Why do laboratories model expected stutter percent by locus?

Why do laboratories model expected stutter percent by locus?

• A. To estimate sequencing error rates.
• B. To calibrate the instrument.
• C. To assess DNA degradation.
• D. To account for expected minor peaks so true alleles are not misinterpreted.

Answer: (D)

Stutter modeling by locus is used to distinguish true alleles from PCR artifacts that show up as smaller, “stutter” peaks. During PCR amplification of STR regions, the DNA polymerase can slip, producing a minor peak that is typically one repeat unit shorter than the main allele. The amount of this stutter peak is not random; it’s predictable and varies from locus to locus and with allele size, so labs build empirical expectations (percent stutter) for each locus. By knowing the expected stutter percentage, analysts can tell whether a nearby smaller peak is likely just stutter or could be a real second allele. This is especially important in mixtures or when two peaks are close in height, where miscalling a stutter as a true allele could lead to incorrect conclusions about an individual’s genotype. Stutter modeling helps set interpretation thresholds and improves the accuracy of peak calls. This focus on stutter is different from sequencing error rates, instrument calibration, or DNA degradation, which involve other aspects of data quality and interpretation.

Stutter modeling by locus is used to distinguish true alleles from PCR artifacts that show up as smaller, “stutter” peaks. During PCR amplification of STR regions, the DNA polymerase can slip, producing a minor peak that is typically one repeat unit shorter than the main allele. The amount of this stutter peak is not random; it’s predictable and varies from locus to locus and with allele size, so labs build empirical expectations (percent stutter) for each locus.

By knowing the expected stutter percentage, analysts can tell whether a nearby smaller peak is likely just stutter or could be a real second allele. This is especially important in mixtures or when two peaks are close in height, where miscalling a stutter as a true allele could lead to incorrect conclusions about an individual’s genotype. Stutter modeling helps set interpretation thresholds and improves the accuracy of peak calls.

This focus on stutter is different from sequencing error rates, instrument calibration, or DNA degradation, which involve other aspects of data quality and interpretation.