Skip to contents

Identify transmission clusters based on the number of shared variants

Source: R/cluster_algorithms.R
get_tn_clusters_sv_index.Rd

Clustering is performed to identify the maximal clusters containing a single intake-positive patient that occurs before all cluster converts. The clustering metric is the number of shared variants, and clusters can have multiple intake-positive patients if they share an identical number of variants with other cluster members or intake-positive patients occur after converts. This clustering also requires that clusters be defined by at least one shared variant that other isolates don't have.

Usage

get_tn_clusters_sv_index(
  dna_aln,
  snp_dist,
  adm_seqs,
  adm_pos_pt_seqs,
  seq2pt,
  dates,
  tree
)

Arguments

dna_aln

A DNA alignment object of class DNAbin. Its first sequence is taken as the outgroup.

snp_dist

A matrix of SNP distances between isolates. See get_snp_dist_matrix().

adm_seqs

A vector of sequence IDs for sequences from patients positive at intake.

adm_pos_pt_seqs

A vector of all sequence IDs from admission-positive patients, whether collected at intake or later; a superset of adm_seqs.

seq2pt

A named vector mapping sequence IDs to patient IDs.

dates

A named vector of isolate dates, named by sequence ID.

tree

A phylogenetic tree of class phylo over the same isolates, e.g. from get_phylo_tree(). Its first tip is taken as the outgroup, matching dna_aln.

Value

A numeric vector giving the cluster each isolate belongs to.

References

Hawken, S. E., Yelin, R. D., Lolans, K., Pirani, A., Weinstein, R. A., Lin, M. Y., Hayden, M. K., & Snitkin, E. S. (2022). Threshold-free genomic cluster detection to track transmission pathways in health-care settings: A genomic epidemiology analysis. The Lancet Microbe, 3(9), e652–e662. doi:10.1016/S2666-5247(22)00115-X