Individuality and Parentage

1
Individuality and Parentage
2
Most species (sexually reproducing organisms)
harbor sufficient genetic variation - molecular
markers can distinguish each individual from all
others with near certainty. Individual
identification Polymorphic molecular markers
(with known pathways hereditary transmission)
parentage verification. Especially suitable for
such analyses are highly variable nuclear genome
markers with specifiable modes of Mendelian
inheritance. These markers include the allozyme
products of numerous protein-coding genes as well
as DNA-level fingerprints such as those
provided by minisatellite loci, RAPDs and
(especially in recent years) microsatellites.
3
Human forensics
The earliest efforts - involved typing various
blood groups and serum proteins - registered only
modest genetic variation - offered limited
evidence on individual identity and
uniqueness. At the DNA level - the first widely
employed approach - RFLP analyses of
minisatellite sequences, - one locus at a time.
DNA fragments at these minisatellite loci were
typically several kilobases in size and were
measured with some error. In practice, grouping
procedures were employed to pool fragments of
similar length into allelic bins whose widths
reflected magnitudes of experimental error across
replicates.
4
VNTR locus often exhibited 10-30 or more
differentiable bins of alleles in a typical
population sample. This extensive variation
carried a key consequence At any locus, the
probability of a genetic match between randomly
chosen individuals was low. Most human forensic
laboratories switched to short tandem repeat
(STR) sequences of microsatellite markers, and
this remains the primary method in use
today. Microsatellites are also highly variable,
but they offer several advantages over
minisatellite sequences Unlimited supply of
loci for examination Shorter fragments and
smaller tandem repeat units, PCR basis for the
procedure
5
A
B
C
D
E
Marker 1
6
DNA fingerprinting has become an integral
component of modern crime laboratories around the
world. DNA typing merely provides physical
evidence that assigns particular individuals to,
or excludes them from, particular tissue samples,
and thus, must be used in conjunction with
additional lines of evidence to resolve legal
issues such as criminal guilt or innocence.
7
In the Western judicial tradition, where a
suspect is considered innocent until proven
guilty. If the result is a match, what is the
probability of a spurious DNA match? At face
value, such probabilities calculated from DNA
forensic data are infinitesimally small. Thus, a
perfect multi-locus match is usually interpreted
as establishing genetic identity beyond
reasonable doubt. However, this type of
conclusion is based on some key assumptions whose
validity was questioned, thus placing DNA
fingerprinting itself on trial, not long after
the techniques debut. Consider an extreme
example in which a suspect belongs to a small
inbred community that differs dramatically in
allele frequency from North American Caucasians
overall. Use of a Caucasian database as a
reference for calculation genotypic probabilities
clearly would be inappropriate, and the direction
of error could work against an innocent defendant
(e.g., the likelihood of a genotypic match
between an innocent suspect and another member of
the local community who may actually have
committed the crime is much higher than the
probability of a match within a broader Caucasian
population).
8
In the late 1990s in the United States, a
national database known as CODIS (Combined DNA
Index System) was implemented. It involves 13
core STR loci that are now employed widely in
forensic analysis and which collectively yield an
average random match probability of less than one
in a trillion among unrelated individuals.
Mitochondrial and Y-linked markers are also
employed routinely when information is required
about particular human matrilines or patrilines.
Today, DNA typing via standardized batteries of
molecular markers is routine practice in human
forensics, and scientific controversies about the
evidentiary power of DNA have mostly faded to a
distant memory.
9
The first criminal conviction in the United
States based in part on DNA evidence came in a
1987 rape trial State v. Andrews, Orange County,
Florida. This case established a legal precedent
for the use of DNA tying to link a suspect to
biological material (blood, semen, or hair
follicles) left at a crime scene. One of the
earliest examples of DNA tying in a homicide case
was also unusually bizarre It involved a
mortuary worker accused of killing and
incinerating his estranged wife at a crematorium
in Kansa. Circumstantial evidence had implicated
the worker in his wifes death, but he staunchly
maintained that she had not been at the mortuary
near the time of her disappearance. However,
bloodstains discovered on the side of the
crematorium proved by DNA typing to match other
remaining tissue from the deceased woman. The
mortuary worker was convicted of aggravated
kidnapping and first-degree homicide. In another
early example of the power of DNA typing methods,
Hagelberg et al. (1991) used PCR to amplify DNA
sequences from the 8-year-old skeletal remains of
a murder victim. By comparing microsatellite DNA
markers in the remains with those of the
presumptive parents, the victims identity was
established.
10
Not all forensic applications of DNA typing
involve crimes this macabre, but molecular
genetic methods have provided powerful physical
evidence in thousands of homicides, rapes,
burglaries, assaults, hit-and-run accidents,
missing persons, identifications of war-atrocity
victims, and other cases. Forensic DNA methods,
validation studies, and empirical examples are
the focus of no less than three major scientific
journals Forensic Science International
(Elsevier Science), International Journal of
Legal Medicine (Springer-Verlag), and Journal of
Forensic Sciences (American Society of Testing
and Materials).
11
A genetic chimera is an individual composed of
a mixture of genetically different cells that
is, cells typically stemming from separate
zygotes. The phenomenon is quite rare in the
biological world. Far more normally in nature,
each multicellular individual is composed of
generically identical cells all tracing back
asexually through mitotic cell divisions to one
fertilized egg.
Chimerism is known even in vertebrate animals,
most notably in marmosets and tamarins. These
primates normally give birth to two fraternal
(non-identical, or dizygotic) twins per
pregnancy. In the first month of pregnancy,
however, the tiny embryos partially fuse for a
time inside the uterus , exchanging blood and
some other body cells. Although the fetuses
physically separate again before birth, molecular
fingerprinting assays (of the common marmoset,
Callithrix jacchus) have confirmed that each
individual continues to be a chimera of its own
blood cells plus those from its genetically
distinct sibling.
12
Molecular procedures for genetic assessment of
parentage are similar in principle to those used
to assess genetic identity versus non-identity,
but with the added complication that rules of
Mendelian transmission genetics must be taken
into account when comparing the genotypes of
sexually produced progeny against those of
putative parents.
Parentage analyses often address some version of
the following question Are the adults who are
associated behaviorally or spatially with
particular young the true biological parents of
the offspring in question? If the answer proves
to be no, a genetic exclusion has been achieved.
13
(No Transcript)
14
Using Mendelian molecular markers, parentage
verification can be achieved by exclusion.
Associated with such genetic exclusion are
statistical probabilities that are a joint
function of the variability of the genetic
markers employed and the biological nature of the
particular parentage problem (e.g., perhaps one
of the two parents is known from independent
evidence, such as pregnancy). Exclusion
probabilities may be either specific or average.
Specific exclusion Consider a neutral
autosomal locus with two equally frequent alleles
(A and B). In a large population at
Hardy-Weinberg equilibrium, about 25 of all
individuals would be homozygous AA, and another
25 homozygous BB. Suppose that molecular markers
how that an AA mother had an AA offspring. All
adult males in the population with genotype BB
can be excluded as the youngsters biological
father (baring mutation), so the specific
exclusion probability in this case is 0.25.
15
Average exclusion probability By contrast, is
the mean probability (or the expected proportion)
of excluded parents for randomly chosen
juveniles. A mean exclusion probability may be
higher than some specific exclusion values and
lower than others because it is calculated by
combining all specific exclusion probabilities
weighted by the expected frequency of each
parent-offspring pair in the population.
Biologists are often particularly interested in
average exclusion probabilities because they
indicate the strength of available genetic
markers for parentage exclusions (values above
0.99 typically are sought) and because these are
useful for comparing statistical power across
published studies. Many other statistical tasks
associated with various nuances of parentage
assessment have been developed. For specified
biological settings, maximum likelihood
approaches are available to categorically assign
individuals to their parents or to fractionally
assign parentage to multiple non-excluded adults.
Computer programs used to implement these or
other methods include CERVUS, FAMOZ, KINSHIP,
PAPA, PARENTE, PATRI, and PROBMAX.
16
Whether the actual mothers and fathers also can
be specified depends on the size and genetic
composition of the pool of candidate parents and
on the level of genetic variability monitored.
Sometimes a biological parent is known from
independent evidence and the problem simplifies
to one of paternity (or maternity) exclusion or
inclusion. In other cases, neither parent is
known with certainty prior to the molecular
study. Parentage analysis utilize cumulative
information from multiple polymorphic loci. In
recent years, microsatellite assays have largely
supplanted earlier methods of parentage
assessment. Even a few hypervariable
microsatellite loci in a population often display
sufficient genetic variation to yield combined
exclusion probabilities well above 0.99, thereby
offering exquisite information on biological
paternity and maternity.
17

• Maternity and paternity both uncertain,
  exclusions attempted.
• T.W. Quinn et al. (1987,1989) compared goslings
  within each of several broods of snow geese (Chen
  caerulescens) against their adult male and female
  nest attendants (putative parents) using genetic
  markers.

From their data, the following observation and
deductions were made. Two goslings in family 2
proved to be homozygous at some loci for alleles
not present in the female attendant. Such cases
excluded the putative mother and they were
interpreted to reveal instances on intraspecific
brood parasitism, or egg-dumping, where other
females (not assayed) must have contributed eggs
to the nest.
18
Other gosling proved to be homozygous for
alleles not present in the male attendant. Such
cases excluded the putative father and were
interpreted to reveal likely instances of
extrapair fertilization by other males. Some
heterozygous loci exhibited one allele not
observed in either nest attendant and a second
allele present in both attendants. Such loci
exclude one of the putative parents, but do no
alone determine which attendant is
disalloqed. Finally, some loci were homozygous
for alleles not observed in either nest
attendant, thus excluding both.
19
Marker A
Marker B
Marker C
Marker D
Father 102 102 153 153 210 216 219 221
Mother 100 106 153 153 210 218 221 221
Chick 1 102 102 153 155 210 216 219 221
Chick 2 102 102 153 157 210 216 221 221
Chick 3 100 106 153 153 210 216 217 221
Chick 4 100 106 153 153 210 216 221 221
Chick 5 100 102 153 155 210 210 221 221
Chick 6 100 102 153 153 206 216 217 217
Chick 7 100 102 153 153 210 216 219 221
20
2. Maternity known, paternity to be decided among
a few candidate males.
Burke et al. (1989) applied multi-locus DNA
fingerprint assays to the dunnock sparrow
(Prunella modularis), a species with a mating
system in which two males often mate with a
single female and defend her territory. In the
DNA fingerprints, those bands in progeny that
could not have been inherited from the known
mother were identified as paternally derived.
Then, the true father was determined by comparing
bands from the fingerprints of candidate sires
against these paternal alleles in progeny.
Mother
Chick
Father1
Father2
Profile of possible father
Marker A
Marker B
21
(No Transcript)