Note that for each pair of paired-end reads, the two reads were mapped independently, and only those reads mapping uniquely to the genome were included in the data submission to FlyBase. In other words, information from one read was not used to resolve ambiguous mapping of its paired read.