A Single Sample, Multiple Sources of DNA
Forensic swabs often carry cellular material from more than one person.
- ~30%: Common mixture rate (of casework samples)
- 2–4: Typical contributors (per mixed stain)
- <1 ng: DNA input (often low template)
- Swab: Collection method (touch or biological trace)
Where mixtures come from
Touch DNA picks up cells from many prior contacts.
Shared objects and surfaces accumulate multiple donors.
Biological stains can mix victim and offender material.
Why it matters early
Mixture status changes every downstream analysis choice.
Analysts flag mixtures before amplification and interpretation begin.
Reading Extra Peaks at Each STR Locus
A single contributor shows at most two alleles per locus.
- 2: Max alleles (1 donor) (per autosomal locus)
- 20+: Loci in core panel (CODIS-compliant kits)
- ~50 RFU: Detection threshold (analytical threshold)
- ~1 bp: Peak width (repeat unit resolution)
Spotting a mixture
Three or more alleles at one locus flags a mixture.
Capillary electrophoresis separates alleles by fragment size.
Reading the trace
Each peak height is measured in relative fluorescence units.
Peak position gives the allele's repeat number.
Peak Heights Encode the Contributor Ratio
Taller peaks generally come from the contributor present in greater quantity.
- >60%: Heterozygote balance (expected within one donor)
- 1:1–20:1: Mixture ratio range (major:minor typical)
- <15%: Stutter ratio (of parent peak height)
- ~150 RFU: Stochastic threshold (below which dropout risk rises)
Ratio estimation
Peak height ratios across loci estimate the mixture proportion.
Consistent ratios across loci support a shared contributor pair.
Complications
Stutter artifacts can mimic a low-level minor contributor.
Degraded DNA lowers heights unevenly across fragment sizes.
Assigning Peaks to Major and Minor Contributors
The tallest allele pair per locus is grouped as the major contributor.
- >65%: Major contributor share (typical major fraction)
- <35%: Minor contributor share (remaining template)
- 2-person: Separation feasible (mixtures most reliable)
- ~1:1: Ambiguous ratio (major/minor indistinguishable)
Grouping logic
Peaks cluster by consistent relative height across loci.
The tallest cluster is provisionally labeled the major profile.
Limits of separation
Even ratios make major and minor peaks hard to distinguish.
More contributors multiply possible allele combinations per locus.
Estimating Individual Profiles from the Combined Signal
Probabilistic genotyping software reconstructs likely single-source contributor profiles.
- Probabilistic: Method (genotyping (e.g. STRmix, TrueAllele))
- Likelihood ratio: Output (per candidate contributor)
- High: 2-contributor confidence (well-separated ratios)
- Low: 4-contributor confidence (combinatorial complexity)
Probabilistic genotyping
Software models thousands of possible genotype combinations.
Each combination is weighted by its fit to peak heights.
Confidence and limits
Confidence drops as contributor number and ratio evenness rise.
Deconvolution output supports comparison against a reference profile.
More contributors and more even ratios sharply lower deconvolution confidence.