Metrics to compare different tools for PE reads merging

Question

[Denis]

Hi there, I'd like to compare different tools for PE reads merging like FLASH, PEAR etc. on my amplicon metagenome data. The naive approach is to use the number of merged reads as metric to choose the best read merger, but in such case i'll not count the False Positives (FP), False Negatives (FN) and so on. Do you know which metrics will be more correct or how i can culculate FN, FP, TP, TN in my data? Thank you in advance, Denis

[GenoMax]

Brian Bushnell had looked at this when he released BBMerge (part of BBMap).