DNA Fingerprinting Lecture Notes
Extra Credit Opportunity
Discussion section activity involves writing and revising multiple-choice questions in small groups.
Based on course learning objectives on the website.
Involves critiquing and revising questions from other groups to prevent "gaming" the system.
Points awarded for question quality and critiques (up to 14 points total).
Good questions may appear on the final exam.
Grading:
Up to 8 points for questions (up to 4 points each).
Up to 6 points for critiques (3 points each).
Introduction to DNA Fingerprinting
Transition from previous topic of kindness to DNA fingerprinting.
One of the most abstract ideas in biology with significant practical application.
Originated in England in the early 1980s.
The First Case of DNA Fingerprinting
1983: Rape and murder case of 15-year-old Linda Mann in England.
Police had no suspects or leads.
Three years later, a similar crime occurred nearby involving another 15-year-old girl.
Evidence suggested the same perpetrator committed both crimes.
A suspect confessed to the second murder but denied involvement in the first.
Biologist Alex Jeffries discovered variable regions in DNA that could be used for identification.
Comparing DNA Fingerprinting to Traditional Fingerprinting
Traditional fingerprinting relies on physical fingerprints left at crime scenes.
DNA fingerprinting uses biological material (tissue, cells, blood) to identify individuals.
White blood cells in blood samples contain DNA.
The Process of DNA Fingerprinting
Involves identifying highly variable regions of DNA.
Transforming the variable DNA into interpretable data for court cases.
Jeffries examined semen samples from both crime scenes.
Goal: Find a part of the DNA that varies significantly between individuals.
Human DNA is mostly identical; the challenge is to find the variable regions.
Analogy: Finding something unique like a UCLA ID number to distinguish individuals.
Ensuring uniqueness to avoid convicting innocent individuals based on chance matches.
Initial Findings
Semen from both crime scenes matched, indicating the same perpetrator.
The DNA did not match the person who confessed.
First exoneration based on DNA fingerprinting.
The Investigation
Police requested voluntary tissue samples from local males aged 18-36/40.
Potential problems with this approach:
The actual perpetrator is unlikely to volunteer.
Testing samples could reveal other unrelated crimes.
In the U.S., such a request without probable cause is considered an unauthorized search.
The Breakthrough
5,000 people volunteered samples, none of which matched.
Someone overheard a conversation in a pub about a man being paid to submit a sample in someone else's name.
The substitute was found, and his DNA matched both crime scenes.
Conviction
Colin Pitchfork was convicted based on DNA evidence.
He had a history of arrests for indecent exposure.
Sentenced to 25 years to life.
Initially granted parole, but re-arrested shortly after for another crime.
Parole was revoked, and his sentence was extended.
Standard Fingerprints vs. DNA Fingerprints
Regular fingerprints are converted into numerical data based on bifurcations, ridges, dots, and ridge endings.
Fingerprints from identical twins are similar but distinguishable.
Influenced by both genes and environment.
DNA fingerprinting is identical in identical twins, as their DNA is the same.
DNA Basics and Variation
Most human DNA is the same across individuals.
Must find the small part that varies to identify individuals.
Looking at differences, like one's UCLA ID card.
DNA fingerprinting is used in criminal cases, paternity, exoneration of wrongly accused, and immigration.
Also used in identifying species in cases of fraud or regulation.
DNA Structure
DNA is a long, linear strand composed of four bases: adenine (A), thymine (T), guanine (G), and cytosine (C).
It is a double helix, with two complementary strands.
A pairs with T, and G pairs with C.
Genetic information is encoded in the sequence of these bases.
The sequence of base pairs is read down the strand and encodes the genes.
Each cell contains 23 chromosomes in duplicate.
DNA Variability
99.9% of DNA is identical between individuals.
0. 1% difference equates to 3 million base pairs.
DNA fingerprinting focuses on these differences.
Goal of DNA Fingerprinting
Determine if DNA from a crime scene matches a suspect's DNA.
Distinguish between a true match and a coincidental match.
Short Tandem Repeats (STRs)
Alex Jeffries focused on short tandem repeats (STRs).
STRs are repeating units of 4-5 nucleotides (e.g., ATCG).
The number of repeats varies between individuals.
Example: ATCG repeated multiple times at a specific location.
Individuals inherit different numbers of repeats from their parents.
More alleles with STRs versus eye color(2) and blood type(3)
Non-Coding Regions
STRs are typically located in non-coding regions of DNA.
90% of DNA does not code for anything.
The function of non-coding DNA is debated (regulation, "junk DNA").
An individual has two alleles for each STR, one from each parent.
Probability of a Match
If an STR has ten equally common alleles, the probability of a random match is calculated.
Example: If a perpetrator has a 10-repeat and 20-repeat STR, the probability of a random person matching is .
This calculation raises questions about sending someone to jail based on a single match.
Example with UCLA Population
With 50,000 people on campus, 500 might match, so that's not enough evidence.
Inheritance of STRs
STRs are inherited like alleles for physical traits.
Parents pass on one allele to their offspring.
Mendel's law of segregation applies.
STR and Mendel
Analogy: A mother with (10, 10) STR genotype and a father with (12, 14) STR genotype, their daughter can only have (10, 12) or (10, 14), following Mendel's laws of segregation.
Combinations of STR
Illustration of number of STR combination between two parents.
Take Home Message:
Multitude of alleles.
Each person has only two like blood type.
Each person puts one of their alleles into their gamete.
At fertilization, you re establish the diploid number.
Everything about this is the same as what we've been talking about for how Mendel's stuff works, the alleles that we do or pedigrees that we're trying to solve.
Except that we're talking about something that's not a physical trait.
Multiple STRs for Increased Accuracy
Different STRs exist in other places in DNA.
STR two, for example, has a different sequence (e.g., AGAT).
Each person has a genotype for each STR.
Cumulative Probability
If STR one has ten repeats and 20 repeats, and STR two has five repeats and 15 repeats, the combined probability is calculated.
The probability for STR two matching with ten different alleles is . With five potential alleles with STR one, the cumulative probability is .
Likelihood Comparison And Jury
The jury and members have less comprehensive knowledge to make an informed decision.
Numbers Game
What are the odds or likelihood with multiple STR. Third STR.
It would be
Should we send them to jail?
You say no, they are 8 millions people in Los Angeles and 8 people in LA probably match it.
Then prosecutor said a male did it, and he ran.
If it is expensive to do the analysis, let's look at the fourth one!
One out of 100 million, one out of 10 billion
330 million people live in US.
I don't know, maybe there will be a few people on earth that would match it, but do we send someone to jail.
But most of them don't have passports and have never been to the US, most of them are female, most of them are not in that age range, you play these numbers game up and down back and forth.
Creating the DNA Fingerprint
The person figured out they can take the DNA they get from the scene and amplify the DNA.
First thing they do
If we heat up the DNA, it separates, then throw in A's and T's and G's and C's.
Pair-ups, splits apart. Then you double.
Doubling it again, double it in 20 minutes.
DNA MULTIPLICATION
Heating it up and cooling it back down quickly made it possible to double the DNA.
Called PCR.
They chop the DNA up on other side of the STR region that is going to repeat.
How big
If you have 2 repeats, and cut it at the beginning of the end, it's short.
If you have 50 repeats and you cut it on either side of the thing, it's pretty long.
You have different length of DNA based on how many times it repeats, and put it into something that is called a gel.
It will be either radioactivity or florescent.
Gel
Put all the DNA in the gel and the jello is microscopic pores.
They put current in the well from negative to positive, then they run a current through dna.
It will be pull to the bottom. The longer it is more gets down. Thing that determine are how long the piece if it goes through the pores.
The baby ones get through first.
Then dump it on top to see dna from the str.
When it is picture it will look like real dna in that region.
Then take the specimens and look, then look at STR 1, 2,3and 4.
Then you have what you saw at the court.
Convict someone here.
Look at alibi, check the motive, check things that this doesn't exist.
How it looks if babies got mixed up
DNA finger print shows gels and patterns with 2 bands, because there are 2 alleles.
Looking at these to figure out came from which parents.
Has to follow mendel's law of segregation, and say A has 2 band, one from their mother and father, who does that comes from.
Can come from a mix up at the hostpital.
What they are also paid for.
Sometime STRs overlap. This could be is this band and that band the same or are they far enough apart.
If STRs are a little similar.
How many do you compare?
At first, people only use two or three str.
Now US law dictates that you have to do 13 of them.
13 STRs MEANS 26 DIFFERENT ALLELES
The frequencies can keep in codis.
Subpopulation Similarity
What if we were looking at the DNA? Let's say we knew for certain that the person who committed the crime was from Ireland and the suspect we have was also from Ireland, but they didn't do it. Do you think they're more likely to have the same STR alleles just based on common ancestry, that they are Irish, or is everyone in the world equally likely to have them?
There are subpopulation similarities where everyone from Ireland for this SDR, they all have one or the other of these two rather than all ten.
It's tricky to decide. You want to make sure that whatever numbers you're using, they come from a reasonable dataset.
Do humans make mistakes?
Errors can be intentional.
Drop his hair in the hair brush in blood from the crime.
Then, the prosecutor will say, Well, hey, we're going to send it out to another lab and they get the same result. We're going to send it out to a third lab, a completely independent lab. They get the same result. Of course they will because it's already been contaminated.
Chain of of evidence or that it has been contaminated.
Chain of custody.
eyewitness were wrong 30% of the time.
Nonetheless, as I said, this is the most important and useful addition to criminal justice ever.