Fig. 1

Study design with monozygotic twins and data generation. a Overview of the study design. Briefly, DNA reference materials were constructed from immortalized cell lines of a Chinese Quartet with father (F7), mother (M8), and monozygotic twin daughters (D5 and D6). They were sequenced by four short- and three long-read platforms at seven labs. Small variant and structural variant benchmark calls were integrated from massive sequencing datasets. Performance of a test dataset can be evaluated by comparing with benchmark calls or genetic built-in truth within the Quartet family. b Schematic overview of short-read sequencing datasets. Three replicates for each of the Quartet DNA reference materials were sequenced in nine batches, by both PCR and PCR-free libraries on four sequencing platforms at six labs, resulting in 108 WGS libraries. c Schematic overview of long-read sequencing datasets. One replicate for each of the Quartet DNA reference materials was sequenced per batch by PacBio Sequel, PacBio Sequel II, and ONT. Eleven combinations of three mappers and five callers were used to detect structural variants, resulting in 120 variant calling datasets