A DNA fingerprint is a visual pattern of genetic markers unique to an individual, displayed either as dark bands on a gel image or as colored peaks on a digital chart. Reading one comes down to comparing the position of those bands or peaks between two or more samples. If the markers line up at every tested location, the profiles match. If even one location differs, the person is excluded as the source.
The process is straightforward once you understand what you’re looking at, but the details matter. Here’s how to interpret both classic band patterns and the modern digital profiles used in forensic labs today.
What a DNA Fingerprint Actually Shows
Your DNA contains millions of short, stuttering sequences where a small chunk of code (two to six letters long) repeats over and over. These are called short tandem repeats, or STRs. At any given spot in your genome, you might have the sequence CTAG repeated five times, while someone else has it repeated seven times. These differences are harmless, they don’t affect your health, but they vary enormously from person to person.
A DNA fingerprint measures the number of repeats at multiple specific spots, or loci. You inherit one version from each parent, so at every locus you carry two alleles. Some people carry two identical alleles (homozygous), others carry two different ones (heterozygous). The combination of alleles across all tested loci creates a profile so distinctive that the chance of two unrelated people sharing it is astronomically small.
In the United States, the FBI’s CODIS database now requires testing at 20 core loci. The original standard was 13, but seven additional loci were added in 2017 to make profiles even more discriminating. Each locus is like an independent question, and the more questions you ask, the harder it is for two people to give identical answers by coincidence.
Reading Bands on a Gel or Autoradiograph
The older, classic DNA fingerprint looks like a barcode: a column of dark horizontal bands against a light background. Each band represents a DNA fragment of a particular size. Smaller fragments travel farther through the gel during electrophoresis, so bands near the bottom are shorter fragments and bands near the top are longer ones.
To compare two samples, they’re run side by side in separate lanes on the same gel. A standard reference ladder (a lane with fragments of known sizes) is usually included so you can measure exactly where each band falls. Here’s how to read it:
- Count the bands in each lane. At each locus, a person will show either one band (homozygous, two copies of the same allele) or two bands (heterozygous, two different alleles).
- Compare band positions across lanes. A band in one lane lines up horizontally with a band in another lane only if those fragments are the same size, meaning they contain the same number of repeats.
- Check every locus. For a match, every band in the evidence sample must have a corresponding band at the same position in the suspect’s sample. A single misaligned band at any locus excludes that person.
In paternity testing, the logic adds one step. First, identify which bands the child shares with the known mother. The remaining bands must have come from the biological father. If the alleged father’s profile contains all of those remaining bands, he is included as the potential father. If any are missing, he is excluded.
Reading Peaks on a Modern Electropherogram
Most forensic labs today produce digital charts called electropherograms rather than physical gels. Instead of dark bands, you see a series of colored peaks along a horizontal axis measured in base pairs (the size of the DNA fragment). The vertical axis shows peak height, which reflects how much of that fragment was in the sample.
Each tested locus occupies its own labeled region on the chart. Different loci are tagged with different colored dyes during testing, so peaks can overlap in position without being confused. Here’s what to look for:
- One peak at a locus means the person is homozygous, carrying two identical alleles. The peak will typically be taller because both alleles stack into a single signal.
- Two peaks at a locus means the person is heterozygous. The two peaks should be close in height and sit at the same baseline. A large imbalance between paired peaks can signal a problem with the sample.
- Peak position tells you the allele’s size. A peak sitting at 120 base pairs represents a shorter fragment (fewer repeats) than one at 160 base pairs. Labs convert these positions into allele numbers (like “12” or “15”) that indicate the repeat count.
- Peak height and shape matter for quality. Clean peaks are sharp and well-defined with a strong signal. Broad, noisy, or irregularly shaped peaks suggest degraded DNA, low quantities, or technical issues.
Comparing two electropherograms works the same way as comparing gel bands. At every locus, the allele numbers in the evidence sample must match the allele numbers in the known sample. A full match across all 20 loci is an inclusion. Any discrepancy is an exclusion.
What “Match,” “Exclusion,” and “Inconclusive” Mean
A match (or inclusion) does not mean the DNA definitely came from that person. It means the person cannot be eliminated as the source, and the statistical weight of the match is then calculated. Analysts determine the random match probability: the likelihood that a randomly selected, unrelated person from the population would share the same profile. With 20 loci, this probability is often less than one in a billion or even one in a trillion.
An exclusion means the profiles are different and the person did not contribute that DNA. Importantly, as the National Institute of Justice notes, exclusion doesn’t mean the person wasn’t involved in a crime. Someone can be present at a scene without leaving detectable biological material.
An inconclusive result means the data isn’t clean enough to call either way. This happens more often than people expect, particularly with degraded, tiny, or mixed samples.
Why Mixed Samples Are Harder to Read
A single-source DNA profile is relatively simple: one or two peaks at each locus, clean and balanced. Mixed samples from two or more people are far more complicated. Instead of a maximum of two peaks per locus, you might see three, four, or more, and separating which peaks belong to which contributor becomes a serious challenge.
Several problems compound the difficulty. In sexual assault cases, for instance, the victim’s DNA often vastly outnumbers the suspect’s, so the suspect’s alleles show up as tiny peaks that can be hard to distinguish from background noise. If a sample sits at a crime scene for a long time, one person’s DNA may degrade more than another’s, creating uneven signal strength. When the total amount of DNA is extremely small (under about 200 trillionths of a gram), random effects during testing can cause alleles to drop out entirely or appear at inconsistent heights.
Contamination introduces additional phantom peaks. DNA from investigators, lab technicians, or even residue on lab equipment can show up in the profile. These “drop-in” alleles are often difficult to distinguish from real contributor alleles, especially in low-quantity samples. Stutter peaks, which are technical artifacts one repeat shorter than the true allele, can also mimic minor contributor alleles and further muddy the picture.
Because of these complications, determining the exact number of contributors to a mixed sample is never 100% certain with current technology. Specialized software now handles much of the statistical heavy lifting, using probabilistic models rather than the simple visual comparison that works for single-source profiles.
Practical Tips for Interpreting a Profile
If you’re looking at a DNA fingerprint for a class, a paternity report, or a forensic document, keep these principles in mind. First, always orient yourself with the reference markers. On a gel, find the size ladder. On an electropherogram, check the labeled loci and the base-pair scale along the bottom. These are your rulers.
Second, focus on one locus at a time. Identify the alleles in each sample at that locus before moving to the next. Trying to take in the whole profile at once leads to confusion, especially with multiple samples on the same image.
Third, remember that peak height on an electropherogram reflects quantity, not identity. A short peak and a tall peak at the same position represent the same allele; one sample simply had more DNA. What matters is horizontal position (size) not vertical height, unless you’re evaluating sample quality or trying to distinguish major from minor contributors in a mixture.
Finally, a single mismatch rules out a match. DNA fingerprinting is powerful precisely because it’s binary at each locus: the alleles are either the same number of repeats or they aren’t. There’s no “close enough.” Two profiles that share 19 out of 20 loci but differ at one are from different people.