Figure 1

Amplicons computation pipeline. Step I groups sequenced amplicons according to their most similar reference sequence (or sequences: more than one reference accepted in the case of close similarity of references). Step II clusters groups of sequences belonging to the same references while applying certain thresholds and heuristics. Step III detects punctual mutations and avoids false positives due to homopolymers. A database stores sequence data and computation results (intermediate and final), and fully separates the steps of the pipeline.