config.yaml

# Configuration file for pipeline

# specify one of the mutation-annotated trees in `mat_trees` as the "current" one that is
# linked to by default as the best current estimates.
current_mat: gisaid_2024-04-24

# Links to dataframes from Jesse's repository: jbloomlab/SARS2-mut-fitness
url_founder: https://raw.githubusercontent.com/jbloomlab/SARS2-mut-fitness/refs/heads/main/results_gisaid_2024-04-24/clade_founder_nts/clade_founder_nts.csv
url_counts: https://media.githubusercontent.com/media/jbloomlab/SARS2-mut-fitness/refs/heads/main/results_gisaid_2024-04-24/expected_vs_actual_mut_counts/expected_vs_actual_mut_counts.csv

# Mapping for grouping clades into clusters
clade_cluster:
  {'ancestral': ['20A', '20B', '20C', '20E', '20G', '21C'],
 '20H': ['20H'],
 '20I': ['20I'],
 '20J': ['20J'],
 '21I': ['21I'],
 '21J': ['21J'],
 '21K': ['21K'],
 'BA.2': ['21L', '22C'],
 'BA.4-5': ['22A', '22B'],
 'BA.2.75': ['22D', '23C'],
 'XBB': ['22F', '23A', '23B', '23D', '23E', '23F', '23H'],
 'late': ['23I'],
 'all': ['20A', '20B', '20C', '20E', '20G', '21C', '20H', '20I', '20J', '21I', '21J', '21K', '21L', '22A', '22B', '22C', '22D', '22F', '23A', '23B', '23C', '23D', '23E', '23F', '23H', '23I'],
 }

 # Mapping for grouping clades into clusters
cluster_founder:
  ancestral: 20A
  20H: 20H
  20I: 20I
  20J: 20J
  21I: 21I
  21J: 21J
  21K: 21K
  BA.2: 21L
  BA.4-5: 22A
  BA.2.75: 22D
  XBB: 22F
  late: 23I
  all: 20A

  # Orf1ab to Nsp numbering (amino-acid start in Orf1ab) from
# https://github.com/theosanderson/Codon2Nucleotide/blob/main/src/App.js
orf1ab_to_nsps:
  nsp1: [1, 180]
  nsp2: [181, 818]
  nsp3: [819, 2763]
  nsp4: [2764, 3263]
  nsp5 (Mpro): [3264, 3569]
  nsp6: [3570, 3859]
  nsp7: [3860, 3942]
  nsp8: [3943, 4140]
  nsp9: [4141, 4253]
  nsp10: [4254, 4392]
  nsp12 (RdRp): [4393, 5324]
  nsp13: [5325, 5925]
  nsp14: [5926, 6452]
  nsp15: [6453, 6798]
  nsp16: [6799, 7096]

# How to handle nucleotide sites that overlap among sites: exclude them from
# amino-acid fitness estimates or retain those sites in estimates. If retained, estimates
# can be confounded by not knowing which gene the selection is acting on.
gene_overlaps:
  exclude:
    - [ORF1a, ORF1ab]  # 11 sites of overlap near ribosomal frameshifting location
    - [ORF7a, ORF7b]  # 1 site of overlap corresponding to ORF7a stop codon
  retain:
    - [N, ORF9b]  # ORF9b is a 291 out-of-frame overlapping reading frame in N

# Pseudocount for calculating amino-acid fitnesses
fitness_pseudocount: 0.5

# Genes order on the viral genome
genes: ['ORF1ab', 'nsp1', 'nsp2', 'nsp3', 'nsp4', 'nsp5 (Mpro)', 'nsp6', 'nsp7', 'nsp8', 'nsp9',
  'nsp10', 'nsp12 (RdRp)', 'nsp13', 'nsp14', 'nsp15', 'nsp16', 'S', 'ORF3a', 'E', 'M', 'ORF6',
  'ORF7a', 'ORF7b', 'ORF8', 'ORF9b', 'N', 'ORF10']

# Define common names of Nexstrain clades
clade_synonyms:
  19A: B
  20A: B.1
  20B: B.1.1
  20C: B.1.367
  20E: B.1.177
  20F: D.2
  20G: B.1.2
  20I: Alpha
  20H: Beta
  20J: Gamma
  21C: Epsilon
  21F: Iota
  21I: Delta
  21J: Delta
  21K: Omicron BA.1
  21L: Omicron BA.2
  22A: Omicron BA.4
  22B: Omicron BA.5
  22C: Omicron BA.2.12.1
  22D: Omicron BA.2.75
  22E: Omicron BQ.1
  22F: XBB
  23A: XBB.1.5
  23B: BB.1.16
  23C: CH.1.1
  23D: XBB.1.9
  23E: XBB.2.3
  23F: EG.5.1
  23G: XBB.1.5.70
  23H: HK.3
  23I: BA.2.86
  24A: JN.1
  24B: JN.1.11.1
  24C: KP.3
  24D: XDV.1
  24E: KP.3.1.1
  24F: XEC
  24G: KP.2.3

# minimal domain for amino-acid fitness heatmap
aa_fitness_heatmap_minimal_domain: [-6, 2] 

# Default minimum expected count for amino-acid fitness estimates
min_predicted_count: 10

# minimum count for correlation calculation
cluster_corr_min_count: 500000  

# For GitHub Pages docs of plots
docs_url: https://neherlab.github.io/SARS2-refined-fitness  # base URL for GitHub pages site
docs_plot_annotations: data/docs_plot_annotations.yaml  # lists plots and their annotations