Request Information | Testimonials | Family Law Directory | Links | Careers | About Us | Contact Genelex
On-line Ordering Now Available - Click Here
Cover
Preface
Introduction
Chapter One:
The DNA WARS SCIENCE MEETS THE LAW
Chapter Two:
DNA AS EVIDENCE
Chapter Three:
DNA IN THE LABORATORY
Chapter Four:
DNA IN PARENTAGE TESTING
Chapter Five:
DNA IN THE COURTROOM
Chapter Six:
PEOPLE V. ORENTHAL JAMES SIMPSON
AFTERWORD
Appendix:
DNA TESTING IN 50 STATES
Glossary
Recommended Resources
Copyright
 |
 |
 |
 |
 |
|
There are about 100 trillion cells in the adult human body. Most of them have a nucleus, or center, that contains thread-like bundles of chromosomes. In these chromosomes are all of the instructions and information needed to make a human being. Each parent contributes one chromosome to each of the 23 pairs found in all normal people. Within the chromosomes, are up to 100,000 paired genes, the fundamental units of heredity. Each gene can have different versions (as many as 100 or more in rare cases) called alleles, but most are the same from person to person. Genes determine all inherited traits including those that give the individual specific characteristics (blue eyes rather than brown eyes) as well as common characteristics (two eyes, two arms, etc.). Genes are made of deoxyribonucleic acid (DNA). Hence, DNA is the master molecule of life and controls the growth and development of every living thing. It is a polymer, i.e., a long string of simple repeating units. These repeating units are called nucleotides and are of four types: adenine (A), cytosine (C), guanine (G), and thymine (T). Just as the order of the letters of the alphabet determines the information content of words, the order in which these four bases are strung together is what gives DNA its information content. The complete DNA molecule consists of two of these strands of the four bases. In the two strands, A always is across from or paired with T and G always is paired with C. These are the base pairs that are the unit of measurement in determining the size of a given segment of DNA. This structure suggests a natural mechanism for the duplication or replication of the DNA molecule, as occurs during cell division. These pairings are what connect the two strands of DNA together to form a tightly coiled, twisted ladder. This spiral staircase, the famous double helix, is the natural form in which DNA is found within the nucleus of the cells. If uncoiled, the DNA molecules in every human cell would measure six feet in length. That is the total length of the 3.3 billion base pairs that make up the total human genetic complement or genome. Except for identical twins, the sequence of the base pairs within the DNA helix is unique for every person, and forms the individual's genetic code or blueprint. Perhaps the basis of DNA typing can be best understood by comparing the way in which genetic information is stored in the DNA to the way in which printed information is stored in books. For example, if we were to cut all the sentences in forty volumes of the Encyclopedia Britannica into strips, and tape them together end to end, then we would have an amount of information equivalent to that contained in the DNA within each of the cells that make up our bodies. Furthermore, the information would then be in the same physical form as the DNA information, i.e., a long linear strip sometimes likened to a computer punch tape. The genetic information contained in the DNA is organized and packaged into chromosomes, much as printed information is organized into volumes. Just as a specific passage in the encyclopedia can be identified by specifying a volume, page, and line number, a specific genetic passage or location, known as a locus, can be identified. A specific naming system identifies genes by numbers issued by the Human Gene Mapping Committee. For example, if we see the designa- tion D4S139 in a report, then we know exactly what gene has been analyzed, that it is on chromosome four, and that it is the 139th DNA probe to be mapped to chromosome four. A significant difference between the way information is stored in the cell and in the encyclopedia is that there are two copies of the information in each of the cells, one from the mother and one from the father. These two copies of the genetic information which are largely identical, come together at the moment of conception when the sperm and the egg join together. All of the child's cells contain DNA derived from this original fertilized cell, half from the mother and half from the father. It is this basic principle of heredity, first discovered over 150 years ago, that allows us to reliably perform parentage tests.
The DNA and hence the genetic code of humans is almost the same for all individuals. It is the very small amount that differs from person to person that forensic scientists analyze to identify people. These differences are called polymorphisms (from the Greek for "many forms") and are the key to DNA typing. Two major kinds of polymorphisms are most useful to forensic scientists. The first consists of variations in the length of the DNA at specific locations (loci) known as VNTRs (variable number of tandem repeats). These VNTR regions consist of stretches of DNA made up of short repeating DNA sequences. The number of times the short DNA sequence is repeated determines the physical length of the DNA molecule at these specific loci. Each of the many versions that may be found in the population are known as alleles. These variations are examined by the method known as the RFLP technique. The second type of variation in the DNA is simply a difference in the nucleotide letters found at a specific pair of bases. These are examined by the method known as PCR. The goals are the same for both types of DNA tests: to isolate a distinctive DNA sequence and record its presence in a way that can be examined visually. The basic procedure of RFLP analysis is known as "Southern blotting" after Edward Southern, a Scottish bio-chemist who developed the technique in the early 1970s. The basic procedures of PCR based testing were invented in the early 1980s by Kary Mullis while working for a California biotechnology company. Deciding which method to use is determined by the amount of DNA that is available and how deteriorated or degraded it is. RFLP analysis is more often performed because it is more discriminating. If there is only a small amount of material or it is highly degraded, PCR analysis is used, because this method requires less material and can produce a result on DNA of poor quality. As little as two billionths of a gram (2 ng), the amount of DNA contained in about 700 sperm cells will suffice for PCR analysis. The RFLP test requires 20 to 50 ng of DNA. PCR tests can be performed in a matter of days as compared to weeks for RFLP tests.
Once the evidence has been documented and screened in the laboratory, and deemed appropriate for DNA analysis, the initial step is to isolate or purify the DNA. First, it must be removed from whatever object it is attached to and removed from the cell. Unless the major non-DNA constituents of the cell such as proteins, fats and carbohydrates are removed, the enzymes essential in the next step will not be able to do their work. This isolation of DNA begins with the controlled destruction of cellular integrity, releasing the DNA from its nuclear and chromosomal packaging. The cell walls are dissolved with a detergent and the proteins are digested by enzymes. The DNA then is purified by the methods of extraction and precipitation. Once the DNA is isolated and concentrated, a small sample is tested to determine quality and quantity. If intact (high molecular weight) human DNA is present in sufficient amounts, RFLP testing can proceed. If the DNA is degraded, or present in minute amounts, PCR testing is used.
RFLP typing of purified DNA consists of the following six steps:
Restriction enzymes reproducibly cut DNA at specific four or six-base-pair sequences called restriction sites. Hae III (Haemophilus Aegyptius III), the most commonly used enzyme in forensic science, cuts the DNA everywhere the bases are arranged in the sequence GGCC. These sites are found throughout the human genome and are, for the most part, the same in everyone. Hae III cuts human DNA into approximately 12 million different restriction fragments ranging in size from a few hundred to 10,000 or more base pairs in length.
The physical length of DNA restriction fragments in VNTRs is one fundamental biochemical characteristic that varies from person to person. Electrophoresis is the technique by which the different-sized fragments are separated. The DNA is loaded into a hole, or well, at one end of a slab of semi-solid gel, a porous, Jello-like substance. When an electrical field is applied to the gel containing the DNA, the DNA moves toward the positive electrode because it has a negative electrical charge. The sieving action of the gel allows the small restriction fragments to migrate at a faster rate than the larger ones, just as a small rabbit would move faster and farther through a thicket than a large hound in pursuit. The relative position of these fragments within the gel after overnight current application is determined by their length or molecular weight. Gels have several lanes within them which contain DNA, including marker lanes that measure how far fragments move through the gel and a lane for control DNA that produces a known pattern and can be used to verify that the test was properly conducted. The DNA in the gel can be stained and seen under ultraviolet (black) light.
Because the gel is fragile, it is necessary to remove the DNA from the gel and permanently attach it to a solid support. This is accomplished by the process of Southern blotting. The first step is to denature the DNA in the gel which means that the double-stranded restriction fragments are chemically separated into the single-stranded form. The DNA then is transferred by the process of blotting to a sheet of nylon. The nylon acts like an ink blotter and "blots" up the separated DNA fragments. The restriction fragments, invisible at this stage, are irreversibly attached to the positively-charged nylon membrane called the "blot."
Visual observation of an individual's DNA pattern requires the use of DNA probes. The DNA probe, like a guided missile, will seek out and find its target sequence. There are approximately a dozen VNTR DNA probes in common usage in forensic RFLP testing. These probes have been patented and are commercially available from biotechnology suppliers. Each DNA probe can be used to develop the DNA profile at a particular VNTR locus. Most of these are located on different chromosomes, an important factor to be considered when performing a statistical analysis. Most forensic tests use a combination of at least four to six separate DNA probes in a sequential manner. All of these DNA probes have been obtained by the process of cloning, which simply means that they are free of all other human DNA. Therefore, large quantities of probe DNA can be made and labeled with a radioactive or other tracer in the laboratory.
To detect a VNTR locus immobilized on the Southern blot, one uses a DNA probe that has a base pair sequence complimentary to the DNA sequence at the VNTR locus. The double-stranded nature of DNA and a phenomenon called hybridization provide the scientific basis for the usefulness of DNA probes. Double-stranded DNA fragments will separate when heated. When the DNA cools down, the two strands will reconnect and become double-stranded again. This is not a random process. Because of the complementarity of the two DNA strands, the single strands will only reconnect with another strand that has a complimentary sequence. Just prior to its use, the DNA probe is labeled with a radioactive or other tracer and boiled to the single stranded form. The DNA on the blot, also single stranded, is soaked or incubated in a solution containing the DNA probe. The probe hybridizes, or binds, to only the DNA fragments that bear the complimentary sequences of DNA bases. These will be the restriction fragments corresponding to the locus from which the probe was originally cloned.
The excess probe is washed off and the blot placed in contact with a sheet of film. The film, exposed by the radioactive tracer instead of light, is developed and becomes an autoradiograph, commonly known as an autorad. The autorad is the final product of the RFLP analysis. It reveals the overall quality of the testing and can be copied, distributed and interpreted by other DNA experts. The autorad has darkened areas known as bands corresponding to the position of the DNA probes and hence the restriction fragments bound to the membrane. Typically, several probes are used sequentially in order to compile forensically significant DNA profiles. This requires that the blot be stripped of the first radioactive probe, hybridized to a second, washed and exposed to a new film to make another autorad. Each round of hybridization and autoradiography may take up to ten days. This is why RFLP testing takes longer than conventional blood or PCR testing.
Up to this point in DNA analysis, there is little argument about its validity (provided that it is done correctly) because we are dealing with physical reality. The interpretation of the data, what is its meaning, is another story. This requires that inferences based on the science of statistics, population genetics and probability theory be applied to the measurements of physical reality that the autoRADS reveal. Forensic scientists use these mathematical concepts to calculate and report an estimate of how frequently the genetic profile they have observed might be found in major population groups. If the genetic profiles found in two different samples, say one from a piece of evidence and one from a suspect are indistinguishable, they are said to match. A typical population frequency for conventional blood typing might be 1 in 200, for DNA 1 in 5,000,000. This means that only 1 in 5,000,000 people would have the same DNA profile. All others would be excluded from being the source of the matching evidence. Irrespective of what calculation method is used, it is a physical fact that the genetic profiles match and that they would be found at some frequency in the population. In attempting to call the frequency estimates into question, attorneys are fond of pointing out that they are a comparison between the observed profile and randomly chosen individuals, and that a relative of the person with the profile is much more likely to share that profile than any of the random individuals. When forensic scientists provide a report and testimony about the frequency estimates, their job is done. The judgment of guilt or innocence reached by the court may take these estimates into account, but they must be placed into the circumstances surrounding the crime and the intent, motive, means and opportunity available to the defendant.
The meaning of the genetic profile frequencies are often misconstrued by attorneys. For example, the argument might be made that if a pair of matching genetic profiles are found in 1 in 200 individuals then there is a 1 in 200 chance that they came from different sources. Not true. What the chance is that they came from different sources cannot be determined by the genetic evidence alone. It depends on all of the circumstances surrounding the case. Another fallacy is the one heard in the Simpson case preliminary hearing. O,J. Simpson's blood type matched blood found on the sidewalk trailing away from the murder scene. Defense attorneys pointed out that 80,000 people in Los Angeles share that blood type. True, but all of those 80,000 people didn't visit 85 South Bundy where the homicides occurred. The prosecutor's fallacy is to argue that as the genetic profile is found in 1 in 1000 people then there is only a 1 in 1000 chance that the defendant is innocent. Once again this ignores the other facts of the case and asks the scientific evidence to determine guilt or innocence. Science cannot do that. Guilt or innocence must be determined by the judge or jury.
Critics of DNA testing have generated and seized upon disagreements about the best way to interpret the data as a reason not to admit DNA evidence. Forensic scientists themselves have given the critics ammunition by not always following widely used scientific conventions for rounding off and reporting the significance of the numbers that are reported. In evaluating the criticisms that are made, primarily of the validity of the population data on which probability estimates are made, it is important to remember that the critics:
Much of the population genetics controversy has been generated by an elementary misunderstanding of the basics of RFLP testing in the early writing of Eric Lander. Another source of misunderstanding was an article published in 1972 by Richard Lewontin and since repudiated by the author. Long after Lewontin changed his mind, the article continued to be used to support defense motions to quash DNA evidence. What this reveals is that one defense strategy is the shotgun approach; throw up a barrage of flak and hope that the judge or jury accepts the validity of at least one of the arguments. The critics also change their arguments over time. As various lines of attack are countered or exposed as spurious, new objections are raised. Frequency estimates are fertile territory for the attackers because most people are not conversant with the methods of statistics and the meaning of probability.
Before an opinion can be formed as to the forensic significance of a set of DNA profiles, a series of data interpretation steps must be completed.
In forensic science, statistics are used to ensure that the data is interpreted in a conservative fashion, that it can be guaranteed to be an overstatement of the true occurrence of a DNA profile in the population. For example, if we were to test everybody in the entire world then we would know exactly how often a given genetic profile occurs. Let's say that it is 1 in a million. If we report that profile out at 1 in 500,000, that is acceptable because it is an understatement. The data has been systematically skewed in favor of the defendant.
The purpose of the visual inspection is to form an opinion as to which of the DNA samples in the various lanes in the autorad could have come from the same individual source and which could not. Included in the autorad are bands from any DNA that was tested as well as control specimens from known individuals. In order to be valid, the opinion formed at this stage must be confirmed by the following quantitative tests.
Central to the analysis are the series of "sizing ladders" containing up to 30 closely spaced bands. Each of the bands in the sizing ladders corresponds to a fragment of DNA of exactly known size. These are the rulers which computerized video devices use to measure the DNA fragments in each of the sample lanes. The estimated size of the fragments being measured are reported in base pairs. At least one sample of a human DNA, known as K562, is run on virtually every membrane used in forensic casework. The K562 DNA is available as a standard reference material from the National Institute of Standards and Technology. Sizing estimates of the K562 must be within established tolerances before sizing estimates from that membrane are acceptable.
Comparisons are made between the DNA profiles of the known samples (generally blood samples taken from the suspect and/or victim) and DNA profiles of the evidence samples (for example, bloodstains from the crime scene or dried vaginal swabs). There are three possible results:
Interpretation of RFLP data has required the adoption of statistical techniques new to forensic serology because of the numeric nature of the data. In ABO and other typing systems, the data is a non-numeric listing of discrete types. In the RFLP technique, the sizing estimates form a continuous distribution because restriction fragments differing by one VNTR repeat unit cannot be distinguished. Therefore, it becomes necessary to perform statistical calculations to determine if two bands on an autorad can be distinguished from one another. For forensic purposes, two DNA profiles are said to match if they are statistically indistinguishable and are therefore consistent with having been produced by DNA from the same individual. First a quantitative match range for each appropriate VNTR band is calculated. Next the match ranges of corresponding bands are compared lane to lane. If these ranges overlap then the bands are indistinguishable from one another and are said to match. The process continues until all bands at a locus, and all bands at all the loci examined have been compared. In order for two composite DNA profiles to be declared a match and therefore included in the group of DNA profiles that could have come from an individual source, all bands must match. Most laboratories use a match range, or "window", on the order of + 2.5% of the measured size or molecular weight of the band. This match window has been determined by comparing the DNA profiles obtained from vaginal swabs and fresh blood from rape victims. When working with fresh blood samples, such as in a parentage laboratory, the measurement error is less than 1% of the measured size of the bands.
The relatively large size of the forensic match window compensates for the phenomenon known as "bandshifting". When DNA fragments are separated by electrophoresis, each sample is loaded into its own lane on the gel. Sometime |
||||
| << Chapter 2 | Chapter 4 >> | |||
|
||||
