Fig. 1

PtWAVE algorithm flow chart. A It requires four pieces of information as input data: (1) one sequencing trace file from the control (WT) sample, (2) one sequencing trace file from the edited sample, (3) target sequence information, including the protospacer sequence and PAM motif, and (4) parameter settings, including the mutation ranges, fitting algorithms, and variable selection methods. The algorithm first checks the input data and finds an alignment window based on the sequencing quality. The algorithm then aligns the primary peaks between the two samples by referring to the alignment window. The decomposition window was determined on the basis of the alignment window, alignment, and sequence quality. The algorithm also generated EMSPs and output vectors based on the input data to model the DNA sequence trace chromatograms of the edited sample in the decomposition window. There are various choices for modeling chromatograms, such as variable selection modes (all, random, and backstep) and fitting algorithms (NLLS and LASSO models). After modeling the chromatograms, the PtWAVE algorithm provided the composition ratios of the indels of the edited sample and the editing efficiency. B An illustration explaining the key nine steps from alignment window determination through decomposition window determination. These windows are determined based on user input, given sequencing traces, and quality values (phred score) from direct sequencing of pooled alleles, including allele 1, 2, 3, …, N. The borders of the alignment window (blue box) are called “aln_start” and “aln_end.” The borders of the decomposition window (green box) are called “inf_start” and “inf_end.”