config.yaml

# Configuration file

# Pre-built UShER tree, from:
# https://hgdownload.gi.ucsc.edu/hubs/GCF/000/864/105/GCF_000864105.1/UShER_NC_007362.1/
usher_prebuilt: 
  tree: https://hgdownload.gi.ucsc.edu/hubs/GCF/000/864/105/GCF_000864105.1/UShER_NC_007362.1/fluA.GCF_000864105.1.NC_007362.1.latest.pb.gz
  version_info: https://hgdownload.gi.ucsc.edu/hubs/GCF/000/864/105/GCF_000864105.1/UShER_NC_007362.1/fluA.GCF_000864105.1.NC_007362.1.latest.version.txt
  ref: NC_007362.1
  ref_cds_coords: [22, 1728]

# Deep mutational scanning (DMS) data from
# https://github.com/dms-vep/Flu_H5_American-Wigeon_South-Carolina_2021-H5N1_DMS
dms:
  prot: https://raw.githubusercontent.com/dms-vep/Flu_H5_American-Wigeon_South-Carolina_2021-H5N1_DMS/main/results/gene_sequence/protein.fasta
  site_numbering_map: https://raw.githubusercontent.com/dms-vep/Flu_H5_American-Wigeon_South-Carolina_2021-H5N1_DMS/main/data/site_numbering_map.csv
  site_numbering_schemes:  # map name to use to column in site_numbering_map
    sequential: sequential
    mature_H3: reference
    mature_H5: mature_H5
  phenotypes: https://raw.githubusercontent.com/dms-vep/Flu_H5_American-Wigeon_South-Carolina_2021-H5N1_DMS/main/results/summaries/phenotypes.csv
  escape_by_species: https://raw.githubusercontent.com/dms-vep/Flu_H5_American-Wigeon_South-Carolina_2021-H5N1_DMS/main/results/summaries/phenotypes_per_antibody_escape.csv

# for DMS phenotypes, keep this many digits to the right of the decimal
dms_decimal_scale: 3

# For the scatter plots of phenotypes, initial phenotypes for x and y-axis
pheno_scatter_init_phenos:
  x: ferret_sera_escape_increase
  y: frac_divergence

# Filters for analyses: keep just strains fulfilling filter for each analysis
analysis_filters:
  cattle:
    frac_aligned: 0.98  # require >= this fraction of residues to be aligned (not gapped)
    frac_divergence: 0.1  # require divergence among aligned residues to be <= this
    strain_regex: "(?i)cattle(?!_egret)|(?i)dairy_cow"  # require strain to contain this regex
  max-diff-0.04:
    frac_aligned: 0.98  # require >= this fraction of residues to be aligned (not gapped)
    frac_divergence: 0.04  # require divergence among aligned residues to be <= this
  max-diff-0.08:
    frac_aligned: 0.98  # require >= this fraction of residues to be aligned (not gapped)
    frac_divergence: 0.08  # require divergence among aligned residues to be <= this
  max-diff-0.12:
    frac_aligned: 0.98  # require >= this fraction of residues to be aligned (not gapped)
    frac_divergence: 0.12  # require divergence among aligned residues to be <= this

title: Influenza H5 HA annotated by deep mutational scanning phenotypes

description: |
  See https://github.com/jbloomlab/Flu-HA-H5-2.3.4.4-DMS-informed-surveillance for details.