Accurate detection of subclonal single nucleotide variants in whole genome amplified and pooled cancer samples using HaloPlex target enrichment

Table 3 Accuracy of single nucleotide variant detection in non-indexed pools

Sample	TP^a	TN^a	FP^a	FN^a	FDR (%)^b	Novel^c
ALL1_pool2	224	304	1	3	0.4	0
ALL2_pool5	304	227	0	1	0	3
ALL1_pool10	225	296	9	2	3.8	4
ALL1_pool2_rep	224	304	1	3	0.4	0
ALL2_pool5_rep	298	222	5	7	1.7	2
ALL1_pool10_rep	223	291	14	4	5.9	4

^a TP: true positives; TN: true negatives; FP: false positives; FN: false negatives.
^b FDR: false discovery rate, calculated as FP/(TP + FP).
^c Putative novel SNVs called in the previously uncharacterized samples included in the pools. One of the variants called in ALL2_pool5 was not called in the corresponding replicated pool (ALL2_pool5_rep). All other novel variants were identical between replicates.

ISSN: 1471-2164