4 September 2018

Changed

• Added link to release notes in menu

• Moved genomes before collections in navigation

Fixed

• Added Google Analytics ID

v2.1.0

3 September 2018

Changed

• Better searching for genome names and collection titles