Recross-examination of Dr. Robin Cotton

📄 Recross-examination of Dr. Robin Cotton — Thursday, November 14, 1996

Address:

C:\DEPT103\CIVIL\1996\NOV\14\RECROSS-EXAMINATION-OF-DR-ROBI.DOC

TRIAL

▲ Day 15 of 57

◀ Redirect examination of Dr. Robin Cotton Recess ▶

Witness: Dr. Robin Cotton
Examiner: Robert Blasier
Called by: Plaintiff • Date: Thursday, November 14, 1996 • Utterances: 56

Robert Blasier recross-examines DNA expert Dr. Robin Cotton, pressing her on the limitations of forensic DNA statistics. He extracts admissions that match windows are used because measurements can't confirm true identity, that population frequency statistics break down entirely for relatives, and that children of OJ Simpson and Nicole Brown Simpson could share banding patterns with their parents — raising the possibility that DNA found in the Bronco could have come from the children.

# 1 Q:

All right.

Let's just talk about the one percent where there's variations. That's 60 million base pairs; isn't it?

# 2 A:

Yes.

# 3 Q:

And you only look at extremely small percentage of that, don't you?

# 4 A:

Yes that's what I just said a few minutes ago.

# 5 Q:

And the 99 percent, that's the same? That's an estimate too; isn't it?

# 6 A:

Yes, it is.

# 7 Q:

Now, Mr. Lambert asked you a question that you only look at the portion that is unique from person to person. That's not correct, is it? You don't look at the portion that's unique from person to person, do you?

# 8 A:

Well, you --

# 9 Q:

That's -- yes or no?

# 10 A:

No, it's not. I can't answer that yes or no.

# 11 Q:

The fragment that you look at, are they unique only one person has the fragment?

# 12 A:

No. No on these tests, the pieces that we're looking at are not necessarily unique, but they are.

# 13 Q:

That's what you use? Go ahead.

# 14 A:

They are as a group, part of what makes someone unique.

I didn't mean to imply that a PM result for example, was unique to a given individual and hopefully I haven't given you that impression.

# 15 Q:

And now, let's talk about medical application transplants. There's a is substantial difference between forensic use of DNA technology and medical use; isn't there?

# 16 A:

There are some differences.

# 17 Q:

You don't, for transplants, you don't go and scrape something, a piece of blood off the ground, do you, to analyze it?

# 18 A:

Presumably that's not how they're getting it.

# 19 Q:

You always have clean samples and you always know where they came from, don't you?

# 20 A:

Actually, you should always have clean samples, yes. And do hospitals occasionally mix things up, yes.

KEY QUOTE

# 21 Q:

Okay. And that gets back to error rates, doesn't it?

# 22 A:

For the hospital, yes.

KEY QUOTE

# 23 Q:

And errors occur much more frequently than 1 in 530 billion.

# 24 MR. LAMBERT:

Objection. Irrelevant.

# 25 MR. BLASIER:

I'll withdraw it.

# 26 Q:

(BY MR. BLASIER) When you're doing bone marrow transplants, you have an unlimited amount of DNA to work with. You can do multiple tests to see if you get the same result, correct?

# 27 A:

I wouldn't say you have unlimited, but you possibly have enough to do multiple tests.

# 28 Q:

Okay. You're not estimating a frequency in a population for a bone marrow transplant? You never do that?

# 29 A:

There's no need to do that. All you need to do is make sure that the donor and recipient have different patterns.

# 30 Q:

So the whole part of this technology that generates these large numbers is relatively unique to the forensic area; isn't it?

# 31 A:

Possibly so, yes.

# 32 Q:

And this technology wasn't developed in the forensic community, was it? It was all developed in the medical and research community?

# 33 A:

It was all developed in the research community.

# 34 Q:

Now, you used a phrase, as long as these fragments are close enough to be considered a match. Isn't it true, what you mean by that is because of the measurements and precision, I can't tell you whether they are the same or not so I'm going to allow myself a window to call something a match when I can't really tell you it's the same, correct?

# 35 A:

Yes.

# 36 Q:

Now, did the steering wheel -- Are your saying that might be DNA from another person, some other time and not blood?

# 37 A:

I'm just saying, yes, there's no way for me to know. I didn't collect that sample, number one. So even if there was blood on that steering wheel, could there be something under that, some other cells under that? Sure. I don't -- it's a steering wheel. People are going to grab onto it with their hands.

# 38 Q:

That could be true for the console too, couldn't it?

# 39 A:

I suppose it could be, sure.

# 40 Q:

Do you have kids in a car that sneeze or bloody noses and wipe them on the console, you're going to get their DNA there, aren't you?

# 41 A:

If they do that, you would get their DNA there.

# 42 Q:

O.J. Simpson and Nicole Brown Simpson's kids are going to have very similar if not the same banding patterns as their parents?

They get half of their DNA from one, and half from the other, correct?

# 43 A:

Banding patterns, they will share characteristics, but the banding patterns wouldn't be the same as either Nicole Brown or Mr. Simpson for the PCR tests.

The kids, theoretically, since those characteristics are not so powerful at distinguishing one from another, could any of the kids have as a type that ends up being the same as one or the other parent, yes, they could.

# 44 Q:

And these statistics don't work at all where you're talking about possible sources from relatives, do they?

# 45 A:

Well, you would do a different calculation.

# 46 Q:

These calculations don't work for that, do they?

# 47 A:

It isn't that they don't work, it's that they wouldn't be appropriate.

# 48 Q:

They don't give you an accurate answer, do they? They don't give you an accurate estimate, do they?

# 49 A:

They don't tell you anything about relatives.

KEY QUOTE

# 50 Q:

Okay. Let's talk about number 29, the blood on the steering wheel. There was a substrate control taken, that is a sample taken from right next to that blood stain, correct?

# 51 A:

I believe so.

# 52 Q:

Was there any DNA on that?

# 53 A:

No.

# 54 Q:

Isn't that an indication that what was taken in that blood stain, that the blood in there from an unidentified third person, is in that blood stain and wasn't there before? Wouldn't you, if it had been there before, you'd expect to find it on the substrate control, wouldn't you?

KEY QUOTE

# 55 A:

You might.

# 56 MR. BLASIER:

Thank you. No further questions.

Temperature

procedural

Key Quotes (4)

Robert Blasier

Isn't it true, what you mean by that is because of the measurements and precision, I can't tell you whether they are the same or not so I'm going to allow myself a window to call something a match when I can't really tell you it's the same, correct?

Forces Cotton to confirm that a DNA 'match' is not a true identity determination but an approximation within a measurement tolerance window.

Dr. Cotton

Yes.

Her one-word confirmation of the above is a significant concession on the fundamental reliability of match calling.

Dr. Cotton

They don't tell you anything about relatives.

Concedes that the astronomical population statistics are meaningless when the question involves family members — directly relevant to the Simpson children's DNA in the Bronco.

Robert Blasier

Isn't that an indication that what was taken in that blood stain, that the blood in there from an unidentified third person, is in that blood stain and wasn't there before? Wouldn't you, if it had been there before, you'd expect to find it on the substrate control, wouldn't you?

Uses the clean substrate control result to argue that the unidentified DNA component in sample 29 was deposited at the same time as the blood, not a background contaminant.

Evidence (3)

Informal

Sample 29 — blood stain from the Bronco steering wheel

discussed

Informal

Substrate control taken adjacent to steering wheel blood stain — returned no DNA

discussed

Informal

Bronco console as potential source of children's DNA

discussed

Notable Exchanges (3)

Robert Blasier•Dr. Cotton

Blasier walks Cotton through the distinction between forensic and medical DNA use, extracting that bone marrow transplant work never generates population frequency statistics — that large-number statistical framework is unique to forensics and was not developed there.

strategic

Robert Blasier•Dr. Cotton

Blasier raises the possibility that Simpson and Nicole's children share banding patterns with their parents, and Cotton concedes the PCR statistics do not apply to relatives at all — they 'don't tell you anything about relatives.'

revealing

Robert Blasier•Dr. Cotton

Blasier uses the clean substrate control for sample 29 to argue that unidentified third-party DNA in the steering wheel stain was deposited contemporaneously with the blood, not pre-existing background contamination. Cotton gives a hedged 'you might' rather than a firm denial.

strategic

Light Moments (1)

Robert Blasier

Blasier asks whether kids in a car who sneeze or get bloody noses and wipe them on the console would leave DNA there; Cotton confirms flatly 'If they do that, you would get their DNA there.'

Credibility Attacks (1)

⚔ Dr. Cotton

concession extraction

Blasier systematically draws out that DNA 'matches' involve an imprecision window, that forensic population statistics fail entirely for relatives, and that the technology producing billion-to-one odds was never designed for or validated in forensic contexts.

Objections

1 objections (0 sustained, 0 overruled)

Proceeding 8269 • 56 utterances • Plaintiff witness

Civil Trial

Department 103