DNA Profiling Explained: STRs, Matching, and the Limits of Interpretation

Forensic DNA profiling is a powerful tool for identifying individuals based on the unique patterns of DNA found in their cells. The technique centers on short tandem repeats (STRs) – tiny stretches of DNA that vary in length among people – and uses statistical methods to decide whether two DNA samples could come from the same source. This article explains the biology of STRs, walks through the matching process, and highlights the key factors that limit interpretation.

1. What is DNA profiling?

DNA profiling (also called DNA typing) is a laboratory method that creates a genetic “fingerprint” for a biological sample. The fingerprint consists of the sizes of a selected set of STR loci, each representing a specific position on the genome. Because the number of repeat units at each STR locus differs widely among individuals, the combined profile of several loci can distinguish one person from virtually any other person in the world.¹

2. Short Tandem Repeats (STRs)

2.1 Definition and structure

An STR is a short DNA sequence (typically 2–6 base pairs) that is repeated head‑to‑tail in a row. For example, the repeat unit “AGAT” might appear 10 times in a row at one locus (written as AGAT₁₀). The number of repeats is called the allele at that locus.

2.2 Why STRs are useful for forensic work

• High variability: The number of repeats at a given STR can range from a few to dozens, creating many possible alleles.
• Compact size: STR regions are only a few hundred base pairs long, making them easy to amplify even from degraded samples.
• Standardized kits: Internationally agreed panels (e.g., the CODIS core loci in the U.S.) contain 13–20 STRs that have been validated for forensic use.²

3. From Sample to Profile: The Laboratory Process

The basic workflow is the same for most forensic labs and follows four steps (isolation, processing, typing, interpretation).³

1. DNA extraction: Cells are lysed and DNA is purified from the evidence (blood, saliva, hair, etc.).
2. Quantification: The amount of DNA is measured to decide how many PCR cycles are needed.
3. Polymerase‑chain reaction (PCR): Primers that flank each STR locus amplify the region. The PCR products are then separated by capillary electrophoresis, which sorts fragments by size.
4. Allele calling: The electropherogram shows peaks; the height of each peak corresponds to an allele size. The lab records the repeat number for each locus, producing a profile such as “D5S818: 11,12; FGA: 22,24; …”.

4. Matching Two Profiles

4.1 Direct comparison

If the suspect’s reference sample and the crime‑scene sample share the same allele at every examined STR locus, the profiles are said to be a match. A mismatch at any locus generally excludes the suspect, because the chance of two unrelated individuals sharing the same allele at that locus is extremely low.

4.2 Statistical interpretation – Random Match Probability (RMP)

Because a perfect match can still occur by chance, forensic scientists calculate the probability that a random, unrelated person from the relevant population would have the observed profile. The calculation multiplies the individual allele frequencies for each locus (assuming independence) and then squares the product for heterozygous loci (Hardy‑Weinberg equilibrium). The result is the random match probability (sometimes expressed as a “one in X” figure).

Example: If the combined frequency of the observed alleles across 13 CODIS loci is 1 × 10⁻¹⁵, the RMP is “one in a quadrillion”. This figure is used by courts to convey the weight of the DNA evidence.²

4.3 Likelihood ratios and the prosecutor’s vs. defense’s perspective

Some jurisdictions report a likelihood ratio (LR) – the probability of the evidence if the suspect is the source divided by the probability if the suspect is not the source. An LR of 10⁸, for instance, means the evidence is 100 million times more likely if the suspect contributed the DNA.

5. Limits of Interpretation

5.1 Population genetics considerations

Allele frequencies differ among ethnic and geographic groups. Using an inappropriate reference database can over‑ or underestimate the RMP. Modern guidelines require labs to select a population database that matches the ancestry of the evidence donor as closely as possible.¹

5.2 Sample quality and mixtures

• Low template DNA: Very small amounts of DNA increase stochastic effects (e.g., allele drop‑out), leading to incomplete profiles.
• Mixtures: When DNA from two or more people is present, the electropherogram shows extra peaks. Interpreting mixtures requires additional statistical models and may lower the evidential weight.

5.3 Laboratory error and contamination

Human error (mis‑labeling, pipetting mistakes) and contamination (foreign DNA introduced during extraction) can produce false matches or exclusions. Accredited labs follow strict quality‑control procedures, but a small error rate remains.⁴

5.4 Biological limits

Identical twins share the same STR profile, so DNA profiling cannot discriminate between them. Moreover, mutations in STR regions are rare but possible, which could cause a true donor to appear as a mismatch.

6. A Worked Example

Scenario: A burglary scene yields a bloodstain. The suspect’s buccal swab is available. The lab analyzes the 13 CODIS STR loci and obtains the following simplified allele frequencies (illustrative only):

Assuming independence, the combined genotype frequency for each locus is calculated as 2 × p × q for heterozygotes (Hardy‑Weinberg). For D5S818: 2 × 0.12 × 0.15 = 0.036. For FGA: 2 × 0.08 × 0.10 = 0.016. For TH01: 2 × 0.20 × 0.05 = 0.020.

Multiplying the three loci gives an overall frequency of 0.036 × 0.016 × 0.020 ≈ 1.2 × 10⁻⁵, or about 1 in 83,000. Adding the remaining 10 CODIS loci (each typically contributing a factor of ~10⁻³) pushes the RMP to roughly 1 × 10⁻¹⁵ – a “one in a quadrillion” match, which would be considered overwhelming evidence if the sample were single‑source and the population database appropriate.

However, if the bloodstain were a mixture of two contributors, the interpretation would be more complex, and the evidential weight would be reduced.

7. Summary

DNA profiling relies on the high variability of STR loci to generate a statistical fingerprint. Matching involves direct allele comparison and the calculation of a random match probability or likelihood ratio. Interpretation must consider population allele frequencies, sample quality, mixtures, laboratory error, and biological limits such as identical twins. Understanding both the power and the constraints of STR profiling enables informed use of DNA evidence in forensic and legal contexts.

8. Next steps for learners

Explore how probabilistic genotyping software evaluates complex DNA mixtures, or study the emerging use of single‑nucleotide polymorphism (SNP) panels for ancestry inference and degraded samples.