[Expert Explanation] Why Is Paternity Testing After a Baby Is Born Cheaper?

Rewritten: January 21, 2026

There are two types of paternity testing: postnatal and prenatal. The price difference stems from the different DNA markers used—STR versus SNP—and the difficulty of the analysis technology involved. In this article, experts explain in detail, from a molecular genetics perspective, the principles behind each method and why their costs differ.

There are two methods of paternity testing: "testing performed after the baby is born" and "prenatal fetal DNA testing performed before birth." Many people wonder why postnatal paternity testing is relatively inexpensive. Looking at typical market rates, postnatal DNA paternity testing is often performed for around 25,000 yen, while prenatal fetal DNA testing costs around 100,000 to 250,000 yen—a substantial difference. This price gap can be four to ten times, which may come as a surprise to those considering testing for the first time. However, this difference is not simply a matter of pricing policy; there are scientific reasons rooted in the principles of paternity testing, the DNA markers used, and the difficulty of the analysis technology. Between postnatal and prenatal testing, the state of the DNA being analyzed, the analytical equipment used, and the computational processing power required are all fundamentally different. This article explains in detail, from a molecular genetics perspective, the DNA analysis technologies used in both postnatal paternity testing and prenatal fetal DNA testing, and clarifies why such a large cost difference arises. Our goal is to help those considering paternity testing correctly understand the basis for the cost and choose a test with confidence. [ref:1]

What Are the Basic Principles of Paternity Testing?

Paternity testing is based on the fundamental principle of heredity (Mendel's laws): "half of a child's DNA comes from the father, and the other half from the mother." Human somatic cells contain 46 chromosomes, 23 of which are inherited from the father and the remaining 23 from the mother. Paternity testing centers on statistically evaluating whether the candidate father's DNA information is contained within the child's DNA sequence, using this principle. Specifically, during the cell division process called meiosis, sperm and eggs each become haploid, possessing only 23 chromosomes. Upon fertilization, these combine to form a diploid individual with 46 chromosomes again. Through this mechanism, a child always inherits half of its genetic information from each of its father and mother. [ref:5] The test statistically evaluates whether the child's DNA sequence is genetically consistent with the DNA of the presumed parents. Specifically, an index called the "Paternity Index (PI)" is used to quantify the probability that the subject is the biological father. Generally, if the paternity probability is 99.99% or higher, the relationship is determined to be "confirmed paternity," and if inconsistencies are found across all loci, it is concluded to be "excluded." Calculating the paternity index requires allele frequency data at each genetic locus. Based on the frequency of each allele in the relevant population, the ratio (likelihood ratio) between the probability of a coincidental match and the probability of a match based on an actual parent-child relationship is calculated, and by combining this across multiple loci, the final probability of the parent-child relationship is derived. [ref:2] What matters here is "which DNA regions are being compared," and the type of DNA marker used differs fundamentally between postnatal and prenatal testing.

What Is "STR," Used in Postnatal Paternity Testing?

The DNA marker primarily used in postnatal paternity testing is called STR (Short Tandem Repeat). STR refers to regions where a short DNA sequence of about 2 to 6 bases, such as "AGAT" or "TCAT," is repeated a certain number of times at a specific locus, and this repeat count varies between individuals. For example, at a given locus, Person A might have "AGAT" repeated 12 times while Person B has it repeated 15 times—each person showing a different pattern. STR markers are believed to exist at hundreds of thousands of locations across the human genome, and in the fields of forensic science and DNA paternity testing, loci that are particularly polymorphic and offer excellent discriminatory power between populations are selected for use. These loci have been standardized by the international forensic science community and are used as a common foundation by testing institutions worldwide. [ref:2] STR has the following characteristics.

• Large individual variation and high discriminatory power
• Inherited regularly from parent to child (codominant inheritance)
• High discriminatory power even with a small number (about 16 to 24 loci)
• Can be easily amplified by PCR (polymerase chain reaction)
• Analysis methods are internationally standardized
• Relatively low mutation rate, allowing stable inheritance across generations
• Diverse alleles exist, resulting in high heterozygosity

STR has therefore been used for many years in forensic science, criminal investigations, and paternity testing. STR markers are also adopted as the foundational technology in CODIS (Combined DNA Index System), managed by the FBI. CODIS originally used 13 STR loci, but was expanded to 20 loci in 2017, achieving even higher discriminatory power. [ref:6] The analysis methods and evaluation criteria are internationally established, enabling highly accurate determinations with a small number of markers. Multiplex kits that analyze around 20 STR loci simultaneously are now commercially available, allowing efficient results to be obtained in a single reaction. Representative kits include Applied Biosystems' GlobalFiler™ and Promega's PowerPlex® Fusion, which are widely used by forensic science laboratories and DNA testing institutions around the world. [ref:2]

Why STR Analysis Costs Can Be Kept Low

There are several clear reasons why STR analysis can be performed relatively inexpensively. First, the equipment required for STR analysis (such as capillary electrophoresis instruments) is already widely available at many testing institutions, keeping the running cost per sample low. Capillary electrophoresis instruments spread rapidly from the late 1990s onward and are now standard equipment at forensic science laboratories and genetic testing institutions worldwide. In addition, since samples are collected using oral mucosa swabs (rubbing a cotton swab against the inside of the cheek), collection is non-invasive and simple, requiring no special medical procedure. Because no physician needs to be present for collection, there are no coordination costs with medical institutions. Furthermore, the DNA obtained from oral swabs is of stable quality, resulting in a low analysis failure rate, which is also an economic advantage. Additionally, because the number of loci analyzed is limited to 16 to 24, the data processing burden is small, and automated analysis pipelines produce results in a short time. In modern STR analysis, the entire process—from PCR amplification through capillary electrophoresis to data analysis—is nearly fully automated, allowing a small number of skilled technicians to efficiently process many samples. Moreover, because reagent kits for STR analysis are mass-produced, the reagent cost per sample is kept to only a few thousand yen. The combination of these factors makes it possible to offer postnatal paternity testing at a relatively affordable price of around 25,000 yen.

\Resolve your paternity concerns in as little as 2 days/

What Is "SNP," Used in Prenatal Paternity Testing?

Prenatal fetal DNA testing, on the other hand, primarily uses SNP (Single Nucleotide Polymorphism) rather than STR. SNP refers to a position in the DNA sequence where only a single letter (one base) differs. For example, a variation in which most people have "A (adenine)" at a given locus while some people have "G (guanine)" is a SNP. There are believed to be approximately 4 to 5 million SNPs across the entire human genome, representing extremely rich diversity. Large-scale genome research projects such as the International HapMap Project and the 1000 Genomes Project have mapped the distribution and frequency of SNPs across the human genome in detail. [ref:5] Although the difference at each individual SNP is small, SNPs exist in extremely large numbers throughout the genome. While STR markers can have dozens of alleles at a single locus, SNPs typically have only two alleles (for example, A and G). Therefore, a single SNP alone does not provide enough information for individual identification or paternity determination. However, by analyzing many SNPs simultaneously and statistically integrating that information, it becomes possible to achieve discriminatory power comparable to or exceeding that of STR. Because prenatal testing requires analyzing the trace amounts of fetal-derived cfDNA (cell-free DNA) contained in maternal blood, judgments are made by statistically integrating information from hundreds to thousands of SNP sites. This method has developed rapidly since Dr. Lo and colleagues first reported the presence of fetal-derived cfDNA in maternal plasma in 1997. Their groundbreaking discovery demonstrated that fetal-derived DNA fragments circulate in the mother's peripheral blood, opening the way for non-invasive prenatal testing. [ref:3] [ref:7]

The Technical Hurdles of cfDNA Analysis

The proportion of fetal-derived cfDNA in maternal blood (the fetal fraction) varies depending on gestational age and individual differences, but it is generally only about 5 to 20% of the total maternal cfDNA—an extremely small amount. In other words, most of the free DNA in the blood is the mother's own, and the small amount of fetal DNA information must be accurately identified and analyzed from within it. The fetal fraction tends to increase as pregnancy progresses, but it also varies depending on the mother's BMI (body mass index) and placental condition, among other factors. Multiple studies have reported that pregnant women with higher BMI tend to have a lower fetal fraction, which is thought to be because increased cfDNA released from maternal adipose tissue relatively lowers the proportion of fetal-derived cfDNA. To accurately read such trace amounts of DNA, a high-performance genetic analysis device called a next-generation sequencer (NGS) is required. NGS can simultaneously read hundreds of thousands to hundreds of millions of DNA fragments at once, making it especially well-suited to analyzing fragmented DNA such as cfDNA. cfDNA circulates in the blood as extremely short fragments averaging about 166 base pairs (bp), which made efficient analysis difficult with conventional Sanger sequencing (first-generation sequencing). NGS's massively parallel sequencing technology has, for the first time, made it possible to read large volumes of such tiny fragments. However, NGS-based analysis requires expensive reagents and consumables, and a single analysis can take several days. For example, with Illumina's sequencing platforms—a representative NGS platform—the cost of the reagent kit for a single run can reach hundreds of thousands to millions of yen. In addition, powerful computers and specialized bioinformatics expertise are essential to process the resulting massive volume of sequence data. A single NGS run can generate several hundred gigabytes of data, and processing this data in real time and accurately mapping and analyzing it requires dedicated server infrastructure and sophisticated algorithms. These technical requirements are a major factor driving up the cost of prenatal testing.

\Find out who the father is while the baby is still in the womb/

Why Is Prenatal Testing So Expensive?

The high cost of prenatal testing stems from the following complex technical challenges.

1. Fetal DNA is mixed with maternal DNA, making separation difficult. 80-95% of the cfDNA in maternal plasma is maternal in origin, and the small remaining amount of fetal-derived cfDNA must be accurately identified.
2. The amount of fetal DNA is extremely small, requiring highly sensitive detection technology. Ordinary PCR amplification alone is insufficient, and advanced methods such as target enrichment or digital PCR may be required.
3. Advanced sequencing technology is required to simultaneously analyze hundreds to thousands of SNP sites. This requires the use of an NGS platform along with appropriate library preparation protocols.
4. Bioinformatics technology and quality control to process the enormous volume of analysis data are essential. Multiple stages of computational processing—raw data filtering, alignment, variant calling, and statistical analysis—are required.
5. Because the test involves the medical procedure of drawing maternal blood, coordination with a medical institution is necessary. Blood collection must be performed by a physician or nurse, incurring management costs associated with partnering with a medical institution.

This technology originally developed in the field of NIPT (Non-Invasive Prenatal Testing), and requires highly accurate sequencing technology and computational processing. A large-scale study reported by Bianchi et al. in 2014 in the New England Journal of Medicine demonstrated that prenatal testing using cfDNA showed higher sensitivity and specificity than conventional screening tests. This study was a landmark achievement in establishing the clinical utility of NIPT technology and triggered its subsequent worldwide adoption. [ref:5] Prenatal paternity testing (NIPPT: Non-Invasive Prenatal Paternity Testing) applies this NIPT technology, and unlike chromosomal abnormality screening, it adds the additional analytical step of matching against the candidate father's DNA information, requiring even more advanced technical capability. While NIPT detects abnormalities in fetal chromosome copy number, NIPPT requires a more complex computational process: matching fetal-derived SNP information within maternal cfDNA against the candidate father's SNP profile and statistically evaluating the parent-child relationship. [ref:4]

Cost Comparison Between Postnatal and Prenatal Testing

The main differences between postnatal and prenatal testing are summarized below.

As shown, the cost difference is directly tied to the maturity of the technology, the complexity of the analysis, and the cost of the required equipment. STR analysis, used in postnatal testing, is a mature technology with decades of history, while cfDNA-SNP analysis, used in prenatal testing, is a relatively new technology, and the costs involved in performing the test are fundamentally different. To add further detail, postnatal testing can be completed with standard multiplex PCR kits and a capillary electrophoresis instrument, so initial investment is relatively small and the running costs for testing institutions remain low. Prenatal testing, on the other hand, requires an initial investment on the scale of tens of millions of yen to introduce an NGS platform, and ongoing maintenance fees, software license fees, and data storage costs are also incurred continuously. Because these infrastructure costs are spread across each test performed, the result is a higher price for prenatal testing.

A Deeper Look at the Molecular Biological Differences Between STR and SNP

To gain a deeper understanding of the differences between STR and SNP, let's take a closer look at each one's characteristics from a molecular biology perspective. STR (short tandem repeat), also called microsatellites, is found predominantly in non-coding regions of the genome (regions that do not code for proteins). Variation in STR repeat count is believed to arise through a phenomenon called slippage during DNA replication. When DNA replication enzymes (DNA polymerase) replicate a repeated sequence, the newly synthesized strand or the template strand can form a loop structure, resulting in an increase or decrease in the repeat count. Because this phenomenon occurs at a low frequency in each generation, diverse allele patterns have accumulated within populations over time. [ref:6] SNPs (single nucleotide polymorphisms), on the other hand, are distributed uniformly across the genome, occurring in both coding regions (exons) and non-coding regions (introns, intergenic regions). SNPs mainly arise from base substitution errors during DNA replication, repair errors following base damage from chemicals or ultraviolet light, and deamination reactions. In particular, deamination following methylation of cytosine at CpG dinucleotides (consecutive cytosine-guanine sequences) is known to be a major cause of C-to-T mutations. [ref:5] The main reason SNPs, rather than STRs, are used in prenatal testing lies in the physical characteristics of cfDNA. As noted above, cfDNA circulates in the blood as extremely short fragments averaging about 166 bp. This length corresponds roughly to the length of DNA wrapped around one nucleosome unit (a structure in which DNA is wound around histone proteins), a result of DNA being cleaved into nucleosome units during the process of cell apoptosis (programmed cell death). Because most STR markers are designed as PCR products of about 100 to 400 bp, there is a high likelihood that the target STR region will not be fully contained within a short cfDNA fragment of around 166 bp. SNPs, by contrast, involve a variation at just a single base, so information can be reliably read even from very short DNA fragments. This is the scientific basis for choosing SNPs in cfDNA analysis.

The Statistical Foundation Supporting the Accuracy of Paternity Testing

The results of paternity testing cannot be obtained simply by checking whether DNA sequences match or not. Determining the result requires rigorous mathematical models based on population genetics and probability and statistics. In STR analysis used for postnatal testing, a "Paternity Index (PI)" is calculated individually for each locus. PI is the value obtained by dividing "the probability of observing this STR pattern if the test subject is the true father" by "the probability of observing this pattern if the test subject is not the father (i.e., the probability that an unrelated man would randomly show this pattern)." This calculation references an allele frequency database for the relevant population. The value obtained by multiplying the PI across all loci is the "Combined Paternity Index (CPI)," and converting this into a posterior probability using Bayes' theorem yields the final "paternity probability." Generally, assuming a prior probability of 0.5 (50%), a CPI of 10,000 or higher corresponds to a paternity probability of 99.99% or higher, resulting in a determination of "confirmed paternity." [ref:2] Prenatal testing requires an even more complex statistical model. Because the cfDNA in maternal plasma is a mixture of maternal and fetal origin, the fetal genotype at each SNP locus cannot be directly observed. Therefore, advanced statistical methods such as Maximum Likelihood Estimation and Hidden Markov Models are used, combining the estimated fetal fraction, the maternal genotype, the candidate father's genotype, and the population allele frequency of each SNP, to select the most probable hypothesis regarding the parent-child relationship. Accurately carrying out this kind of complex statistical processing requires powerful computing infrastructure and specialized bioinformatics pipelines, which is another factor driving up the cost of prenatal testing. [ref:7]

Summary

The reason postnatal paternity testing can be offered inexpensively is that it uses the mature DNA marker STR, and the target of analysis is well-defined. STR analysis has decades of track record, low equipment and reagent costs, and can achieve accuracy of 99.99% or higher by examining just a small number of genetic loci. The analysis process is highly automated, with the entire workflow, from sample collection to reporting of results, designed for efficiency. Prenatal fetal DNA testing, on the other hand, is a medical-grade test requiring advanced analysis using large numbers of SNPs, resulting in higher costs. Because accurate determination is only possible by reading the small amount of fetal cfDNA in maternal blood using a next-generation sequencer and performing advanced bioinformatics analysis, the technical costs are reflected in the price. To summarize once more the main factors behind the cost difference between the two:

• Differences in the type of DNA marker used (STR vs. SNP) and the number of loci that need to be analyzed
• Differences in the cost of analytical equipment (capillary electrophoresis vs. next-generation sequencer)
• Differences in sample type and collection method (non-medical procedure vs. blood draw as a medical procedure)
• Differences in the complexity of data processing and the scale of computing infrastructure required
• Differences in the maturity of the technology and market scale (standardized general-purpose technology vs. cutting-edge specialized technology)

What's important is that a higher cost does not mean you are "losing out." The cost of prenatal testing includes cutting-edge technology and rigorous quality control that guarantee highly accurate results. These technical investments are essential to achieving highly accurate, non-invasive paternity determination while ensuring the safety of both mother and fetus. At the seeDNA Genetic Medical Research Institute, we value creating an environment in which people can choose their test with confidence by accurately conveying this technical background. Whether it is prenatal or postnatal testing, we provide reliable results based on scientific evidence. If you have any questions or concerns, please feel free to contact our specialized staff.

Frequently Asked Questions

Q1. Why is postnatal paternity testing about 25,000 yen, and so inexpensive?

A. Postnatal paternity testing uses STR (short tandem repeat), a long-established DNA marker. The number of loci that need to be analyzed is small, around 16 to 24, and automated analysis using capillary electrophoresis instruments is possible. Because samples can also be easily collected using oral mucosa swabs, costs can be kept low from both a technical and operational standpoint. Mass production of multiplex PCR kits also keeps reagent costs low, and the combination of these factors makes it possible to offer testing at an affordable price.

Q2. Why does prenatal paternity testing cost 100,000 to 250,000 yen?

A. Prenatal testing requires analyzing the extremely small amount of fetal-derived cfDNA (cell-free DNA) contained in maternal blood using a next-generation sequencer. Because hundreds to thousands of SNPs are examined simultaneously and advanced bioinformatics analysis is performed, all of the reagent, equipment, and labor costs are higher than for postnatal testing. The initial cost of introducing an NGS platform is on the scale of tens of millions of yen, and reagent kits alone can cost several hundred thousand yen or more per run. This technical cost is reflected in the price.

Q3. What is the difference between STR and SNP?

A. STR is a marker that looks at "differences in the number of repeats of a short base sequence," showing large individual variation and high discriminatory power even with a small number of loci. A single locus can have dozens of different alleles. SNP, on the other hand, is a marker that looks at "a difference of a single base within the DNA sequence." While the difference at each individual site is small, SNPs exist at millions of locations across the entire genome, so combining many of them enables highly accurate analysis. Because cfDNA in prenatal testing consists of extremely short fragments of about 166 bp, SNPs, which allow information to be read from short sequences, are better suited.

Q4. From what week of pregnancy can prenatal paternity testing be performed?

A. Generally, testing is considered feasible from around 7 to 9 weeks of pregnancy onward, once the concentration of fetal cfDNA in maternal blood has risen sufficiently. However, if the fetal fraction is low, an accurate determination can be difficult, so we recommend checking the gestational age recommended by the testing institution. The fetal fraction increases as pregnancy progresses, but it also varies with individual differences such as the mother's BMI. At the seeDNA Genetic Medical Research Institute, we provide guidance on the optimal testing timing through prior consultation.

Q5. Is prenatal paternity testing safe? Is there any risk to the baby?

A. Prenatal paternity testing (NIPPT) is a non-invasive test that only requires drawing blood from the mother. Unlike conventional amniocentesis or chorionic villus sampling, which require inserting a needle into the uterus, there is no direct risk to the fetus, such as a risk of miscarriage. The test can be performed with the same level of burden as an ordinary blood draw. This "non-invasiveness" is the greatest advantage of prenatal testing using cfDNA.

Q6. Is there a difference in accuracy between postnatal and prenatal paternity testing?

A. Both tests are capable of determining the parent-child relationship with an accuracy of 99.99% or higher. However, because the analytical approaches differ, the technical mechanisms are quite different. Postnatal testing involves direct comparison of STRs and calculation of the combined paternity index, while prenatal testing involves statistical integration of SNPs and determination via maximum likelihood estimation—but the final accuracy achieved is equally high in both cases.

Q7. What kind of device is the next-generation sequencer (NGS) used in prenatal testing?

A. A next-generation sequencer (NGS) is a high-performance genetic analysis device capable of reading hundreds of thousands to hundreds of millions of DNA fragments simultaneously and in parallel. Whereas conventional Sanger sequencing could only read one DNA fragment at a time, NGS uses massively parallel processing to acquire large volumes of DNA sequence information in a short time. In prenatal testing, this massively parallel reading capability is essential for detecting fetal-derived SNP information from the trace amounts of cfDNA in maternal plasma.

[References]