pangoLEARN data release 2021-05-27

Release notes

• Update to pangoLEARN to now include lineages.hash.csv for assignment of previously designated sequences
• Update to no longer have lineages.metadata.csv as a file hosted
• Model from pango-designation version 1.2.6
• Compatible with pangolin version 3.0 or greater.

pangoLEARN data release 2021-05-19

Release notes

• pangoLEARN decision tree model trained from pango-designation release v1.2.6.
• Now implementing an assignment hash of the designations.

pangoLEARN data release 2021-05-11

Release notes

Updated decision tree model addressing issue #189 in pangolin.

pangoLEARN data release 2021-05-10

Release notes

New pangoLEARN decision tree model corresponding to pango-designation release v1.2, with large scale gate in N content implemented.

pangoLEARN data release 2021-04-28

Release notes

Update to pangoLEARN decision tree model for pango-designation version 1.1.23.

Now model comes with pango-designation version as part of the __init__.py file.

pangoLEARN data release 2021-04-23

Release notes

• Update to training model to remove duplicates from B.1.617.2.

pangoLEARN data release 2021-04-21

Release notes

Training based on pango-designation v1.1.19, now including sublineages of B.1.617. Full details found at cov-lineages/pango-designation#49.

pangoLEARN data release 2021-04-14

Release notes

• New decision tree model trained for pango-designations up to v1.1.14.
• Updated lineages and lineages added:

B.1.526
B.1.526.3 (new)
B.1.618 (new)
B.26
B.1.177.2
B.1.177.88 (new)
B.1.177.89 (new)
N.10 (new)
P.1.1 (new)
A.2.5
A.2.5.1 (new)
B.1.1.29 (reassigned from B.1.1.439 to fix pangothon error)
B.1.429.1 (new)

pangoLEARN data release 2021-04-01

Release notes

• New decision tree model trained for pango-designation v1.1.6 and v.1.1.5
• Now assigns lineage B.1.617 circulating in India (VOI related to pango-designation issue #38).
• Now assigns sublineages of B.1.214. Now 4 sister lineages, B.1.214.1 from Congo, B.1.214.2 from Belgium with a defining 9bp insertion in the NTD of the spike gene, B.1.214.3 a French lineage also found in Mayotte and B.1.214.4 a Danish lineage.

pangoLEARN data release 2021-03-29

Release notes

• pangoLEARN trained with downsample. Sequences dowsampled from 382,806 designated seqs to 177,916 seqs, maintaining diversity representing every node in the tree.
• Downsampling took 2.5 hours: training time went from >20 hours to 3 hours.
• We're hosting both the designated metadata here as well as the final downsample that went into the training.
• 1,249 lineages, 1,169 lineages downsampled as they had greater than 10 sequences designated.