DNA in the Courtroom

8/13/2019 DNA in the Courtroom

1/75

DNA in the Courtroom - Preface

PREFACE

This book was originally conceived as a fifty-page technical guide forreporters covering the Simpson trial. As we assembled our material andtalked with people, it became clear that just providing information onthe scientific aspects was not enough, and that more than a mediaguide was required. n order to understand what is happening with !"Ain the courtroom, it is necessary to e#amine the conte#t andenvironment in which science and the law meet.

The !"A war is a clash of the different approaches and interests of lawand science. t would be difficult to think of two disciplines that havemore disparate approaches to seeking the truth. Scientific truth

evolves by the building of consensus through peer review andreplicating e#periments. $or the law, truth is less absolute and morerelative to the case at hand. %egal truth is achieved through adversarialargument and judgement. The !"A war is a dramatic, hard foughtconflict between these two worlds.

n this short book we have tried to provide a brief overview of !"Aissues in advance of the Simpson trial. &e are simultaneously workingon a more comprehensive book, Silent &itness' !"A, (rime and )usticein America, which will tell the full story of the !"A war, the people whofight the battles, and the true crime stories that put a human face on

forensic science.

%et us emphasi*e at the start that we strongly believe that !"A testingproperly performed belongs in the courtroom. &e share this view withthe +.S. (ongress ffice of Technology Assessment, the "ationalAcademy of Sciences and the vast majority of scientists in America.The continuing controversy over the admissibility of !"A evidencemakes little sense e#cept to the very few people who benefit fromcontinued confusion.

INTRODUCTION

!"A was discovered in /, at about the same time that the Austrianmonk 0regor 1endel formulated the laws of heredity, which laid thefoundation for the modern science of genetics. A major breakthroughwhich eventually made possible forensic uses of !"A occurred in theearly 234s when )ames &atson and $rancis (rick solved the pu**le of!"A5s structure and precipitated the revolution in molecular biologywhich followed.


2/75

The 6io-Technology 7evolution plays a major and ever-e#panding rolein our life. 1edical genetics and gene therapy already have given ustremendous tools to diagnose and treat inherited diseases. &ith !"A,parentage testing has become a virtually certain procedure. &e use!"A to improve animal and vegetable foodstocks, protect endangered

species, verify animal pedigrees, detect war crimes and identify humanremains. 1any 8&-1A families have had the satisfaction of finallylearning the fate of their loved ones through !"A testing of bonesrecovered in 9iet "am. !"A testing was used to make sure that thecorrect remains were returned to families following the war with raq.After the inferno at the 6ranch !avidian comple# in &aco, Te#as, !"Awas the only way to confirm identities from many of the charredremains.

ur knowledge of the past is considerably strengthened byanthropological and paleontological !"A testing. &hile )urassic 8ark-

style reincarnations are fantasy, !"A has been e#tracted from suchremnants of ancient life as fossils, prehistoric bones, and mummies. nthe past year, Scotland :ard analy*ed a skull fragment and confirmedthe identity of 7ussia5s last c*ar. ronically, a few months later, acontemporary pretender to the 7omanoff name and fortune wasunmasked by !"A testing. &hile these are notable uses of the tests,many people believe !"A5s most fascinating use is in criminalidentification. (ertainly no methodology to emerge thus far from the6io-Technology 7evolution has had a greater impact on the public.

The basic techniques of !"A typing were developed and proven in

research and medical laboratories long before they were used in thecrime lab. 6ut as we try to reach a fair and just result in the trial of .).Simpson, where !"A evidence plays such a prominent role, and as weevaluate the use of !"A testing in the criminal justice system, it helpsto remember that the first use of !"A in an American courtroomoccurred only seven short years ago.

Since forensic !"A testing was introduced in 2/, it already has beenused in appro#imately 34,444 criminal investigations in the +.S. alone.!"A evidence is most often found in se#ual assaults. About three-fourths of the !"A evidence e#amined by the $6 and other labs

consists of semen specimens. About a third of all $6 tests e#oneratewrongfully accused men. A quarter of the tests are inconclusive, andabout ;4< match the suspect5s profile.

1ost of the !"A casework is performed in 34 or so federal, state andlocal police crime laboratories =appro#imately >44 are foundthroughout the +.S.?. As a result, !"A testing is often regarded as aprosecution tool. (ertainly, when !"A is used in court and reported in


3/75

the media it almost always is being used to prove the prosecutor5scase. @owever, it is important to remember that thousands of tests areperformed annually which eliminate innocent persons as suspects.

There also are now almost a score of cases in which !"A testing has

been used to free previously convicted and incarcerated individuals,some of whom have been in prison for up to twelve years. Anotheradvantage of !"A typing is that it leads to considerable savings in thecriminal justice system because suspects confronted with !"Aevidence often will plead guilty, thereby saving the considerable costsof a trial.

!"A evidence is a powerful tool for justice to be used by both theprosecution and the defense to benefit us all.

Chapter 1:

The DNA Wars: cience !eets the "a#

The publication of this book coincides with the beginning of the trial ofthe century. The 8eople of (alifornia versus renthal )ames Simpsoncombines the celebrity and media circus aspects of the %indberghkidnapping case with the science under attack scenario of the Scopes1onkey trial over the teaching of evolution in Tennessee public

schools. The celebrity element in the Simpson case is especiallyfascinating because it involves the suspect instead of the victim. Therehas never been any trial in history that cuts across as many issues. tcovers se# and gender, fame and the media, public opinion and thejury system, race and violence, entertainment and commerciali*ation,even of murder.

6efore the start of the actual trial, Simpson5s lawyers are putting !"Atesting on trial. f !"A loses everyone loses, because forensic !"Atesting is such a revolutionary improvement to fairness in the criminaljustice system. very year hundreds of thousands of !"A tests are

performed in both forensic and paternity cases. These replace older,often uncertain, testing methods, or help to solve cases where othertesting would not have been possible.

!espite the proven value of !"A testing, its reception by the courtscan be characteri*ed by the ebb and flow of an ongoing war betweenprosecutors and defense attorneys and their !"A e#perts. The blamefor this war lies partly with the laboratories which developed and


4/75

introduced the testing, partly with the contentious and fragmentednature of our legal system, and partly with inaccurate media coverage.The !"A showdown in the Simpson case could be the last major battlein the !"A war or it could be just another skirmish in this e#pensiveand senseless war fought with academic trivia and specious

arguments.

6oth sides have e#perienced !"A lawyers to argue their positions anda stable of veteran e#pert witnesses. ven as the adversaries preparefor battle, the battlefield is shifting with almost weekly court decisionsand scientific events that impact the testimony and arguments. t iscertain that the pre-trial evidentiary hearing will go on for weeks asboth sides commit all their forces to persuading )udge %ance to to rulein their favor.

In the $e%innin%

The forensic use of !"A started with the work of Alec )effreys, ageneticist at the +niversity of %eicester in 6ritain5s 1idlands. n 2;,)effreys invented the techniques that took human identification fromthe laboratory to the courtroom. &ith his co-workers, he alsodemonstrated that forensic samples, dried stains several years old,contained sufficient !"A to yield conclusive results. )effreys provedthat even small fragments of !"A molecules were virtually unique toindividuals. &ith appropriate dramatic flair, he called the process heinvented B!"A fingerprinting,B a term most forensic scientists dislikebecause it is confusing and can be misleading.The BfingerprintB produced by the test bears a superficial resem-blance to a supermar-ket bar code with the differences between indi-viduals revealed by the spacing between the 3 or C4 lines calledbands. The differences between specimens are measured by a processcalled 7estriction $ragment %ength 8olymorphism =7$%8? analysis.)effreys5 new form of genetic typing and the law were linked from thebeginning. @e sought high-profile forensic tests for his brainchild. $irst,he applied it to an immigration case. A boy from 0hana sought toemigrate to 6ritain, claiming that his mother was already a resident.(onventional blood tests were not conclusive beyond confirming thatthe two could be related. !"A analysis showed beyond reasonabledoubt that the relationship was as claimed, and the @ome ffice put itsstamp of approval on the new technology.

Fin&in% the Pitchfor' in the (a)stac'

A detective in the ast 1idlands read of the case and sought )effreys5help in solving the vicious murder and rape of two 6ritish schoolgirls.


5/75

The police held a prime suspect in the case, a kitchen porter at aninsane asylum who had confessed to one of the murders. They brought)effreys semen samples from the murder scenes and a blood samplefrom the suspect. )effreys confirmed that the same person committedboth crimes but it was not the suspect the police held. n "ovember

C, 2/, the kitchen porter became the first person in the world tohave his innocence proven by !"A testing.6oth the police and villagers in the area felt strongly that the killer wassomeone in their midst. 8olice were prompted to try something entirelynew. All male residents between the ages of D and >; were asked tovoluntarily submit a blood sample. &ithin a month, a thousand menhad been Bblooded.B 6y 1ay 2D the number had risen to more than>,/44. Summer turned to $all, it seemed that this e#periment wasdestined to be as unproductive as the previous, more conventionalefforts.Then the police received an une#pected tip. A bakery manager

chatting in a pub with some of her employees learned that one of theircolleagues, (olin 8itchfork, had convinced another baker to be bloodedin his stead. After four long years and the disappointment of theporter5s false confession, the detectives felt this was the break theywere looking for. They went to 8itchfork5s home and moments afterarresting him, he confessed. @e became the ;,3>rd and last man tobe tested in the hunt for the 1idlands killer. @is sample provided aperfect match to the sperm taken from his two young victims. t wasSeptember of 2D and forensic !"A was on its way.

Commercia* De+e*opment in the Unite&tates

t also was in 2D that the 6ritish firm of (ellmark !iagnostics openeda branch in 0ermantown, 1aryland and introduced )effreys5 methods inthe +nited States =(ellmark is the private laboratory performing testingfor the prosecutor in the Simpson case?. The firm is part of mperial(hemical ndustries, a giant 6ritish company, which previouslyestablished a !"A laboratory in Abingdon, ngland. &hen (ellmark setup its operation in this country, it had only one other competitor,%ifecodes (orporation of 9alhalla, "ew :ork. $ounded in 2C,

%ifecodes began forensic !"A testing in 2D and took the lion5s shareof the market early.%ifecodes performed the tests in the first case in the +nited States inwhich a criminal was identified by !"A. The trial of accused rapistTommy %ee Andrews began in rlando, $lorida on "ovember >, 2D.A scientist from %ifecodes and a 1..T. biologist testified that semenfrom the victim matched Andrews5 !"A, and that Andrews5 print wouldbe found in only in 4 billion individuals. n "ovember /, 2D, the


6/75

jury returned a guilty verdict and Andrews was subsequentlysentenced to CC years in prison.This case was heavily reported by the press, creating a media blit*favor- able to the new technology. !efense attorneys were caught offguard by the technology and largely accepted it without question.

ther cases quickly followed with the same result. )udges and jurieswere clearly impressed with this new technology.The introduction of !"A methods to the courtroom by privatecompanies was unique in the history of forensic science. The sharpcompetition, the proprietary approach of the industry and their desireto keep their products and processes under wraps did not create afavorable environment in which to launch a new technology with suchvast potential for changing the criminal justice system. 8atentchallenges, litigation, and technology-licensing questions became thenorm and continue to impede the introduction of technologicalimprovements.

The usual methods of testing new scientific methods are publicationand peer review. The requirements for standardi*ation and replicationin multiple labs and evaluation of test performance under e#actingenvironmental conditions are of paramount importance in thevalidation of a forensic test. These did not occur as the commerciallaboratories maintained secrecy while rushing to get a return on theirsubstantial investment and start-up costs.n essence, the major private companies were racing with each otherto the courtroom. They hoped to license their procedures and sell theirproprietary materials and reagents to as many crime laboratories asthey could. They used different tools that produced incompatible

results which precluded comparison. As !"A testing becameestablished, some labs were overwhelmed with casework. Systemswere not yet in place to ensure quality control, nor had the labsperformed sufficient validation studies. They were run like researchlabs, having been started by academic scientists, not forensicscientists. &hile the juggernaut of !"A seemed unstoppable, the veryspeed with which it was moving boded ill.

Rush to ,u&%ment

(ase after case involving !"A evidence was won by the prosecution onthe basis of testing and testimony provided by %ifecodes and (ellmark.The two companies, while competing for the same business, oftenjoined forces to promote the new technology to the bench, bar, andlaw enforcement. Their job was made easier by an adulatory press thatwrote numerous stories about the miracle technology that fingeredcriminals with unerring accuracy.


7/75

)udges accepted the assertions of industry witnesses at face value andjuries were wowed by the big numbers they were offered. n the wordsof a 1assachusetts Supreme (ourt justice, !"A had acquired an Bauraof infallibility.B ne juror in Eueens put it succinctly when he said, B:oucan5t argue with science.B )udge )oseph @arris of Albany, ".:., after

sentencing a defendant on a murder and rape charge that hinged on!"A evidence, called it the Bsingle greatest advance in the search fortruth since the advent of cross-e#amination.BThe reaction within the defense bar ranged from bemusement to shellshock. ne $lorida prosecutor commiserated with attorneysrepresenting guilty clients. Bf they print your guy with this stuff, you5redead. :ou can5t combat it. There is no defense to it.B !efense attorney7obert 6rower5s assessment was equally unequivocal. @e felt that !"Aevidence threatened the constitutional right to a fair trial. Bn rapecases, when the semen has been matched with the defendant5s andthe chance that it came from another person is >> billion to , you

don5t need a jury.BAcross the board, the new technology was changing the criminaljustice system, and defense attorneys didn5t like this development. fcourse, they could hope that at least some of the !"A convictionswould be reversed on appeal. n the meantime, they were clearly onthe defensive as they never had been before.

Ne# or' +. Castro: The Chin' in theArmor

&hen police arrived at the 6ron# apartment of )effrey tero in$ebruary 2D, they discovered a scene of terrible carnage. 9ilma8once, tero5s seven months pregnant common-law wife, lay on theliving room floor, nude from the waist down. She was perforated bymore than si#ty knife wounds. n the bathroom, police found the bodyof her two-year-old daughter, "atasha, also repeatedly stabbed.8olice interviewed )ose (astro, the janitor of a neighboring buildingwho fit tero5s description of the suspect. The detective noticed whathe thought might be a dried bloodstain on (astro5s watch and asked ifhe could retain it for e#amination. Shortly thereafter, (astro wasarrested and charged with the double murder. The dried blood on )ose

(astro5s watch and how it was handled led to the first notablecourtroom challenge to !"A typing.8olice turned the watch, along with blood samples from (astro and thetwo victims, over to the %ifecodes (orporation. Scientists analy*ed thedried blood and during the 3 week long pre-trial evidentiary hearing,testified that the !"A from the stain matched that of 9ilma 8once, andthat the frequency of her patterns in the @ispanic population were'2,C44,444.


8/75

The defense undertook a thorough e#amination of the genetic analysesand mounted the first e#tended =and eventually successful? effort tohave !"A evidence e#cluded. &hat also occurred in the (astro casethat contributed to this turn of events was an unprecedented out-of-court meeting between two defense and two prosecution scientific

witnesses after they had testified. These scientists all agreed that%ifecodes had failed to use generally accepted scientific techniques inreaching their results matching the blood found on (astro5s watch withthat of 9ilma 8once. The quality of the data they produced was poorand they did not even follow their own procedures for interpreting thedata.ne key player in this drama was ric %ander, an academician whoreceived his doctorate in mathematics from #ford +niversity and nowdirected a genetics research institute at the 1assachusetts nstitute ofTechnology. %ander is a powerful personality. ven his friends admitthat %ander is arrogant, just as his enemies concede that he is brilliant.

As a result of the testimony of %ander et al, the judge ruled that theinclusionary tests suggesting that 8once was the source of the bloodstain were inadmissible, while allowing the e#clusionary evidence thatthe blood did not come from (astro. After almost one hundred caseswhere !"A evidence met little or no resistance and never was ruledinadmissible, the defense obtained their first victory. %ater that year, inwhat was to be the anti-clima# to the case, (astro confessed to themurders, admitting that the blood on his watch came from 9ilma8once, and pled guilty.

The F$I an& !ounties Ri&e to the Rescue

$ortunately, the $ederal 6ureau of nvestigation and the 7oyal(anadian 1ounted 8olice entered the picture at about this time, with asalutary effect. The $6 saw the potential for the forensic use of !"Atesting at about the same time that Alec )effreys was conducting hisbreakthrough e#periments. Along with the "ational nstitutes of @ealth,the $6 began collaborative research and in 2D, set up its ownresearch unit to establish !"A identification techniques for the 6ureau.After one year of testing, ending in late 2, the $6 set up their own!"A laboratory at their 8ennsylvania Avenue headquarters. The 7(18soon followed with their own !"A laboratory.The 6ureau benefitted from the e#perience of !"A lab pioneers hereand in urope, and was not locked into a single technology or product.&hen the $6 lab went on-line, it used a combination of four different!"A probes, including those developed by 0ene%e#, !r. 7aymond&hite of the @oward @ughes 1edical (enter, %ifecodes, and (ellmark.!"A probes and primers are the key patented biochemicals used toidentify individual genetic differences.


9/75

The main result of the $6 and 7(18 beginning their own !"A testingwas standardi*ation of a chaotic industry. The $ederal agencyestablished detailed laboratory protocols, performed appropriatevalidation studies, and cut through the competing systems, methods,and tools to establish a standardi*ed system that is used in almost all

"orth American laboratories today. The raising of standards becameeasier once standardi*ation was achieved.

Defense $ar trate%)

As a result of the evidentiary hearing in the (astro case, the "ationalAssociation of (riminal !efense %awyers ="A(!%? set up a !"A Task$orce in the $all of 22. The new group was headed by (astro5s =andnow .). Simpson5s? !"A defense team, 6arry Scheck, a professor at6enjamin ". (ardo*o %aw School, and 8eter "eufeld, a private attorneyin 1anhattan. They asserted that the evidence introduced in the (astro

case did more than prove that the !"A industry was fallible. They feltit was simply a case of Bthe emperor having no clothes.B As .).Simpson5s !"A attorneys they will continue their crusade against !"A,having already filed a more than one hundred page motion to e#cludeall !"A testing evidence from the trial.Time has shown that (astro provided a needed psychological boost todefense attorneys and cannon fodder for !"A critics, but otherwisehad very little effect on the legal system. The prosecution rendered thekey opinion of the trial court moot by agreeing that some of theevidence was inadmissible. Since (astro pled guilty, there was noreview on appeal. At the time, however, at least to hopefulcommentators, (astro looked like a &aterloo for %ifecodes and theprosecutors who depended on the services of that and similar labs Thetask force announced that its first effort would be to try to reopen allconvictions involving evidence tested by %ifecodes. "eufeld evensuggested that the thousands of court orders in paternity suits decidedby !"A testing also were of questionable validity because of the signalvictory in (astro.&hile this wholesale repudiation of !"A testing has remained anunreali*ed dream, "A(!%5s !"A task force has been a large factor inthe !"A war, by leveling the playing field, escalating the conflict, andkeeping the conflict alive against all odds. 8art of their catching upwith the competition consisted of taking the offensive and mounting apublic relations campaign. Some of the press printed new articleswhich were as critical of !"A typing as previous accounts had beenenthusiastic.

Unite& tates +. ee


10/75

n the Spring of 224, Scheck and "eufeld began to prepare for a trialin hio, +nited States vs. :ee, that would be a rigorous judicial inquiryinto the soundness of !"A testing.Three members of the (leveland chapter of the @ell5s Angelsmotorcycle gang, Steven :ee, 1ark 9erdi, and )ohn 6onds, were

accused of killing !avid @artlaub of Sandusky, hio. The defendantsallegedly killed @artlaub because they mistakenly thought him to be amember of the rival utlaw5s 1otorcycle (lub, with whom they werehaving a turf war. The victim was shot fourteen times with a silenced1A( 4 machine gun inside his own van.1ost of the blood was later determined to be @artlaub5s, but bloodtyping tests revealed that some was not. !etectives theori*ed that aricocheting bullet had hit one of the suspects. !"A analysis by the $6showed a match between blood from the van, from :ee5s car, and from)ohn 6onds. t was this evidence that the defense planned to challenge.n a way, (astro was a dress rehearsal for :ee. 1any of the rising stars

in the !"A constellation appeared as witnesses for one side or theother at the )une 224 hearing in Toledo before +.S. 1agistrate )ames(arr. The prosecutor called si# witnessesF including Thomas (askey ofthe 6aylor (ollege of 1edicine, who had just finished serving as headof a panel that e#amined forensic !"A for the (ongressional ffice ofTechnology AssessmentF Genneth Gidd, a :ale geneticistF and 6ruce6udowle, the $65s main !"A scientist. The defense had five e#pertsincluding 7ichard %ewontin, and !aniel @artl, geneticists at @arvardand &ashington +niversity, respectively. The court called ric %ander,the 1..T. mathematician-turned-geneticist who had testified in (astro.The defense launched a full-scale attack on the $6 and its work. They

claimed that the 6ureau5s published articles on its matching criteriawere ambiguous or inconsistent and sought access to the supportingdata. The prosecutor fought the motion for discovery, but the courtgranted the defense access to these materials. The e#perts pored overthe data and had a field day, citing what they claimed to be errors,omissions, lack of controls, and faulty analysis.)ames &ooley, the federal prosecutor, countered the criticism byreiterating two telling points. $irst, regardless of disputes over matchcriteria, the multi-probe match produced by the $6 was highly unlikelyif specimens came from two different people. @ence the chance of aninnocent person being incriminated was virtually nil. Second, =and the

court noted that there were Btroublesome questions about the qualityof the 6ureau5s workB?, whatever deficiencies e#isted went to theweight not the admissibility of the evidence. The magistrate concurredand the !"A matches were admitted. They also were admitted at thehomicide trial in state court and passed muster on appeal.6arry Scheck considered the :ee case to be far more significant than(astro to the defense bar. The documents procured from the $6 andothers through the discovery process were to find their way into many


11/75

courtrooms throughout the land. &hile they had lost the battle atToledo, Scheck, "eufeld, and their cohorts had sei*ed reams ofammunition and gained new recruits for the many battles that layahead.

/oo& Intentions $ac'fire: The NRC Report)ust after the (ongressional ffice of Technology Assessmentpublished a report recommending the continued use of forensic !"A in22, a second federal study was initiated in response to Ba crescendoof questions.B $unding was procured for a more definitive study of theproblems, this time by the "ational Academy of Science5s "ational7esearch (ouncil ="7(?. ne of the deans of American genetics, 9ictor1cGusick, was picked to chair the fourteen person panel. They begantheir work in )anuary 224.$rom the beginning, the courtroom battles over !"A were refought in

the "ational Academy of Sciences committee rooms, and with nearlythe same ferocity. The most contentious issue was the matter of howto calculate statistical probability, the odds that a match between !"Afound at the crime crime and !"A taken from the suspect could be theresult of coincidence. To find a match, crime labs look at several siteswhere the !"A is known to vary. f these sites match, there is ane#tremely high probability that the samples came from the sameperson. To quantify these findings, investigators calculate thefrequency with which each variation occurs in the suspect5s populationgroup. The frequencies for each site are then multiplied together toarrive at a figure for the complete !"A profile. !atabases of !"Aprofiles for various populations have been gathered for use in makingthese calculations.To some population geneticists, there5s the rub. $riction over this pointprovided the spark that was fanned by the theory of the twogeneticists, 7ichard %ewontin and !aniel @artl. They maintained thatthe frequencies of genetic markers in sub-groups could differ widelyfrom the frequencies found in larger population groups. f this is so,then any estimates calculated using the widely accepted $6 match-binning methods could be considerably wide of the mark. 8roponentsof this theory insisted that e#tensive and e#pensive population studiesmust be completed before reliable estimates could be introduced intothe courtroom, even if this takes a decade or more. &hile ric %anderdidn5t go all the way down this road, he was a fellow traveler enough ofthe distance to become the champion of e#treme caution, representingthe e#treme view on the panel.At the other end was Thomas (askey, then president of the AmericanSociety of @uman 0enetics, an advisor to the $6, and developer oftechnology used in !"A analysis. (askey took the widely predominant


12/75


13/75

A major problem with the classical systems that test proteins found inblood, is that very few of these proteins are also found in semen andother body fluids. 6ecause the majority of cases that require genetictyping are se#ual assaults, the lack of a definitive set of useful geneticmarkers in semen has long been a great handicap to the scientific

analysis of rape evidence.!"A is found in almost every cell in the body. !"A typing has greatlye#panded the sources of evidence that can be tested, whilesimultaneously reducing the amount of evidence necessary to performa conclusive test.

ources of DNA for Testin%

$*oo&

The surface or substrate on which a bloodstain is found can profoundlyaffect the ability to successfully perform an analysis. #tracting !"Afrom dried blood found on glass, metal, hard plastics or lightweightcloth is straightforward. !enim, vinyl, carpet, automobile seats andother dense and heavily colored fibers require additional steps. Theyoften contain substances that inhibit the !"A testing process andwhich must be removed by a variety of purification methods. (oncreteand other similar porous surfaces are difficult simply because it isdifficult to separate the blood from the concrete. Soil is an almostimpossible substrate from which to obtain usable !"A because currente#traction methods are not capable of releasing the !"A from soil.

6lood stains may be mi#tures of blood from two different people andcan produce !"A profiles that are more comple# than those from asingle individual. !"A profiling may be the only way to determine if agiven stain is a mi#ture. !"A testing can also determine if specimensare from individuals of a different se#. 0ender is the only physicalcharacteristic that can be determined by forensic !"A analysis.Specimens drawn from suspects or victims are called known e#emplarsand usually consist of liquid blood. The simplest and most reliablesamples to work up are liquid whole blood. &hile the best storagemethod for blood is fro*en, !"A has been successfully isolated afterrefrigeration for several months. The most common way in which

forensic blood samples are handled, following collection, is as a seriesof dime-si*ed stains on washed cotton sheeting. This method makesfor convenient transport and storage of large numbers of samples.ccasionally buccal =inside cheek? swabs are supplied as e#emplars.All +.S armed forces recruits give both blood and buccal cell samples.This process has the advantage of being non-invasive. t can be usedwith infants to determine paternity and with other individuals whocannot have blood drawn for medical or religious reasons.


14/75

#emplars may also be obtained from the B0uthrie (ards,B widelycollected at birth for newborn screening of genetic disease and savedby many states. These have been used in body identification. Similarlya seventeen-year-old band-aid was used to determine the paternity ofa young man killed in a car wreck. The question of paternity came up

after he was buried, as part of a dispute over insurance benefits.

emenSemen stains are the most common evidence to be submitted for !"Aanalysis and are typically e#amined and tested by conventionalmethods prior to !"A analysis. Sperm cell specific staining andmicroscopy help determine if a stain is the result of vaginal drainage.9aginal swabs or stains resulting from post-coital drainage will typicallycontain sperm cells mi#ed with vaginal cells. Sperm cells as driedstains on furniture or clothing may be identifiable for years.Semen and condoms are surprisingly common in many environments.

dentification of one or a few sperm cells, for e#ample, is notconsidered highly probative. Se#ually active women will often havesperm present in the vagina for DC or more hours following intercourse.That is why it is useful to also test the boyfriend or consenting se#ualpartner of rape victims.1i#tures of cells are usually a confounding factor when performingstandard blood typing. n the case of !"A analysis, the process ofdifferential lysis =a controlled disintegration of the cells?, is used toseparate the sperm cells or male fraction from the non-sperm cells,Bthe female fraction.B

Tissuessolation of !"A from tissues taken at autopsy is a simple,straightforward process. 1ore often, soft tissues received in thelaboratory are from partially decomposed bodies. !"A survival timesare low in soft tissues such as liver and kidney, longer in muscle andbrain, and longest yet in dense bone and teeth.1inute remnants of brain or other tissues that have been scattered bygunshot or other trauma or tissue that might still adhere to the weaponor a bullet can sometimes yield sufficient !"A for testing. 6its of tissuesuch as these can be used if a body is not recovered and thenidentified by Breverse paternity,B by comparison with surviving family

members.

Chemica**) Treate& TissuesThe two common chemical treatments are formaldehyde fi#ation formedical purposes and embalming. !"A can be successfully testedfrom both tissues. 6odies that have been buried for several years caneven yield results. 1ost commonly, e#humations are required in


15/75

paternity determinations, but they may also provide a known e#emplarto compare with evidence in a forensic case.

(air Rootsne to five hair roots can contain sufficient tissue for 7$%8 analysis.

Shed, or telogen, hairs contain only trace amounts of !"A, and aregenerally not suitable for testing by current methods in routineforensic use. A highly speciali*ed method known as mitochondrialsequencing is beginning to be used more and more. t was this methodthat was recently used to identify one of "apoleon5s hairs andmembers of the 7omanoff family.

a*i+a!"A can be typed from saliva deposited on envelope flaps or stamps,as was done in the "ew :ork &orld Trade (enter bombings. t can alsobe taken from cigarette butts found at crime scenes, and even from

cups, bottles, telephone mouthpieces, bite marks and penile swabs.

UrineSuccess with !"A analysis of urine is less common, as healthyindividuals do not shed nucleated cells into the urine. (onventionaltesting is more likely to yield results, and is used in disputes oversamples that have been tested for drugs.

Pro&ucts of ConceptionThe non-living products of conception most often analy*ed areabortuses. +sually these cases stem from allegations of criminal

paternity where a teenaged girl has been impregnated by a relative orother adult. ther cases might involve a prison guard accused ofhaving se#ual relations with an inmate or a rape resulting inpregnancy.

Chapter 2:

DNA IN T(E "A$ORATOR

$asic /enetics: Wh) DNA T)pin% Wor's

There are about 44 trillion cells in the adult human body. 1ost ofthem have a nucleus, or center, that contains thread-like bundles ofchromosomes. n these chromosomes are all of the instructions andinformation needed to make a human being. ach parent contributesone chromosome to each of the C> pairs found in all normal people.&ithin the chromosomes, are up to 44,444 paired genes, thefundamental units of heredity. ach gene can have different versions


16/75

=as many as 44 or more in rare cases? called alleles, but most are thesame from person to person. 0enes determine all inherited traitsincluding those that give the individual specific characteristics =blueeyes rather than brown eyes? as well as common characteristics =twoeyes, two arms, etc.?.

0enes are made of deo#yribonucleic acid =!"A?. @ence, !"A is themaster molecule of life and controls the growth and development ofevery living thing. t is a polymer, i.e., a long string of simple repeatingunits. These repeating units are called nucleotides and are of fourtypes' adenine =A?, cytosine =(?, guanine =0?, and thymine =T?. )ust asthe order of the letters of the alphabet determines the informationcontent of words, the order in which these four bases are strungtogether is what gives !"A its information content. The complete !"Amolecule consists of two of these strands of the four bases.n the two strands, A always is across from or paired with T and 0always is paired with (. These are the base pairs that are the unit of

measurement in determining the si*e of a given segment of !"A. Thisstructure suggests a natural mechanism for the duplication orreplication of the !"A molecule, as occurs during cell division. Thesepairings are what connect the two strands of !"A together to form atightly coiled, twisted ladder. This spiral staircase, the famous doubleheli#, is the natural form in which !"A is found within the nucleus ofthe cells.f uncoiled, the !"A molecules in every human cell would measure si#feet in length. That is the total length of the >.> billion base pairs thatmake up the total human genetic complement or genome. #cept foridentical twins, the sequence of the base pairs within the !"A heli# is

unique for every person, and forms the individual5s genetic code orblueprint.8erhaps the basis of !"A typing can be best understood by comparingthe way in which genetic information is stored in the !"A to the way inwhich printed information is stored in books. $or e#ample, if we wereto cut all the sentences in forty volumes of the ncyclopedia 6ritannicainto strips, and tape them together end to end, then we would have anamount of information equivalent to that contained in the !"A withineach of the cells that make up our bodies. $urthermore, theinformation would then be in the same physical form as the !"Ainformation, i.e., a long linear strip sometimes likened to a computer

punch tape.The genetic information contained in the !"A is organi*ed andpackaged into chromosomes, much as printed information is organi*edinto volumes. )ust as a specific passage in the encyclopedia can beidentified by specifying a volume, page, and line number, a specificgenetic passage or location, known as a locus, can be identified. Aspecific naming system identifies genes by numbers issued by the@uman 0ene 1apping (ommittee. $or e#ample, if we see the designa-


17/75

tion !;S>2 in a report, then we know e#actly what gene has beenanaly*ed, that it is on chromosome four, and that it is the >2th !"Aprobe to be mapped to chromosome four.A significant difference between the way information is stored in thecell 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 thefather. These two copies of the genetic information which are largelyidentical, come together at the moment of conception when the spermand the egg join together. All of the child5s cells contain !"A derivedfrom this original fertili*ed cell, half from the mother and half from thefather. t is this basic principle of heredity, first discovered over 34years ago, that allows us to reliably perform parentage tests.

aria3*e DNA: The 4e) to DNA T)pin%

The !"A and hence the genetic code of humans is almost the same forall individuals. t is the very small amount that differs from person toperson that forensic scientists analy*e to identify people. Thesedifferences are called polymorphisms =from the 0reek for BmanyformsB? and are the key to !"A typing.Two major kinds of polymorphisms are most useful to forensicscientists. The first consists of variations in the length of the !"A atspecific locations =loci? known as 9"T7s =variable number of tandemrepeats?. These 9"T7 regions consist of stretches of !"A made up ofshort repeating !"A sequences. The number of times the short !"A

sequence is repeated determines the physical length of the !"Amolecule at these specific loci. ach of the many versions that may befound in the population are known as alleles. These variations aree#amined by the method known as the 7$%8 technique. The secondtype of variation in the !"A is simply a difference in the nucleotideletters found at a specific pair of bases. These are e#amined by themethod known as 8(7.The goals are the same for both types of !"A tests' to isolate adistinctive !"A sequence and record its presence in a way that can bee#amined visually. The basic procedure of 7$%8 analysis is known asBSouthern blottingB after dward Southern, a Scottish bio-chemist who

developed the technique in the early 2D4s. The basic procedures of8(7 based testing were invented in the early 24s by Gary 1ulliswhile working for a (alifornia biotechnology company.!eciding which method to use is determined by the amount of !"Athat is available and how deteriorated or degraded it is. 7$%8 analysisis more often performed because it is more discriminating. f there isonly a small amount of material or it is highly degraded, 8(7 analysisis used, because this method requires less material and can produce a


18/75

result on !"A of poor quality. As little as two billionths of a gram =Cng?, the amount of !"A contained in about D44 sperm cells will sufficefor 8(7 analysis. The 7$%8 test requires C4 to 34 ng of !"A. 8(7 testscan be performed in a matter of days as compared to weeks for 7$%8tests.

DNA E+a*uation

nce the evidence has been documented and screened in thelaboratory, and deemed appropriate for !"A analysis, the initial step isto isolate or purify the !"A. $irst, it must be removed from whateverobject it is attached to and removed from the cell. +nless the majornon-!"A constituents of the cell such as proteins, fats andcarbohydrates are removed, the en*ymes essential in the ne#t step

will not be able to do their work. This isolation of !"A begins with thecontrolled destruction of cellular integrity, releasing the !"A from itsnuclear and chromosomal packaging. The cell walls are dissolved witha detergent and the proteins are digested by en*ymes.The !"A then is purified by the methods of e#traction andprecipitation. nce the !"A is isolated and concentrated, a smallsample is tested to determine quality and quantity. f intact =highmolecular weight? human !"A is present in sufficient amounts, 7$%8testing can proceed. f the !"A is degraded, or present in minuteamounts, 8(7 testing is used.

Restriction Fra%ment "en%thPo*)morphism 5RF"P6 Ana*)sis

7$%8 typing of purified !"A consists of the following si# steps'1. Cutting the DNA into pieces (Restriction Enzyme Digestion);2. Separating the DNA by size (e! E!ectrophoresis);

". #rans$erring the DNA to a so!i% support sur$ace (Southern #rans$er);

&. #argeting an% 'isua!izing the DNA o$ interest (ybri%ization an%

Autora%iography);. Rea%ing the DNA pro$i!e (*nterpretation o$ Data); an%

+. Determining the rarity o$ the DNA pro$i!e (,opu!ation enetics an% -reuency

Estimates).

Restriction En7)me Di%estion


19/75

7estriction en*ymes reproducibly cut !"A at specific four or si#-base-pair sequences called restriction sites. @ae =@aemophilus Aegyptius?, the most commonly used en*yme in forensic science, cuts the !"Aeverywhere the bases are arranged in the sequence 00((. These sitesare found throughout the human genome and are, for the most part,

the same in everyone. @ae cuts human !"A into appro#imately Cmillion different restriction fragments ranging in si*e from a fewhundred to 4,444 or more base pairs in length.

/e* E*ectrophoresis

The physical length of !"A restriction fragments in 9"T7s is onefundamental biochemical characteristic that varies from person toperson. lectrophoresis is the technique by which the different-si*ed

fragments are separated. The !"A is loaded into a hole, or well, at oneend of a slab of semi-solid gel, a porous, )ello-like substance. &hen anelectrical field is applied to the gel containing the !"A, the !"A movestoward the positive electrode because it has a negative electricalcharge.The sieving action of the gel allows the small restriction fragments tomigrate at a faster rate than the larger ones, just as a small rabbitwould move faster and farther through a thicket than a large hound inpursuit. The relative position of these fragments within the gel afterovernight current application is determined by their length ormolecular weight.

0els have several lanes within them which contain !"A, includingmarker lanes that measure how far fragments move through the geland a lane for control !"A that produces a known pattern and can beused to verify that the test was properly conducted. The !"A in the gelcan be stained and seen under ultraviolet =black? light.

outhern Transfer

6ecause the gel is fragile, it is necessary to remove the !"A from the

gel and permanently attach it to a solid support. This is accomplishedby the process of Southern blotting. The first step is to denature the!"A in the gel which means that the double-stranded restrictionfragments are chemically separated into the single-stranded form. The!"A then is transferred by the process of blotting to a sheet of nylon.The nylon acts like an ink blotter and BblotsB up the separated !"Afragments. The restriction fragments, invisible at this stage, are


20/75

irreversibly attached to the positively-charged nylon membrane calledthe Bblot.B

DNA Pro3es9isual observation of an individual5s !"A pattern requires the use of!"A probes. The !"A probe, like a guided missile, will seek out andfind its target sequence. There are appro#imately a do*en 9"T7 !"Aprobes in common usage in forensic 7$%8 testing. These probes havebeen patented and are commercially available from biotechnologysuppliers. ach !"A probe can be used to develop the !"A profile at aparticular 9"T7 locus. 1ost of these are located on differentchromosomes, an important factor to be considered when performing astatistical analysis. 1ost forensic tests use a combination of at least

four to si# separate !"A probes in a sequential manner. All of these!"A probes have been obtained by the process of cloning, whichsimply means that they are free of all other human !"A. Therefore,large quantities of probe !"A can be made and labeled with aradioactive or other tracer in the laboratory.

()3ri&i7ation

To detect a 9"T7 locus immobili*ed on the Southern blot, one uses a

!"A probe that has a base pair sequence complimentary to the !"Asequence at the 9"T7 locus. The double-stranded nature of !"A and aphenomenon called hybridi*ation provide the scientific basis for theusefulness of !"A probes. !ouble-stranded !"A fragments willseparate when heated. &hen the !"A cools down, the two strands willreconnect and become double-stranded again. This is not a randomprocess. 6ecause of the complementarity of the two !"A strands, thesingle strands will only reconnect with another strand that has acomplimentary sequence.)ust prior to its use, the !"A probe is labeled with a radioactive orother tracer and boiled to the single stranded form. The !"A on the

blot, also single stranded, is soaked or incubated in a solutioncontaining the !"A probe. The probe hybridi*es, or binds, to only the!"A fragments that bear the complimentary sequences of !"A bases.These will be the restriction fragments corresponding to the locus fromwhich the probe was originally cloned.


21/75

Autora&io%raph)

The e#cess probe is washed off and the blot placed in contact with a

sheet of film. The film, e#posed by the radioactive tracer instead oflight, is developed and becomes an autoradiograph, commonly knownas an autorad. The autorad is the final product of the 7$%8 analysis. treveals the overall quality of the testing and can be copied, distributedand interpreted by other !"A e#perts.The autorad has darkened areas known as bands corresponding to theposition of the !"A probes and hence the restriction fragments boundto the membrane. Typically, several probes are used sequentially inorder to compile forensically significant !"A profiles. This requires thatthe blot be stripped of the first radioactive probe, hybridi*ed to asecond, washed and e#posed to a new film to make another autorad.

ach round of hybridi*ation and autoradiography may take up to tendays. This is why 7$%8 testing takes longer than conventional blood or8(7 testing.

Interpretation of Data

+p to this point in !"A analysis, there is little argument about itsvalidity =provided that it is done correctly? because we are dealing withphysical reality. The interpretation of the data, what is its meaning, is

another story. This requires that inferences based on the science ofstatistics, population genetics and probability theory be applied to themeasurements of physical reality that the auto7A!S reveal.$orensic scientists use these mathematical concepts to calculate andreport an estimate of how frequently the genetic profile they haveobserved might be found in major population groups. f the geneticprofiles found in two different samples, say one from a piece ofevidence and one from a suspect are indistinguishable, they are said tomatch. A typical population frequency for conventional blood typingmight be in C44, for !"A in 3,444,444. This means that only in3,444,444 people would have the same !"A profile. All others would

be e#cluded from being the source of the matching evidence.rrespective of what calculation method is used, it is a physical factthat the genetic profiles match and that they would be found at somefrequency in the population. n attempting to call the frequencyestimates into question, attorneys are fond of pointing out that theyare a comparison between the observed profile and randomly chosenindividuals, and that a relative of the person with the profile is muchmore likely to share that profile than any of the random individuals.


22/75

&hen forensic scientists provide a report and testimony about thefrequency estimates, their job is done. The judgment of guilt orinnocence reached by the court may take these estimates intoaccount, but they must be placed into the circumstances surroundingthe crime and the intent, motive, means and opportunity available to

the defendant.

CO!!ON FA""ACIE

The meaning of the genetic profile frequencies are often misconstruedby attorneys. $or e#ample, the argument might be made that if a pairof matching genetic profiles are found in in C44 individuals thenthere is a in C44 chance that they came from different sources. "ottrue. &hat the chance is that they came from different sources cannot

be determined by the genetic evidence alone. t depends on all of thecircumstances surrounding the case.Another fallacy is the one heard in the Simpson case preliminaryhearing. ,). Simpson5s blood type matched blood found on thesidewalk trailing away from the murder scene. !efense attorneyspointed out that 4,444 people in %os Angeles share that blood type.True, but all of those 4,444 people didn5t visit 3 South 6undy wherethe homicides occurred.The prosecutor5s fallacy is to argue that as the genetic profile is foundin in 444 people then there is only a in 444 chance that thedefendant is innocent. nce again this ignores the other facts of the

case and asks the scientific evidence to determine guilt or innocence.Science cannot do that. 0uilt or innocence must be determined by thejudge or jury.

EA"UATIN/ T(E AR/U!ENT

(ritics of !"A testing have generated and sei*ed upon disagreementsabout the best way to interpret the data as a reason not to admit !"Aevidence. $orensic scientists themselves have given the critics

ammunition by not always following widely used scientific conventionsfor rounding off and reporting the significance of the numbers that arereported.n evaluating the criticisms that are made, primarily of the validity ofthe population data on which probability estimates are made, it isimportant to remember that the critics'

*$ they are geneticists at a!!/ stu%y non0human organisms or came to the stu%y o$

human genetics $rom other $ie!%s. None are $orensic scientists.


23/75

se the same assumptions in their on or3 that they argue against in $orensic

science.

4ase their critiues on o'er%ran hypothetica! an% theoretica! arguments $or

hich no %ata e5ists.

*gnore the sa$eguar%s an% ua!ity programs that are routine!y use% by $orensic

DNA !abs.

1uch of the population genetics controversy has been generated by anelementary misunderstanding of the basics of 7$%8 testing in the earlywriting of ric %ander. Another source of misunderstanding was anarticle published in 2DC by 7ichard %ewontin and since repudiated bythe author. %ong after %ewontin changed his mind, the articlecontinued to be used to support defense motions to quash !"Aevidence.&hat this reveals is that one defense strategy is the shotgun approachF

throw up a barrage of flak and hope that the judge or jury accepts thevalidity of at least one of the arguments. The critics also change theirarguments over time. As various lines of attack are countered ore#posed as spurious, new objections are raised. $requency estimatesare fertile territory for the attackers because most people are notconversant with the methods of statistics and the meaning ofprobability.

Rea&in% the Autora&

6efore an opinion can be formed as to the forensic significance of a setof !"A profiles, a series of data interpretation steps must becompleted.

-irst/ the e5aminer inspects the case speci$ic %ata an% recor%s to %etermine i$

proce%ures ha'e been $o!!oe% an% interpretab!e resu!ts ha'e been pro%uce%.

Secon%/ a 'isua! inspection o$ the autora% is ma%e an% an opinion o$ the testing

resu!ts is $orme%. *s the suspect inc!u%e% or e5c!u%e% $rom the group o$

in%i'i%ua!s ho cou!% be the source o$ the e'i%ence6

#hir%/ computerize% e!ectronic measurements an% statistica! ca!cu!ations are ma%e

that must con$irm the e5aminer7s 'isua! 8match8 ca!!s.

-ina!!y/ probabi!ity ca!cu!ations using popu!ation genetic %ata are use% toca!cu!ate a conser'ati'e estimate o$ the occurrence o$ the DNA pro$i!e in ma9orpopu!ation groups.

n forensic science, statistics are used to ensure that the data isinterpreted in a conservative fashion, that it can be guaranteed to bean overstatement of the true occurrence of a !"A profile in the


24/75

population. $or e#ample, if we were to test everybody in the entireworld then we would know e#actly how often a given genetic profileoccurs. %et5s say that it is in a million. f we report that profile out at in 344,444, that is acceptable because it is an understatement. Thedata has been systematically skewed in favor of the defendant.

isua* Interpretation

The purpose of the visual inspection is to form an opinion as to whichof the !"A samples in the various lanes in the autorad could havecome from the same individual source and which could not. ncluded inthe autorad are bands from any !"A that was tested as well as controlspecimens from known individuals. n order to be valid, the opinionformed at this stage must be confirmed by the following quantitative

tests.

DNA Fra%ment i7e Estimation

(entral to the analysis are the series of Bsi*ing laddersB containing upto >4 closely spaced bands. ach of the bands in the si*ing ladderscorresponds to a fragment of !"A of e#actly known si*e. These are therulers which computeri*ed video devices use to measure the !"Afragments in each of the sample lanes. The estimated si*e of the

fragments being measured are reported in base pairs.At least one sample of a human !"A, known as G3/C, is run onvirtually every membrane used in forensic casework. The G3/C !"A isavailable as a standard reference material from the "ational nstituteof Standards and Technology. Si*ing estimates of the G3/C must bewithin established tolerances before si*ing estimates from thatmembrane are acceptable.

$an& Comparisons 3) 8!atch Criteria8

(omparisons are made between the !"A profiles of the knownsamples =generally blood samples taken from the suspect andHorvictim? and !"A profiles of the evidence samples =for e#ample,bloodstains from the crime scene or dried vaginal swabs?. There arethree possible results'

#here is a match beteen to specimens being compare%. *$ the match is


25/75

1. beteen a suspect an% the DNA in the e'i%ence/ then that suspect is inc!u%e% in

the group o$ in%i'i%ua!s ho cou!% be the source o$ that e'i%ence;2. #here is not a match beteen the samp!es. #he suspect cou!% not be the source o$

the e'i%ence. #hey are e5c!u%e% $rom the group o$ in%i'i%ua!s ho cou!% ha'e

contribute% the specimen; or

". #he %ata is inconc!usi'e/ meaning that it is not possib!e to ma3e a %etermi0 nation.#his can be cause% by a 'ariety o$ $actors/ an% usua!!y occurs hen the DNA is

o!% an% hea'i!y contaminate%.

nterpretation of 7$%8 data has required the adoption of statisticaltechniques new to forensic serology because of the numeric nature ofthe data. n A6 and other typing systems, the data is a non-numericlisting of discrete types. n the 7$%8 technique, the si*ing estimatesform a continuous distribution because restriction fragments differingby one 9"T7 repeat unit cannot be distinguished. Therefore, it

becomes necessary to perform statistical calculations to determine iftwo bands on an autorad can be distinguished from one another.$or forensic purposes, two !"A profiles are said to match if they arestatistically indistinguishable and are therefore consistent with havingbeen produced by !"A from the same individual. $irst a quantitativematch range for each appropriate 9"T7 band is calculated. "e#t thematch ranges of corresponding bands are compared lane to lane. fthese ranges overlap then the bands are indistinguishable from oneanother and are said to match. The process continues until all bands ata locus, and all bands at all the loci e#amined have been compared. norder for two composite !"A profiles to be declared a match and

therefore included in the group of !"A profiles that could have comefrom an individual source, all bands must match.1ost laboratories use a match range, or BwindowB, on the order of IC.3< of the measured si*e or molecular weight of the band. This matchwindow has been determined by comparing the !"A profiles obtainedfrom vaginal swabs and fresh blood from rape victims. &hen workingwith fresh blood samples, such as in a parentage laboratory, themeasurement error is less than < of the measured si*e of the bands.

$an&shiftin%The relatively large si*e of the forensic match window compensates forthe phenomenon known as BbandshiftingB. &hen !"A fragments areseparated by electrophoresis, each sample is loaded into its own laneon the gel.


26/75

Chapter 9:

DNA IN PARENTA/E TETIN/

UPDATED FOR T(E WE$ EDITION Apri* 0;;;

n the +nited States, establishing paternity has become a majorindustry. 0overnment efforts to recover child support funds bring inover five billion dollars a year at a cost of one-and-a-half billion dollars.State and local child support agencies, working under Title 9! of thefederal Social Security Act, Aid to $amilies with !ependent (hildren,administer these programs for the benefit of unsupported children.This statute requires that when a woman obtains public assistance forher children, she 1+ST name the possible fathers. These programsgenerate the bulk of more than CC4,444 paternity tests performed

annually. n a country where >4< of births are to single mothers, and3< of fathers of record are not the biological father, the large numberof tests is not surprising.

8arentage testing cases are numerically the largest user of !"Atesting. 1ore than C44,444 of the paternity tests performed in the+nited States every year use !"A testing e#clusively or as asupplement to conventional blood testing. @%A =@uman %eukocyteAntigen? and other classical blood typing tests often can resolve aroutine paternity case without resorting to !"A testing. The @%A bloodtypes are the types that must be matched before an organ transplant,such as bone marrow, kidney or heart, can succeed. %imitations ofconventional blood testing are that the blood must be fresh whentested, and only a limited amount of genetic information is obtainable.Also non-blood cell types, such as those found in semen, do not havean @%A and most other common blood types, rendering this type oftesting much less useful in forensic cases. The advantage of !"Atesting is that many more variable genes, up to thirty or more, areavailable. n effect, our ability to do paternity testing is no longerlimited by the amount of genetic information that we are able toobtain.

1ost paternity testing is done for financial reasons, i.e., to establishlegal responsibility and provide for child support. ven more importantare the emotional and social issues. &hen testing can demonstrateconclusively to a man that he is the father of a child, then he is morelikely to provide not only financial support, but emotional support aswell. @e may bond with the child and take an active part in its life. Theimportance of establishing paternity early is nowhere more clearly


27/75

shown than in the testing of adult children, some of whom are inmiddle age. B&ho is my fatherJB is a question that may have hauntedthem their entire lives.

t wasn5t very long ago that obtaining a paternity test was an

uncertain, e#pensive and inconvenient process. t was hard to find aplace to have the blood drawn to initiate the testing and redraws werecommon as the blood had to reach the laboratory in one or two days. Adoctor5s or a lawyer5s involvement might have been required, and,even after the effort and e#pense, the results often came back with aconsiderable degree of uncertainty. Also, there were no inspection andcertification programs available to parentage testing laboratorieswhich provide an assurance of quality.

!"A testing has made the process convenient and the resultsconclusive. The testing is still e#pensive and takes a week or two to

complete, but anyone can order it. n all but the rarest of instances, the!"A test results provide a level of certainty so high that paternity will,for all practical purposes, be proven or disproved.

ne of the greatest benefits !"A has brought to parentage testing isthe ability to solve many more types of cases. n the usual case, testsinclude all three parties, i.e., the mother, child and alleged father. &ith!"A, it is not necessary to have the mother5s sample in order toprovide a conclusive test result. t is possible to distinguish betweentwo brothers provided they are not identical twins. n many cases, it ispossible to perform the !"A tests when the alleged father is deceased

or otherwise unavailable. t also is possible, using !"A, to performtesting before or at the time of birth. Tests also can be performed onsome very unusual samples such as envelope flaps, cigarette butts,and very old blood stains.

What (appens at the amp*eCo**ection

Throughout this country, and the entire world, is a network ofindividuals licensed and e#perienced in the legal sample collectionsthat are necessary in paternity testing. These collections differ frommedical collections by the e#tent to which they are documented.(lients who are being tested for paternity will be asked to show apicture !, fill out and sign release forms and identification labels, anda photo or thumbprint will be taken. n this way the chain of custody=((? that must be maintained throughout the transport and testing


28/75

can be assured. n order for the (( to have ma#imum integrity thetested parties should not be involved in the shipping or handling of thetransport kit. t should be sent directly to the collection site or personalphysician, and include pre-addressed return airbills with instructions.nly a small volume of blood, a teaspoon or less, will be taken in a

draw. (hildren, even babies, usually tolerate it quite well. The bloodsamples from babies also are drawn from the arm because it is lesspainful and much quicker than a heel or finger stick. +nlike formerpractices, blood now can be drawn any day of the week, and all partiesneed not be drawn at the same time or place.

$ucca* or Chee' #a3s an& (ome DNAPaternit) Tests

1ore and more collections are performed using a Bbuccal swab,B a E-tip that has been rubbed on the inside of the cheek. This type ofcollection has made the Bhome paternity !"A testB possible. The hometest is the ultimate in convenience. +nfortunately, it has minimal legalstanding because it is generally impossible to independantly verify theidentity of the individuals who have been tested. There is no chain ofcustody. Another type of home !"A test overcomes these obstaclesbecause the collection is performed in home by a medical professionalwho comes to the home. This level of service is not availableeverywhere but is becoming more available all the time.

PARENTA/E TETIN/ APP"ICATION

Prenata* an& Ne#3orn Paternit) Testin%

$ormerly, it was impossible to test babies until they were si# monthsold. Testing blood from the umbilical cord of a newborn now is one ofthe most convenient and effective ways to obtain a paternity test. Atthe time of birth labor and delivery staff will take blood from theumbilical cord after the baby is born, collect from the mother and ship

the samples to the laboratory. Arrangements must be made inadvance and if the alleged father has been drawn prior to the birth,testing can start immediately.ne of the fastest growing areas !"A paternity tests is prenataltesting. &hile it has been possible for many years to perform aprenatal test for paternity, it has not been routinely done because ofthe small risk to the fetus and the long turn around time =si# to eightweeks.? That is changing as the amniocentesis and chorionic villus


29/75

sampling procedures become more accepted and routine. These arethe procedures for obtaining cells from the unborn baby which, in thepast, have been limited to women over thirty-five or for families with ahistory of genetic disease. &ith the latest in !"A testing methodsresults can be obtained in a matter of days.

Decease& A**e%e& Father

This type of !"A testing often is needed to obtain Social Security orinsurance benefits, to settle an inheritance dispute or to establishstanding in a wrongful death action. There are two basic approaches tosolving these cases. The first is to obtain a postmortem sample of thefather5s !"A. The second is to reconstruct his genotype based onsurviving family members. Samples from the deceased may be

obtained from the medical e#aminer, coroner or other pathologist whomay have performed an autopsy, an associated to#icology or medicallab, the funeral home before or after embalming, or even followingburial for up to several years.

4inship Ana*)sis - !issin% A**e%e&Father

f samples from the deceased are not readily available then the allegedfather5s genotype can be reconstructed by testing close relatives. Themost straightforward of these tests is a grandpaternity test in whichthe alleged father5s parents are tested. f one or both of his parents aremissing, then samples from his siblings or other children can be usedand e#tended typing performed. Any combination of three or four ofthese close relatives will result in a highly conclusive test. &hen fewerrelatives are available for testing, the ability to provide a conclusiveresult decreases. 7elatives farther removed from the alleged fathersuch as cousins are not related closely enough to provide usefulinformation. ne caution to this type of testing is that if therelationships are misrepresented to the laboratory, a false negativemay result.

i3ship Ana*)sis


30/75

The most difficult kinship analysis is to determine if two individuals arefull or half siblings when those individuals are the only ones availablefor !"A testing. f the number of tested siblings increases or otherrelatives become available, the degree of certainty increasesdramatically. t also is easier to determine if two individuals are full

siblings or completely unrelated. Advanced genetic methods andspeciali*ed computer programs are required to reliably solve thesecases. t may be necessary to test twelve or fifteen or more genes inorder to make a determination.

Crimina* Paternit)

The most common and most disturbing criminal paternity cases arethose involving the se#ual abuse of a child. %aboratory testing is not

possible in many of these cases because the abuser is a relative orclose friend of the family, and the crime comes to light after thephysical evidence is gone. Sometimes these unlawful se#ual contactsdo result in a pregnancy, especially if they continue over an e#tendedperiod of time. n this instance, the !"A testing is usually routine andcan be performed on the child or an abortus that is older than si#weeks. !"A testing may be especially useful if there are closelyrelated suspects, such as a father and son or two brothers. (riminalcases also may require the application of non-routine statistical andreporting techniques depending on the requirements of the court.(ases where the victim of se#ual abuse is underage rarely go to court.

The offender usually will receive a lower prison term if he pleads guilty.(onviction in the face of !"A evidence is almost inevitable. n caseswhere the victim is an adult and consent is a defense, the defendantmay change his story when he reali*es that the !"A testing results arealmost impossible to refute. The jury is usually unaware of this changeof story and so often will vote to acquit the defendant.!"A parentage testing has been questioned by the courts in only ahandful of cases. Typically, less than or C percent of civil cases resultin a case contested in trial. The opportunity to have testing repeatedby the same or another laboratory usually can settle most disputes tothe satisfaction of all tested parties. n criminal cases, when the testing

is scrutini*ed e#tensively in court, it is almost universally accepted.

/enetic Reconstruction or Re+ersePaternit)


31/75

These techniques often are used in homicide investigations. Supposethat there is a tissue sample from an unidentified deceased individual,or a stain from a crime scene and investigators have an idea of whomay be the sample5s source. f the decedent5s parents or children areavailable, it will be possible to compare the related survivors5 !"A

profiles with that of the deceased to conclusively determine whom theremains came from.

Test Resu*ts

8aternity testing is possible because everyone has two copies of theirgenetic information, half from each of the parents. f a child has a genethat didn5t come from the mother, then it must have come from thatchild5s biological father. f the alleged father does not have that gene,

the obligate gene as it is called, then he cannot possibly be the father.@e is e#cluded from the group of men who could be the father of thechild. There is one e#ception. f a mutation has occurred, which meansthe gene changed when it was passed from the father to the child.These mutations rarely occur, but they occur often enough thatstatistical methods are available for dealing with them and two !"Ae#clusions are required before reporting out a !"A paternity testresult.The other possible outcome is that the alleged father has the obligategene and so is included in the group of men who could be the father ofthe child. n effect, what paternity testing does is e#amine a series of

genes until the group of men who could be the father of the child isnarrowed down to the point that we can, for all practical purposes, becertain that he is the father. n !"A testing, this can be accomplishedby testing as few as three highly variable genes.Two different ways are used to report the two possible testingoutcomes. An e#clusion is reported as a simple statement that thealleged father is e#cluded from the group of men who could be thefather. n the case of an inclusion, a statistical analysis is performedbased on how common or rare are the obligate genes. These are thegenes that the alleged father could have contributed to the child andthat must have come from the child5s biological father. The

inclusionary results are e#pressed in the following two ways'

The Paternit) In&e requires that the facts or data presented be Bof a typereasonably relied on by e#perts in the particular field.B 7ule ;4>e#cludes evidence that would cause undue prejudice or confusion.8roponents of the $ederal 7ules approach to admissibility believe thattaken together, these rules address all the concerns embodied in the$rye rule.

Dau3ert +. !erre** Do#

(ritics of the $ederal 7ules fear that the courts may be openingthemselves to Bjunk scienceB by rela#ing $rye, but a landmark caseheard by the Supreme (ourt in 22> rejected that claim. n !aubert v.


38/75

1errell !ow 8harmaceutical, nc., the court unanimously held that the$rye rule was incompatible with and had been superseded by theadoption of the $ederal 7ules. t found that Bvigorous cross-e#amination, presentation of contrary evidence, and careful MjuryNinstruction are the traditional and appropriate means of attacking

shaky but admissible evidence.B Trial courts also could still rendersummary judgments and directed verdicts where there was aninsufficient showing of reliability.The effect of !aubert on states where rules mirror federal standardshas yet to be felt. @owever, it is reasonable to assume that e#perttestimony on !"A will be admissible after a threshold finding that it isrelevant and reliable. !efendants in these jurisdictions will have aharder time suppressing !"A evidence, although stiff challenges to itsadmissibility will undoubtedly continue, at least for the near future.!aubert will have little or no effect on states where $rye still prevails.Some of the states even have supreme court decisions affirming $rye.

The most recent state to affirm $rye is (alifornia in the %eahy case,decided in ctober 22;. n these states, new high court decisions orlegislation are the only means to change admissibility standards.

"e%is*ate& A&missi3i*it)

6y the $all of 22;, eleven states had statutes mandating theadmissibility of !"A evidence. 1aryland became the first state to do sofollowed by 1innesota, %ouisiana, and "evada, all in 22. 1ost of the

legislation contains language that !"A testing is acceptable Bwithoutantecedent e#pert testimonyB that it is Ba trustworthy and reliablemethod.B Arguably, these laws do not cover !"A analysis methodsintroduced after their passage, and the defense may still challengelaboratory performance and the statistical interpretation of results. Asa more sophisticated defense bar mounts increasing numbers ofe#pensive challenges to !"A evidence, it is likely that additional statelegislatures will address this issue.

E


39/75

worst of them. t is difficult not to conclude that some of theseindividuals are willing to stretch or ignore the facts, distort the science,and become Bliars for hire.B 1any of these witnesses derive asubstantial amount or even the bulk of their income from testifying,which should be considered in determining their credibility and

weighing their testimony. ne (alifornia judge bemoaned the use ofsuch witnesses by candidly calling them the beneficiaries of Ba welfaresystem for academics.B A recent article on the ethics and responsibilityof e#pert witnesses suggests the following criteria for qualification'

n%ergra%uate an% gra%uate %egrees in the re!e'ant $ie!% o$ e5pertise;

Specia!ize% training in the sub9ect area as it re!ates to $orensic science;

Some training in $orensics;

,ro$essiona! !icenses or certi$ications reuire% by pro$essiona! groups in the

e5pert7s %iscip!ine;

E'i%ence o$ e5perimentation/ teaching/ an% pub!ication ithin the specia!ty area;

an%

,rior %iscip!inary e'i%ence %irect!y re!e'ant to the issues being consi%ere%.

ther elements that help to determine an e#pert5s qualificationsinclude' post-graduate training, publication in peer-reviewed scientificjournals, the development of accepted tests and procedures,membership or leadership in appropriate scientific societies, and, onlylastly, e#perience as an e#pert witness.#en years ago the Ca!i$onia case ,eop!e '. 4ron a%%e% criteria that has pro'en %i$$icu!t

to app!y/ ie./ that the itness 8must a!so be impartia!/7 that is/ not so persona!!y in'este%in estab!ishing the acceptance o$ a techniue that he might not be ob9ecti'e about

%isagreements ithin the re!e'ant scienti$ic community.8 Neither shou!% a itness be soin'este% in %enigrating a techniue that he e5aggerates the %isagreement ithin thescienti$ic community. ,robab!y the best ay o$ gaining the testimony o$ impartia!

itnesses is $or courts/ rather than !itigants/ to appoint an% pay $or e5pert itnesses. Such

is the practice in many other countries.


40/75

=C? the statistical interpretation of data. The focus of the attacks onadmissibility have changed over time. As one objection was knockeddown, !"A opponents came up with another. The quality andrelevancy of the arguments and of the e#perts is decreasing, havinggone from population geneticists to bio-statisticians to statisticians

from completely unrelated fields. &hat follows is a summary of themost frequently heard complaints about !"A typing and the responsesto them that might be e#pected from forensic scientists. 1oretechnical objections, such as bandshifting and laboratory qualityassurance, are addressed in (hapter Three.

Conf*ict of Interest

8rosecution and defense e#pert witnesses in !"A cases are arguably

the most contentious and disparaging in the business. Several judgeshave remarked that $rye hearings over !"A can be e#tremely vicious.Among the charges and countercharges hurled back and forth is thatthe opposing witnesses should be disqualified from testifying becauseof a conflict of interest. To a certain e#tent, both sides are correct. Theprosecution believes that defense witnesses in !"A hearings oftenhave a vested interest in making sure that the subject stayscontroversial so that they can continue the lucrative practice oftestifying. The defense often believes that a practicing forensicscientist has a built-in bias or predisposition toward the prosecution5sside because of the close working relationship between crime labs and

law enforcement. ndeed, criminalists often are police employees.f the forensic scientists are leaders in their field, they may be subjectto a further conflict of interest. f they have developed or inventedtechniques or tools, they may have a proprietary interest in advancing!"A testing. They may have financial holdings in !"A labs or mayhave received grant funding from public or private agencies. (ertainly,a jury is entitled to know about all of these connections which shouldbe fully disclosed. At the same time, courts acknowledge that, Bsimplybecause learned e#perts earn a living with their e#pertise should notprohibit the admissibility of their opinions,B as the court ruled in arecent "ew )ersey case.

Inte%rit) of pecimen

pposing e#pert witnesses try to raise doubts about the way !"Aevidence was gathered and tested claiming that contamination mayhave occurred. The usual argument is that the underlying procedures


41/75


42/75

laboratory performance. n most cases these steps should lead tocorrective action long before a catastrophic error has occurred.1inimi*ing laboratory errors requires a quality control program such asthe ones which already are in place on a voluntary basis in the forensiclaboratories. Almost all forensic !"A laboratories participate in

programs which include proficiency testing and confirm that aminimum level of performance has been achieved. #ternal proficiencytesting also provides an ongoing comparison of inter-laboratorymeasurement error. These programs, led by the American Society of(rime %aboratory !irectors %aboratory Accreditation 6oard are rapidlygaining momentum.n the forensic field, the final arbiters of quality are the courts wheree#perts under e#amination and cross-e#amination submit their resultsto the scrutiny of the opposing e#perts and the judge and jury. Thisadded level of scrutiny is necessary to ensure quality forensic work andincludes review of casework, retesting, and observation of particular

tests by opposing e#perts. 6oth sides need to have equal access toforensic e#pertise in the interests of fairness and justice.

Popu*ation /enetics Estimates

The most contentious debate in forensic !"A involves the use ofstatistics to estimate the rarity of a given !"A profile. This is to bee#pected because the e#traordinary rarity of these profiles is whatgives them their conclusiveness as evidence. The rarities of the

genetic profiles depend on the number of genes e#amined =usuallyfour or five, often more?. The frequencies of the results of each geneare multiplied to reach a combined profile frequency or the finalestimate which is presented to the court.(ritics contend that among certain ethnic sub-groups, there may bearrangements of gene frequencies that differ markedly from thosefound in the general population.

Chapter >:

PEOP"E . ORENT(A" ,A!E I!PON

Peop*e +. Orentha* ,ames impson: TheTria* of the Centur)

At no time in history has someone as well-known and well-liked as .).Simpson been accused of such a heinous crime as slitting his e#- wife5sthroat, almost decapitating her, and the knifing to death of her friend


43/75

who happened on the scene. $or the American public, obsessed as it iswith celebrity, the Simpson case is proving to be a riveting spectacle.As this book goes to press in !ecember 22;, the trial itself has yet tostart and but every legal proceeding e#cept for jury selection has beenbroadcast live, covered thoroughly, and analy*ed e#haustively.

There is no escaping the Simpson trial. 1anagement consulting firmsare predicting that once the trial starts, an epidemic of BSimpsonitisB isgoing to lower productivity in the American workplace. The case is thestaple of television and radio shows, articles, editorials, andcommentaries, jokes, and cartoons. ven !oonesbury has assignedseveral characters to cover the trial and they are staying in a Bbigmedia encampment called .). (ity.B The case is being used as thebasis for law classes at @arvard, :ale, +.(.%.A. and several other lawschools =.). 4?. The trial will be broadcast on S8", the all-sportscable network. t is likely that by the time the trial is over, it will havereceived more news coverage than any event since the 9iet "am &ar.

As part of this media fren*y, !"A typing, the crucial evidence in thecase, is receiving unprecedented attention. To people in the forensicsciences, this attention is greeted with the hope that the Simpson casewill be the acid test for !"A in the courts, at least in the court of publicopinion. At the same time, there is some concern that the defense ismounting the most concerted and well-funded attack ever on !"Aevidence. &ill they succeed in confusing the judge or the juryJ Thedollars versus !"A showdown at the Simpson evidentiary hearing andthe trial will be a real test of the criminal justice system.The pre-trial hearing, called the Gelly-$rye hearing in (alifornia, wherethe judge will decide on the admissibility of the !"A testing was

initiated by the defense motion to e#clude the !"A evidence. Thatdocument was submitted on ctober 3 and runs 4 pages. Themotion is well written and goes back to the beginnings of !"A testingto cite a wide variety of articles, affidavits, and testimony whichchallenge the validity of !"A. t quotes two "obel laureates, one ofwhom will testify for the defense that the 8(7 test is not suitable forforensic use. This motion was rebutted by two documents from theprosecution, one asking the court to admit the 7$%8 testing and theother 8(7.t is highly unusual for jury selection to precede the Gelly-$rye hearing,as was done in the Simpson case. 6oth sides usually want to know the

outcome of the hearing prior to selecting a jury because the judge5sruling on admissibility affects the way in which they questionprospective jurors. )udge to did not really have a choice in this case,however. All of the !"A test results were not yet back from thelaboratories, so the full e#tent and kinds of evidence could not yetbeen determined. At this stage, an evidentiary hearing would havebeen able to address only generic and not specific issues.


44/75

The Gelly-$rye hearing, originally scheduled for "ov. , and then !ec.C, 22;, has now been postponed until )anuary 3, 223 at theearliest. ven this date might prove unrealistic because 8eter "eufeld,co-counsel of the Simpson !"A defense team, is tied up with a murdertrial in "ew :ork. Another unrealistic e#pectation is that the Gelly-$rye

hearing will only take a month. 6etween C4 and >4 witnesses havebeen identified. f they averaged two days of testimony apiece, a twoto three month hearing would ensue. n addition to defense andprosecution witnesses, )udge to may call his own e#perts to testify. fhe does, ric %ander from 1..T. is bound to be one of them. to is beingadvised on !"A issues by )udge !ino $ulgoni, a %os Angeles Superior(ourt judge and former prosecutor speciali*ing in !"A issues.6oth sides did get some early insight into how prospective jurors feltabout !"A evidence. As part of jury selection, all members of the jurypool were required to complete a D-page questionnaire probing theirviews on such subjects as race, religion, sports, politics, science, and,

of course, !"A. f the C2; questions, the eight dealing with !"A areprinted at right.The jury that was chosen for the Simpson case is composed of eightAfrican-Americans, two @ispanics, one American ndian-(aucasian, andone (aucasian. $our are men and eight are women. The jury includestwo people with more than a high school education. 6ased on theirquestionnaires, the Simpson jury, by and large, does not find !"Aevidence to be reliable.n the Simpson case, it is a victory for the defense to have the juryselection come first. The longer the jury sits, the more chance there isfor a mistrial, especially if the jury is sequestered. very time there is a

mistrial, it is a victory for the defense even though the prosecutionmore often than not convicts the ne#t time around.)udge to was probably thinking about the length of the trial and thepotential for BcontaminationB of the j

Documents

DNA in the Courtroom