tfpscan: Compute scanner statistics for nodes in tree.
In oliviabboyd/tfpscanner_preview: Transmission fitness polymorphism scanner

View source: R/tfpscanner.R

tfpscan

R Documentation

Compute scanner statistics for nodes in tree.

Description

Takes standard inputs in the form of a rooted phylogeny and data frame with required metadata (see below). Computes a logistic growth rate and a statistic for outlying values in a molecular clock (root to tip regression). Comparison sample is matched by time and region. If using parallel computation (ncpu>1), the code should be called via mpirun, e.g. mpirun -n <ncpu+1> R –slave -f <script>

Usage

tfpscan(
  tre,
  amd,
  min_descendants = 100,
  max_descendants = 20000,
  min_cluster_age_yrs = 1/12,
  min_date = NULL,
  max_date = NULL,
  min_blen = 1/30000/2,
  ncpu = 1,
  output_dir = paste0("tfpscan-", Sys.Date()),
  num_ancestor_comparison = 500,
  factor_geo_comparison = 5,
  Tg = 7/365,
  report_freq = 50,
  mutation_cluster_frequency_threshold = 0.75,
  test_cluster_odds = c(),
  test_cluster_odds_value = c(),
  root_on_tip = "Wuhan/WH04/2020",
  root_on_tip_sample_time = 2020
)

Arguments

`tre`	A phylogeny in ape::phylo or treeio::treedata form. If not rooted, must provide an outgroup (see below)
`amd`	A data frame containing required metadata for each tip in tree: sequence_name, sample_date, region. Optional metadata includes: sample_time(numeric), lineage, mutations.
`min_descendants`	Clade must have at least this many tips
`max_descendants`	Clade can have at most this many tips
`min_cluster_age_yrs`	Only include clades that have sample tips that span at least this value
`min_date`	Only include samples after this data
`max_date`	Only include samples before and including this date
`min_blen`	Only compute statistics for nodes descended from branches of at least this length
`ncpu`	number cpu for multicore ops
`output_dir`	Path to directory where results will be saved
`num_ancestor_comparison`	When finding comparison sample for molecular clock stat, make sure sister clade has at least this many tips
`factor_geo_comparison`	When finding comparison sample based on geography, make sure sample has this factor times the number within clade of interest
`Tg`	Approximate generation time in years. Growth rate is reported in these units.
`report_freq`	Print progress for every n'th node
`mutation_cluster_frequency_threshold`	If mutation is detected with more than this frequency within a cluster it may be called as a defining mutation
`test_cluster_odds`	A character vector of variable names in `amd`. The odds of a sample belonging to each cluster given tis variable will be estimated using conditional logistic regression and adjusting for time.
`root_on_tip`	If input tree is not rooted, will root on this tip
`root_on_tip_sample_time`	Numeric time that tip was sampled
`test_cluster_odds_values`	Vector of same length as `test_cluster_odds`. This variable will be dichotomised by testing for equality of the variable with this value (e.g. vaccine_breakthrough == 'yes'). If NULL, the variable is assumed to be continuous (e.g. patient_age).