Fig. 3
a Sample collection interval between index isolate and paired invasive isolate according to the clinical context of the paired isolate. b–d Variants identified by episode-specific mapping and variant calling (after exclusion of unrelated same-patient isolates). b Correlation between sample collection interval and number of mutations separating the paired isolates from the index isolate for invasive paired isolates (R2 = 0.106, p = 0.016) and paired colonising isolates (R2 = 0.000, p = 0.971). The dotted line represents one mutation. c Number of mutations according to the clinical context of paired isolate (persistent bacteraemia, relapse on treatment or relapse bacteraemia after treatment, paired colonising isolate). d Distribution of mutation types according to the clinical context of the paired isolate