Processed directed evolution data

CSV tables. Rows are full-length genotypes from PacBio sequencing (Cry1ac) or 600-nt paired end Illumina sequencing (TadA). Columns are samples or timepoints. Values are frequencies such that columns sum to 1.
Following Badran et al.'s methods, Cry1ac genotypes were filtered to those with >1% frequency at any timepoint, which roughly corresponded to filtering all genotypes with fewer than 5 reads in every timepoint (see our Online Methods for more info.)
The same filter was applied on TadA data at a 0% frequency cutoff (equivalent to no filtering) due to higher read depth and accuracy from Illumina sequencing. A threshold of 5% was used to determine common mutations. Frequencies were combined for PANCEs but kept separate for each of three PACEs (see Online Methods).