Fig. 2

Quantitation of integration abundance and number is a function of sequencing coverage. a) The total number of sheared fragment length counts (blue) is substantially lower than the number of UMI counts (red) in each of four replicate libraries. b) A single library (#1179) was reanalysed using subsets of read pairs (1000, 3000, 10,000, 100,000 and 300,000 read pairs). Quantitation of the ten most clonal integrations for each of these subsets is shown using unique sheared fragment lengths identified per integration (blue) and UMI counts per integration (red). These values are similar when sampling lower numbers of reads but as the sample size increases, the sheared fragment length counts becomes saturated. c & d) The clonality and normalized clonality calculations for the ten most clonal integrations is calculated for all read subsets using fragment length counts (c) and UMI counts (d). For the lowest samplings (1000 & 3000 read pairs) the clonality and normalized clonality based on fragments (Fig. 2c) and UMIs (Fig. 2d) are very similar whereas a larger number of reads leads to underestimation of fragment length clonality for the most abundant inserts and conversely an overestimation of fragment normalized clonality for less abundant inserts