About the details of calculation

[zwfsnowdream]

If the sgRNA sequence is not provided in the command line (i.e., the --sgRNA parameter is omitted), does this mean that each read will be aligned with the reference sequence, and any read with at least one base differing from the reference sequence will be considered a "modified read"? Additionally, when calculating the results, if we want to exclude reads with low homology to the reference sequence (i.e., not counting them as "unmodified reads"), how can this be achieved?

[kclem]

Hi @zwfsnowdream, Thanks for using CRISPResso.

If the sgRNA is not provided, the entire alignment is used except for the --exclude_bp_from_left and --exclude_bp_from_right, which default to excluding 15bp from each end of the amplicon, but can be set to 0 if you want the entire alignment included. Any bases that differ from the reference sequence (including insertions, deletions, and substitutions) would make the read 'modified'. You can also set the quantification window manually using the --quantification_window_coordinates parameter.

If you want to exclude reads with low homology, you can adjust the --default_min_aln_score parameter. The default is 60, meaning that 60% of bases must match the reference, which will remove reads with low homology to the reference sequence. That is, if you have a 100bp amplicon sequence, deletions longer than 40bp would cause the read to be unaligned and not count toward the 'unmodified' or 'modified' counts. The min alignment score can also be set for each amplicon individually using --amplicon_min_alignment_score.