Application of a target enrichment-based next-generation sequencing protocol for identification and sequence-based prediction of pneumococcal serotypes

Table 1 Percentage of correct identification of pneumococcal serogroup/serotypes based on different cut-off limits on number of reads

Sample category	Definition for interpretation cut-off*	Median no. reads mapped against serotype/group (IQR)	Mean percentage of reads aligned against correct serotype (95% CI)	Mean percentage of reads aligned against second serotype (if present) (95% CI)	No. of samples	Percentage of correct ID
PCR1 (serotype assignment only)	All samples^**	2,162 (1,059–3,691)	80.5% (72.0–89.0%)	8.8% (2.7–14.9%)	24	91.7%
	Samples with >200 reads mapped against serotypes	2,229 (1,328–4,257)	86.2% (83.1–89.3%)	5.0% (3.7–6.34%)	23	100%
	Samples with >500 reads mapped against serotypes	2,245 (1,530–4,371)	86.5% (83.3–89.7%)	4.9% (3.5–6.3%)	21	100%
PCR2 (serotype and ID)	All samples^**	1,658 (807–2,573)	80.7% (72.9–88.5%)	7.5% (4.3–10.7%)	24	87.5%
	Samples with >200 reads mapped against serotypes	1,683 (898–2,574)	83.0% (76.4–89.6%)	6.6% (3.8–9.4%)	23	91.3%
	Samples with >500 reads mapped against serotypes	1,951 (1,178–2,633)	87.1% (85.1–89.1%)	4.8% (2.3–7.3%)	19	100%

^*Criteria for serotype allocation. Each category fulfilled the minimum criterion of >15% reads out of total reads assigned to the sample.
^**Criterion of >15% reads out of total reads mapped assigned to the sample.

ISSN: 1471-2180