and others, in addition to the more general packages described previously (27,58–61). These methods differ principally in the model used to discriminate between alignment errors and true indel calls, often resulting in greatly discrepant indel calls between software and orthogonally generated data (62). Of the many alignment-based indel detection methods, no single program has proven to be completely accurate and all require considerable validation when used clinically. A major drawback of alignment-based methods is the requirement that indels be contained within a read and identified during the initial read mapping and alignment stage, limiting insertion detection to approximately 15% of the total read length (22,59). However, larger deletions may be detected by these methods; this is often referred to as “deletion bias.”
