DNA evidence can prove innocence by showing that a person’s genetic profile does not match the biological material left at a crime scene. That mismatch, called an exclusion, tells a court the accused could not have been the source of the blood, semen, saliva, skin cells, or hair the perpetrator left behind. Hundreds of people in the United States have walked out of prison on the strength of that result, some after more than a decade inside for crimes they did not commit.
Why a DNA Non-Match Carries So Much Weight
About 99.9 percent of human DNA is identical from one person to the next. The remaining fraction is enough to tell any two unrelated people apart, and it stays the same from birth to death. A sample collected twenty years ago will still match its source today, which is why old evidence can still answer new questions.
Forensic labs turn a biological sample into a numerical profile — essentially a genetic fingerprint — and then compare it to a profile from a known person. Either the numbers line up at every tested location or they do not. When they do not, the tested person did not leave that sample. An analyst is not making a judgment call; a non-match is a non-match.
That binary clarity is what separates DNA from the other evidence a jury usually weighs. Eyewitness testimony is subjective and often wrong. Confessions can be coerced. Informants can lie. A clean genetic exclusion, in a case where the perpetrator necessarily left biological material behind, directly contradicts the theory that put someone away. In a sexual assault conviction, for example, DNA from the evidence kit that does not match the defendant is physical proof against the verdict.
Exclusion is not always the whole story. If a crime could have involved multiple people, or if the biological evidence is peripheral to the charge, a non-match may weaken the case without ending it. But when DNA testing excludes a convicted person as the source of the central evidence, prosecutors and courts take the result seriously, and it can lead to dismissed charges, a new trial, or an overturned conviction.
What DNA Exonerations Have Actually Looked Like
As of 2016, at least 342 people in the United States had been exonerated through DNA testing, according to data compiled by the National Institute of Justice. The number has continued to climb. Innocence Project clients, a subset of these cases, had served an average of 16 years in prison before their release, with some imprisoned for decades.
Those cases have exposed a set of recurring failures behind wrongful convictions. Six factors show up again and again:
- Eyewitness misidentification, present in roughly two-thirds of DNA exonerations and the single most common contributor.
- Flawed forensic evidence, meaning outdated or misapplied techniques that pointed at the wrong person.
- False confessions, sometimes from juveniles or people with intellectual disabilities.
- Government misconduct, including withheld evidence, coached witnesses, or fabricated testimony.
- Unreliable informants, often jailhouse witnesses testifying in exchange for a benefit.
- Inadequate defense, where trial counsel failed to investigate or challenge the state’s case.
Where flawed forensic science was a factor, it almost never worked alone: 98 percent of those exonerations involved two to five additional contributing factors. Wrongful convictions tend to be cascades, not single mistakes. DNA exonerations have also revealed stark racial disparities, with Black defendants disproportionately represented among the wrongly convicted.
How a Convicted Person Requests DNA Testing
Post-conviction DNA testing is not automatic, and there is no federal constitutional right to it. The U.S. Supreme Court ruled in 2009 that access to testing after a conviction has to come from a statute. All 50 states and the federal government have since enacted such statutes, but their terms vary.
Federal Motions
Under federal law, a convicted person can file a motion asking for DNA analysis of biological evidence from the case. The applicant has to lay out a theory of defense that would establish actual innocence and show that the requested testing is reasonably expected to produce new evidence supporting it. The motion also has to demonstrate a reasonable probability that the applicant did not commit the offense if the results come back favorable. If the evidence was tested before, the applicant generally has to explain why the earlier technology was not discriminating enough to produce meaningful results.
State Petitions
Most criminal convictions happen at the state level, so state statutes are the more common route. Requirements differ. Many states ask the applicant to show that a favorable DNA result would have changed the outcome of the original trial. Some impose deadlines running from the date the conviction became final or from the discovery of new evidence. Others are more open-ended.
Preservation of the Evidence Itself
The biggest practical obstacle is often whether the biological evidence still exists. Samples can be lost, destroyed, or degraded during the years between trial and petition. Federal law requires the government to preserve biological evidence — sexual assault kits, blood, saliva, hair, and skin tissue — for as long as a defendant remains imprisoned for a federal offense. That protection kicks in only after the defendant is notified and given 180 days to file a testing motion once the conviction is final and direct appeals are exhausted. Miss that window, and the evidence can legally be destroyed. State preservation laws vary widely, and gaps in them have cost some defendants their only route to proving innocence.
Other obstacles are more ordinary but no less real. Many people petition from prison without a lawyer. The motion has to explain what should be tested, what result is expected, and how that result would show innocence. Even when a court grants testing, processing can add months or years.
Database Matches Through CODIS
The Combined DNA Index System, or CODIS, is a national DNA database run by the FBI that lets federal, state, and local labs compare profiles electronically. Crime scene profiles are searched against profiles from convicted offenders and other unsolved scenes. When a search hits on someone other than the person originally convicted, the case can reopen and the investigation can turn toward the actual perpetrator.
Database matches do two things at once. If DNA from a 1995 sexual assault matches an offender who entered the database years later for a different crime, that hit points at the true perpetrator and shows that whoever was convicted for the 1995 offense was wrongly identified. Cold cases where the original investigation lacked DNA technology, or where samples were too small to work with at the time, have been especially responsive to this kind of search as processing methods have improved.
Where DNA Evidence Falls Short
DNA is powerful, not infallible. Two limitations matter most for anyone leaning on it to prove innocence.
Secondary Transfer
Your DNA can end up somewhere you have never been. Secondary transfer happens when genetic material moves from one person to an object through an intermediate carrier — another person’s hands, clothing, or equipment. In one well-documented case, Lukis Anderson was charged with murder after his DNA was found under the victim’s fingernails. Investigators later determined that paramedics who had treated Anderson for alcohol intoxication earlier that evening unknowingly carried his DNA to the victim’s home on a later call.
The effect is easy to reproduce. In a study where two people shook hands for two minutes and then one of them handled a knife, the non-handler’s DNA turned up on the knife 85 percent of the time. In a fifth of those samples, the person who never touched the knife appeared as the primary or sole contributor. As forensic technology gets more sensitive, the risk of contamination through casual transfer goes up, not down.
Mixed Profiles
When a sample contains DNA from more than one person, interpretation gets harder. Analysts have to separate overlapping signals and decide how many contributors are present. Low DNA quantities, uneven contributor ratios, and artifacts from the amplification step all introduce ambiguity. In one proficiency test, labs were given a DNA mixture from a ski mask and asked whether a specific profile was present. Seventy-four of 108 participating labs incorrectly included a person whose DNA had never touched the mask. Human interpretation can shape what is often presented as purely objective science.
These limits cut in both directions. Secondary transfer can place an innocent person’s DNA at a crime scene. A mixed or degraded sample can rob a wrongly convicted person of the clean exclusion they need to challenge a verdict. DNA evidence does its best work as one part of a careful investigation, not as a standalone verdict, and its answers are only as reliable as the collection, analysis, and interpretation behind them.