Parameterize Nextclade dataset in config

[huddlej]

Description of proposed changes

Prototypes an idea of allowing users to choose their own Nextclade dataset for the final alignment with Nextclade. Adds minimal logic and config to allow users to pull in immune escape and ACE2 binding scores to their builds by switching their default dataset.

As with the Nextclade web UI, users must understand that their trees and annotations on those trees will not make sense biologically if they use sequences that don't descend from BA.2 (Nextstrain clade 21L). This PR doesn't implement any checks or force any filters to ensure that input data are reasonable, unlike the RBD levels annotations in this workflow or the ncov-ingest logic to pull in the same phenotypic scores.

Testing

Tested with this minimal build config based on the CI build:

inputs:
  - name: gisaid
    metadata: "s3://nextstrain-ncov-private/100k/metadata.tsv.xz"
    aligned: "s3://nextstrain-ncov-private/100k/sequences.fasta.xz"

nextclade_dataset: sars-cov-2-21L

builds:
  europe:
    subsampling_scheme: nextstrain_ci_sampling
    region: Europe

filter:
  skip_diagnostics: true

subsampling:
  # Custom subsampling logic for CI tests.
  nextstrain_ci_sampling:
    # Focal samples for region
    region:
      group_by: "division year month"
      max_sequences: 50
      exclude: "--exclude-where 'region!={region}'"
      min_date: --min-date 2022-01-01
    # Contextual samples for region from the rest of the world
    global:
      group_by: "year month"
      max_sequences: 50
      exclude: "--exclude-where 'region={region}'"
      min_date: --min-date 2022-01-01

Run the build like so:

snakemake --forceall -p -j 4 --use-conda --conda-frontend mamba --configfile immune_escape.yaml

An example tree with immune escape scores from this build looked like this:

And ACE2 binding:

This tree is a good example where the input sequences should be limited to 21L descendants and haven't been fully.

Release checklist

If this pull request introduces new features, complete the following steps:

• Update docs/src/reference/change_log.md in this pull request to document these changes by the date they were added.

[huddlej]

Since this is a minor backwards-compatible change that got general verbal approval at our recent Nextstrain meeting, I'm going to merge this without review.

[corneliusroemer]

@huddlej thanks a lot for providing a good debug template config and snakemake command - used this to great advantage in #1050! Ran in 5min including mamba installation!