DNA Evidence in Forensic Science

How Forensic DNA Profiling Works

Forensic DNA profiling exploits the fact that while 99.9% of human DNA is identical across all people, the remaining 0.1% contains regions of extraordinary variability. These variable regions, called short tandem repeats (STRs), consist of sequences 2 to 6 nucleotides long that repeat in tandem a variable number of times. At a given STR locus, one person might have 8 repeats while another has 14. By measuring the repeat count at 20 or more independent loci, forensic analysts generate a genetic profile that is effectively unique to each individual.

The current standard in the United States requires analysis of at least 20 core STR loci, established by the FBI's CODIS (Combined DNA Index System) in 2017 when it expanded from the previous 13-locus standard. European systems use a compatible but overlapping set of loci defined by the European Standard Set (ESS). The expansion to 20+ loci dramatically increased discrimination power and enabled international database comparisons.

Every person inherits one copy of each STR from their mother and one from their father, producing a pair of alleles at each locus. A person might be 12,15 at one locus (12 repeats from one parent, 15 from the other), 8,8 at another (homozygous, receiving the same allele from both parents), and 28,31 at a third. The complete set of allele pairs across all tested loci constitutes the individual's DNA profile.

The Laboratory Process

Extraction isolates DNA from the biological material. Different sample types require different extraction methods. Blood and saliva use standard organic extraction or silica-based commercial kits. Bone and teeth require grinding and extended digestion with proteinase K. Sexual assault evidence uses differential extraction to separate sperm cell DNA from epithelial cell DNA, a critical technique for identifying the assailant's profile separately from the victim's.

Quantification measures how much human DNA was recovered and assesses its quality. Real-time PCR (quantitative PCR) uses fluorescent probes specific to human DNA, determining the concentration in nanograms per microliter. Degradation is assessed by comparing amplification of short versus long DNA targets. If the long target amplifies poorly compared to the short target, the DNA is fragmented and the analyst adjusts the subsequent steps accordingly.

Amplification uses PCR (polymerase chain reaction) to copy the STR regions billions of times. Commercial STR amplification kits (GlobalFiler, PowerPlex Fusion, Investigator 24plex) contain primers for all required loci, each labeled with a different fluorescent dye. The thermal cycler alternates between heating (to separate DNA strands) and cooling (to allow primers to bind and polymerase to copy) for 28 to 30 cycles, doubling the target DNA with each cycle.

Capillary Electrophoresis separates the amplified STR fragments by size. Each fragment migrates through a thin glass capillary filled with polymer under an electric field. Smaller fragments travel faster. A laser at the end of the capillary excites the fluorescent dyes, and a camera records the emission. The resulting electropherogram displays peaks at positions corresponding to the size (repeat count) of each allele at each locus.

Interpretation is where the analyst's expertise matters most. Clean single-source profiles from abundant, high-quality DNA are straightforward to interpret. But many forensic samples are challenging: mixtures of DNA from multiple contributors, degraded DNA with missing alleles, or trace amounts with stochastic effects (random variation in which alleles amplify). Probabilistic genotyping software like STRmix and TrueAllele uses mathematical models to deconvolute complex mixtures, calculating likelihood ratios that quantify how much more likely the evidence is if the suspect contributed versus if an unknown person contributed.

Sources of Forensic DNA

Blood is the most common source of forensic DNA. A single drop of blood contains roughly 30,000 white blood cells, each containing a full copy of the individual's genome. Red blood cells do not contain DNA because they shed their nuclei during maturation. Even dried bloodstains years old can yield usable DNA profiles if stored properly away from heat, humidity, and ultraviolet light.

Saliva contains epithelial cells shed from the mouth lining. Cigarette butts, drinking containers, envelope flaps, stamps, bite marks, and breath alcohol mouthpieces all yield salivary DNA. A single cigarette butt typically provides enough DNA for a full profile.

Semen is critical evidence in sexual assault cases. Sperm cells contain dense, tightly packed DNA that is resistant to degradation. The differential extraction technique, which exploits the resistance of sperm heads to standard lysis conditions, separates male DNA from female epithelial DNA in mixed samples. Azoospermic males (those without sperm cells) still deposit epithelial cells and seminal fluid containing male DNA.

Touch DNA (Trace DNA) refers to the small quantities of DNA left by skin contact. Handling an object, gripping a weapon, or touching a surface deposits shed skin cells containing 5 to 50 picograms of DNA per cell. Modern techniques can generate profiles from as few as 5 to 20 cells, but touch DNA presents interpretive challenges because the DNA may have been deposited through direct contact, secondary transfer (a handshake followed by touching an object), or even tertiary transfer.

Hair varies in its forensic DNA value. Hair with an attached root (anagen phase hair, pulled rather than shed) contains nuclear DNA suitable for STR profiling. Shed hair without a root (telogen phase) contains only mitochondrial DNA, which provides less discrimination because it is inherited maternally and shared among maternal relatives.

DNA Databases and CODIS

The Combined DNA Index System (CODIS) is the FBI's national DNA database, operational since 1998. It contains three tiers: the Local DNA Index System (LDIS), the State DNA Index System (SDIS), and the National DNA Index System (NDIS). As of 2026, NDIS contains over 22 million offender profiles, 5 million arrestee profiles, and over 1.3 million forensic profiles from unsolved cases.

When a laboratory uploads a crime scene profile to CODIS, the system automatically searches for matches against offender and arrestee profiles. A "hit" occurs when a crime scene profile matches a known individual, generating an investigative lead. CODIS has produced over 750,000 hits since its inception, solving cases ranging from property crimes to serial homicides.

Forensic-to-forensic searching links unsolved cases to each other. If blood found at a burglary in Chicago matches semen from an assault in Detroit, investigators know the same person committed both crimes even before identifying a suspect. These links are particularly valuable for identifying serial offenders operating across jurisdictions.

Database eligibility laws vary by state. All 50 states collect DNA from convicted felons. Most also collect from individuals arrested for felonies, though some require a judicial finding of probable cause before the sample is analyzed. The Supreme Court upheld arrestee DNA collection in Maryland v. King (2013), comparing it to fingerprinting at booking.

Genetic Genealogy

Investigative genetic genealogy (IGG) identifies suspects and victims whose DNA is not in criminal databases by searching public genealogy databases. The technique gained national attention in 2018 when it identified Joseph James DeAngelo as the Golden State Killer, a serial murderer and rapist who had evaded capture for over 40 years.

The process begins with SNP (single nucleotide polymorphism) genotyping of the crime scene DNA, analyzing hundreds of thousands of genetic markers rather than the 20+ STR loci used for standard forensic profiling. This SNP profile is uploaded to a genealogy database (currently GEDmatch is used most frequently for law enforcement cases). The system identifies individuals who share significant DNA segments with the crime scene sample, indicating a biological relationship.

Genealogists then build family trees for the identified relatives, working backward through generations and forward to living descendants. By cross-referencing genealogical records with investigative information (age, sex, geographic location), they narrow the candidate list to a small number of individuals. Investigators then confirm or eliminate candidates through traditional DNA comparison, typically collecting a "discard" sample (a coffee cup, cigarette butt, or tissue discarded in public) for standard STR profiling.

Since 2018, genetic genealogy has identified suspects or victims in over 500 cases across the United States, predominantly cold cases that had exhausted all other investigative leads. The technique raises significant privacy and ethical concerns because it relies on DNA voluntarily submitted by private individuals for genealogical purposes, not law enforcement. The Department of Justice issued interim guidelines in 2019 requiring that IGG be used only for violent crimes and unidentified remains, and only after other investigative methods have been exhausted.

Challenges and Limitations

Degradation breaks DNA into smaller fragments over time, especially in hot, humid, or UV-exposed conditions. Degraded DNA may produce partial profiles with missing alleles at larger STR loci. Mini-STR analysis, which uses shorter amplification targets, can recover information from degraded samples that fail with standard kits. Next-generation sequencing (NGS) offers even greater sensitivity for degraded samples.

Mixtures containing DNA from multiple contributors complicate interpretation. A doorknob may carry DNA from dozens of people. Sexual assault evidence routinely contains DNA from both victim and assailant. Probabilistic genotyping software has dramatically improved the ability to deconvolute mixtures, but complex mixtures with more than three or four contributors remain extremely challenging.

Transfer and Persistence issues mean that finding someone's DNA on an object does not prove direct contact. DNA can transfer from person A to person B through a handshake, and then from person B to an object. Activity-level reporting, which considers how DNA may have been deposited rather than simply whether it is present, is an evolving area of forensic interpretation.

Identical Twins share the same STR profile because they develop from a single fertilized egg. Standard forensic DNA testing cannot distinguish between identical twins. However, whole-genome sequencing can sometimes identify somatic mutations (mutations that occur after the twins separate) unique to one twin, though this approach is expensive and not yet routine.