Will Duct Tape be Casey’s Sticking Point? – The DNA

Original article posted on October 29, 2009

There are reasons why “rules” are set on the law enforcement side of DNA forensics.  These “rules”, self-imposed from lessons learned over time, have to do with the quality of DNA sample, purity of sample (is it a mixture and can the DNA of multiple donors be extracted?), and quality of overall data return.  These rules have been put in place to prevent “false positives” or, simply put, to avoid convicting an innocent person based on spurious returns.  This guiding principle, that locks down the prosecuting side of a criminal case on what they can and can’t (with confidence) use as DNA evidence against an accused, is why the defense is currently treading on very, very shaky ground.

The first step in analyzing the situation of the “foreign DNA” on the duct tape at the crime scene is to make sure we understand basic genetics and, subsequently, DNA testing.  All we need is an elementary understanding – so that’s what I’m going to aim for imparting.

First let’s talk about genetics and what an “allele” is.  When the gleam in your daddy’s eye met up with the demure of your momma’s smile and created the “you” we all love, your daddy contributed what is called a haploid, via his sperm, and your momma contributed a haploid, via her egg.  A haploid is, in layman’s terms, half a person waiting to be created.  When the fertilization process “took” and created you, the haploid of your daddy found the haploid of your momma and they joined together “until your death do they part” to create a diploid – the whole person that is you.  In this awesome process of the creation of human life a miraculous joining of the “right parts” of the separate haploids took place.  In other words, the chromosome of your daddy that has the gene for hair-color sought out the chromosome of your momma that had the gene for hair-color and they hooked up – and the same for all the other chromosomes that make you – you.  Two chromosomes that carry the same type of genes (for instance, hair-color) are called homologous chromosomes.  So all the little chromosomes (eye-color, hair-color, etc.) found their homologous mates and “hooked up” for life – your life.  (We’ll stay away from eye-color because, believe it or not, it is one of the most complicated physical traits you have from a genetic standpoint.  That’s because your eye color is not determined by just one gene (i.e. one set of alleles) but by a combination of genes (multiple sets of alleles) and then it gets more complicated because you can have “partial expression” of each of the alleles in a gene.  That’s why a brown-eyed girl and a blue-eyed guy can have a green-eyed baby!)

On a chromosome (let’s stick with the hair-color gene, so we’ll just talk about the chromosome that has that one), there are multiple “gene loci” – i.e. the locations for the separate genes.  So on the chromosome that has your hair color, there may also be a gene that decides whether you’ll have curly hair, straight hair, etc.  These gene loci on one haploid mate up with the same gene loci on the other haplois during the haploid-to-diploid process.  So the gene for hair color that your daddy contributed is a “variation of that gene” or an allele.  And the same holds true for the “variation of that gene” your momma contributed.  So for every phenotypical (i.e. physical) trait you have, you get two alleles – one from momma and one from daddy.  These combine to create your “genotype” (i.e. the combination of two alleles that define that gene type for you) and that genotype, depending on the combination of the two alleles and the dominant/recessive natures of them, etc., creates your “phenotype” (i.e. the physical expression of that genetic trait).  So you get brown hair, or blonde hair, or red hair, etc.

For any given genotype/phenotype, there are two alleles – always, unless you’re a mutant.  Since Casey doesn’t have any gross physical mutations apparent, we’re assuming she is not some physically unique mutation and, therefore, she has 2 alleles for each genotype.  The same will hold true for Caylee, Cindy, etc.  After all, as bizarre as this case has been, we’ve not encountered any three-eyed characters in it…yet.  I’m driving this point home because in DNA testing, the results will come back in one of two ways for a pure sample of one person:  1.  two allele readings (represented by two numbers separated by a comma – for instance, 20, 24), or 2. one allele number – like, 20.  Then you have amelogenin testing which looks for the X and Y sex chromosomes for the purpose of identifying sex.  If the amelogenin test returns X, then the sex is female because we have two X chromosomes.  If it returns X, Y, then the sex is male.

Why can it come back one number some times when I just got through pointing out we all have two alleles for every genetic trait?  Because for every genetic trait you have (made up of the two alleles, one from daddy and one from momma), you are either heterozygous or homozygous in that trait.  If you are heterozygous then you got two different alleles from mom and dad (say brown hair and blonde hair) and your DNA result for that trait would return two numbers – i.e. 20,24.  If you are homozygous then you got the SAME allele from mom and dad (say brown hair, brown hair) and your DNA result for that trait would be listed as one number – i.e. 20 (because basically your reading would “overlay” the two alleles, because they are identical, and it would look like a single peak).

If a DNA return on a gene locus (location) returns more than 2 numbers, then you have a mixture of multi-donors.  So let’s talk about how that effects the “quality” or “soundness” of a DNA test result.  There is the potential to do what is called “differential DNA extraction” – which means, from a mixture, separate donor DNA can be “extracted” for the separate donors.  The success at pulling this off depends on a great deal of factors associated with the quality of the sample.  For instance, if the sample is a “trace” sample (just barely there), the odds of doing this can be reduced, because the “return” (i.e. “reading” at the gene locus) falls below the “rules” that law enforcement have placed on themselves to call it a “sound” return.  If the “types of cells” that constitute the DNA contribution from the different donors in a mixture sample are different, then the odds of success at differential DNA extraction is increased.  So, for instance, if it is a rape case, then you have two different types of cells contributing DNA between the alleged perpetrator and the victim.  The perpetrator’s contribution will be sex genes – i.e. sperm.  The victim’s contribution will be epithelial cells (i.e. skin cells).  Because extraction of DNA can be done by “cracking open” the sperm cells, this type of DNA mixture has a high degree of success in differential DNA extraction.  But when you get into a DNA mixture that is, say, made of skin cells from one donor and skin cells from another donor, the success rate of differential extraction starts dropping significantly.

So in the case of a mixture of skin cells from two donors, what the lab will try to do is determine if they can at least identify a primary donor and secondary donor based on the strength of the return for that allele.  So if the lab gets a return of three alleles at one gene locus, thereby confirming a mixture, if two of those alleles are of the same comparative strength, and greater than the third allele return, they can say “these two alleles belong to the primary donor”, this third one is the secondary donor.

So, after discussing these brief points on genetics, returns and “mixtures”, let’s focus back on a return that falls under an acceptable limit that allows it to be used to draw conclusions (this language you will see in a lot of places in the FBI forensics results).  As stated upfront, this is a self-imposed rule on the lab that requires that the strength or quality of a return meets a certain minimum requirement before they “make a call” on a match (i.e. use it to id some one).  In the DNA test reports released on September 29 of this year, and in the following tables used for illustrative purposes, you will see allele returns that are listed in parentheses, for instance (20, 24).  That parenthetical reference denotes that the returns for those readings were below the acceptable limits to be used to draw conclusions.  They may be introduced in court to show a “possibility”, but to present them as “evidence of a certainty” would be an ill-thought plan.

What would cause an allele, or alleles, to return below an acceptable limit.  A lot of factors.  First, the DNA could have been “trace” in the first place, and simply just “not there enough” to return a good reading.  The return may have been affected by “noise” in the testing – which can be from contamination in the sample, air bubbles in the sample, etc.  The return can be “stutter” which is a small peak that can occur before or after a true peak and are caused by the amplification process of the original DNA sample.  There are also other “false returns”, so to speak, such as “spikes” and “blobs”.  When looking at the graphs in the FBI DNA testing, ALL of these artifacts can be found at one place or the other in the myriad of tests performed.  NONE of us would want to be convicted, nor should we want to see some one else convicted, of a crime based on any of these artifacts in DNA testing.  But, in addition to “false returns”, the DNA of a sample (even if it were a “good” sample in the beginning when it ended up on whatever piece of evidence it was on) will degrade over time.  DNA, being  just like our own body, or the cotton discussed before – it decays, it gets attacked, it doesn’t stick around for ever if it is in an unprotected environment.  In the case of the crime scene where little Caylee’s body was found, the DNA at that location was under constant attack for the almost six months it lay in swamp water.  So by the time the FBI lab got the evidence that may have had strong indications of Caylee, or maybe Casey, in the beginning, that DNA had all but gone away from degradation due to the elements.

Are there legitimate ways to still use DNA returns that fall below the acceptable limit?  Yes, there are.  Those techniques include:  1.  Re-running the same test on the same sample (i.e. swab, bone marrow, etc.) to see if the low returns are repeatable.  This doesn’t necessarily make those returns usable for drawing conclusions, but it confirms those low returns were not anomalous returns created by noise, stutter, etc.  2.  Running the same test on a second sample.  So you run the exact same test on a second swab, or second sample of bone marrow, whatever, from the same source.  If the second sample has the same returns (which can vary in strength from the first swab), and you re-run both tests (i.e. you duplicate returns for sample 1 and then you duplicate returns for sample 2 – both samples from the same source), you can call these returns “good” – and basically remove them from the parentheses.  You have proven repeatability in returns for the sample.  This plays in as very important when dealing with degraded DNA – such as Caylee’s DNA samples from her little tibia.  It has degraded so badly that the FBI had to run tests on more than one sample from the tibia because her little alleles were so weak she could barely “speak to us”.  But through proving repeatability, the FBI confirmed it was little Caylee’s tibia.

So let’s take some of the test results and look at them to see how this plays into things.  The following table is constructed by splicing together the results from Caylee’s hairbrush (these results were used as the benchmark for her DNA for matching purposes), Casey’s buccal swab for matching, and the results on the samples from her tibia.  (The separate tables are located on pages 227 and 228 of this released discovery.)

So this particular composite table shows us several examples of what we’ve gone over so far.  First, the hairbrush that was used to establish Caylee’s DNA fingerprint had a mixture of similar cells from both Caylee and Casey.  This is seen in the two lines of the results for Q18-1.  Note there there are three alleles total in the second gene locus (D16S538).  They are listed as 9 and then (12), 13.  9 and 13 are at acceptable levels, 12 is below.  Extraction does not appear to be possible (and probably because of the fact the cells are of the same type).  But then we go to K1-1 and we get a good, pure sample of Casey’s DNA from her buccal swab.  In the D16S538 locus she has alleles 9, 12.  We can now ascertain that the 9, 13 from the hairbrush belong to Caylee.  (The low level 12 appears to come from Casey.)  In this case they would deem Caylee the primary donor and Casey the secondary donor on the hairbrush.  But note something important here about the “trickiness” of mixtures of like cells.  If they did not have the buccal swab of Casey as a benchmark, they would have initially (then not been able to correct it) had the (12), 13 alleles together and the 9 alone.  It was only after comparing to Casey’s DNA sample that they found the 9,13 alleles go together for Caylee.  This is the problem with mixtures…especially when one donor comes in “under the limits”.

Now, moving down our composite table to the last 4 tests listed, we get to the tests performed on Caylee’s little tibia.  To clarify what the separate designations are:

*Q61-1 is a segment of the tibia Q61

*Q61-1(2) is a sample from Q61-1 for testing

*Q61-1(3) is a different sample from Q61-1 for testing

What we see on these lines are the example of degraded DNA that produce returns that are under the acceptable limit, and then subsequent testing to establish repeatability of those weak returns.  So, not only did they test Q61-1(2) twice – to show that the returns were not anomalous – but they also tested Q61-1(3) twice to show that the results were repeatable for two different samples from the same source (Q61-1).  In doing this you will notice that when they confirmed repeatability with the second sample (Q61-1(3)), they no longer made the results parenthetical, because now they have followed proper procedure to confirm the findings.

Because the comparison between the Caylee benchmark testing from the hairbrush with that from the buccal swab of Casey confirmed that D16S538 is, in fact, the pair of alleles, 9, 13, we have a match between the tibia and Caylee’s DNA.

So now, let’s move to another piece of evidence that had a mixture.  The handle on Brian Burners’ shovel.  The following table is created from taking portions from the tables on pages 105 and 106 of the same pdf linked above.  The first thing to note is that this is a different type of test from the previous table.  The above table is from the COfiler output, but this one is from the Profile Plus ID.  You have to be careful because, if you’ll look at the headings on the two separate tests, they look at different gene loci.

Note that in the Profile Plus ID, the ambiguity that existed on the COfiler results is not present when you look at the results for the hairbrush Q18-1.  Caylee’s DNA is above the acceptable limits and is very distinguishable as the primary donor.  Casey’s two alleles are below the limit.  So differentiation was achievable on this test method.  So we have Caylee (Q18-1), Casey (K1-1) and Brian Burner (K3-1).  Then we have the results of the swab taken off the handle of Brian Burner’s shovel.  The results from the shovel “fail” all the way around.  In other words, you most likely won’t see the prosecution (unless further testing resolves the ambiguity, OR—well, you’ll see another scenario) using this test result to say “Casey’s DNA was present on the handle!” or “Caylee’s DNA was present on the handle!”  Why?  Because it falls outside of their rules for use to draw conclusions.  The sample is problematic in several ways.  With the exception of the X chromosome in the amelogenin testing which came back with an acceptable quality of detection, everything else on this test came back under the acceptable limits for drawing conclusions.  Add to this that it is a mixture (most likely of skin cells from Brian’s hands and probably Casey’s hands, not to mention possible other hands like Brian’s son, wife, etc.) that do not appear to be capable of being differentiated, and you have a pretty useless return.

Now, let’s look at the “contaminating DNA” returns on the duct tape pieces Q62, Q63 and Q64 found on Caylee’s skull.  The following composite table is created from tables on pages 366, 409, 423 and 433 of the same pdf.

Okay, the first thing to point out is that there are two samples from the duct tape:  Q62/Q63/Q64-1 and Q62/Q63/Q64-2.  These samples are called “composite samples” because what they did for each of these two samples is take one swab and swab all three pieces of tape and then test the combined swab as one sample.  K1-1 through K5-1 are the FBI employees who were tested to either eliminate or confirm them as sources of the trace DNA on the duct tape.  (It’s trace because everything but the X chromosome is under the limits, and not even all alleles were picked up.  When all alleles are not present it is called a “partial profile”, which we’ll get to in a minute.)  As can be seen, and is marked on the table, K5-1 was confirmed (through a two runs of the test) as the source for the DNA in the first sample.

Now, let’s look at the second Q62/Q63/Q64-2 test.  In the world of unusable DNA test results this qualifies as the most useless test result one can get.  Of all the alleles tested for only one is detected and is below the usable limit.  In addition, even the amelogenin test result showing an X is not strong enough to be used for confirmation.  So let’s talk about what it means to not have all the alleles present.  When the majority of alleles are indetectable on a sample, it means you are working with a source of DNA that is smaller than a single human cell. In other words, the source isn’t even the complete human cell from whomever it came from.  If the prosecution were to try to use a test result such as this one, they would be derided by every DNA expert in the U.S.  This is “nothing” in the world of DNA results.

As can be seen, three of the FBI employees tested have a 17 in this gene locus:  K2, K3 and K4.  The defense has picked up this one allele and hit the media circuit touting that it shows there was potentially “some one other than Casey or Caylee” in contact with the duct tape.  Well, the most logical response (as if the insinuation merited a response) to this tactic is to point out that 3 FBI employees involved in handling evidence have that allele.  And let’s be clear how easily this “less than a human cell” amount of DNA could have gotten on this duct tape:  try a flake of dandruff while down on your knees rummaging through the undergrowth for evidence; try swatting a mosquito off your cheek with a gloved hand that subsequently touches a piece of duct tape; try the natural hair loss process that every person’s body undergoes everyday.

So the defense would like the world to set aside the “rules” of DNA evidence usage that they would kick a fit about if the prosecution were to try to break them, and they want us to accept that a partial profile consisting of a single allele, that can’t even definitively be stated to be a “female”, because the source was less than a complete cell and the Y chromosome may have been indetectable, is sufficient evdience to support their theory that this proves “some one else did it”.

For one minute, let’s return to Brian Burner’s shovel handle.

Under normal circumstances the results of Q29-1 that was a mixture of Brian’s DNA (and possibly Casey’s or Caylee’s) would most likely not be presented in a court of law to try and “prove” anything because of it’s poor quality of returns and inability to differentiate the mixture.  But if the defense continues to pound on the “17-allele”, the fact that a mixture of DNA that has a 17 allele, and a strong return on the female chromosome, and which was taken from a shovel that was known to be handled by the accused, in what could possibly have been “management of the victim’s body”, will come into play.  And when this happens, the defense will have opened the door to allow the prosecution to make a connection between Caylee’s dead body, Casey, and the shovel.

This is why the “17-allele” may be a “catch-22” for the defense.

Valhall.

Several helpful references:

http://www.scientific.org/tutorials/articles/riley/riley.html

http://www.bioforensics.com/articles/champion1/champion1.html

http://www.ndaa.org/publications/newsletters/silent_witness_volume_10_number_2_2006.html

http://www.fathom.com/course/21701758/session2.html

http://www.ftldna.com/glossary.html

An Introduction to Forensic DNA Analysis

Related posts:

1. Will Duct Tape be Casey’s Sticking Point? – The Cloth
2. Will Duct Tape be Casey’s Sticking Point? – The Connections
3. Will Duct Tape be Casey’s Sticking Point? – Loose Ends
4. Will Duct Tape be Casey’s Sticking Point? – A Possible Source – Update
5. About Caylee’s Parentage

Tags: allele, casey anthony, caylee anthony, DNA, duct tape, forensics, genetics