CA - Joey, Summer, Gianni, Joseph Jr McStay Murders - Feb 4th 2010 #15

[Tortoise]

The end is almost in sight. It looks like about 2 weeks left.

I'm just trying to think who there is left to testify -

Randolph Beasley (crime scene sponge guy)
Vlad Jovanovic (cell phones)
Dennis Shogren (accountant)
Gary Robertson (homicide investigator)
Dan Kavanaugh?
Tracey Riccobene?

Did they have anyone lined up for the Faro scan evidence, or anything else I've forgotten?

 

[Karinna]

Yeah, money and being “the” expert. Narcissism at its peak, IMO.

(Can you tell I didn’t get a warm-and-fuzzy feeling from him? Lol.)

Maybe because he is a scientist? They can come across like that because they have so much going on in their heads all the time, IMO.

 

[missy1974]

There were knots in the electrical cord but if we talk about the half bra in the grave results, there were no protected areas. His own testimony means touch DNA on the bra would have been erased.

RSBM

Do you know where the bra was found in the grave? I looked for a photo, I couldn't find one. I noticed that when Daugherty was showing him photo's, he showed him the grave with Joey and the cord (which wasn't all underneath him), but not one of the bra or the red strap... he tried with the red strap but it was objected to because 'there was no way to know it was the A25 strap' or something like that.

I had always assumed that the bra was found as if it was 'on' Summer, but when I went looking the other day, I couldn't find a photo or any testimony stating that.

 

[MrsPC]

I watched a youtube video of Dr Perlin trying to explain exactly how his TrueAllele software works. What stood out to me was the very first step in the process. The user of the software first asks questions (such as are there multiple contributors) & begins with a "prior belief." And then the software works on a likelihood of the DNA based on those questions and prior belief (and the data.) To my knowledge, Pujols nor Perlin were asked what questions and prior belief were asked/input at the initial stage in this case. And for those who say TrueAllele *results* are not subjective I say hogwash.

sub·jec·tive
/səbˈjektiv/
adjective

1. 1.
   based on or influenced by personal feelings, tastes, or opinions.
   "his views are highly subjective"
   synonyms: personal, personalized, individual, internal, emotional, instinctive, intuitive, impressionistic; More

 

[missy1974]

Maybe because he is a scientist? They can come across like that because they have so much going on in their heads all the time, IMO.

He reminded me of Dr. Rudin but he could actually put a sentence together LOL A bit condescending... to both attorney's ... way smarter than all of us, but comes across like an a$$ lol

I still think it's interesting that 90% of his cases are for the prosecution, and it's normally defense attorney's that are crying it's not valid, it's not reliable, not scientific. I found a great research paper online sometime in the last few days, it looked at the pro's and con's of probabilistic genotyping and TrueAllele, it was informative, but even it is looking more at it from a defense attorney/defendant perspective I think.

https://jolt.law.harvard.edu/assets/articlePDFs/v31/31HarvJLTech275.pdf

There is actually some references to some studies that were done dealing with manual interpretation and bias that is interesting, what results are when those that are doing the testing know details and don't know details, I jumped into some rabbit holes when I first read it looking for the other studies because they sounded interesting and it seems that in every trial with any sort of testing, it always comes up whether the expert did or didn't know what they were testing or if they were blind testing. I think very very rarely is the person 'blind testing' in the trials I have watched.

 

[MrsPC]

He reminded me of Dr. Rudin but he could actually put a sentence together LOL A bit condescending... to both attorney's ... way smarter than all of us, but comes across like an a$$ lol

I still think it's interesting that 90% of his cases are for the prosecution, and it's normally defense attorney's that are crying it's not valid, it's not reliable, not scientific. I found a great research paper online sometime in the last few days, it looked at the pro's and con's of probabilistic genotyping and TrueAllele, it was informative, but even it is looking more at it from a defense attorney/defendant perspective I think.

https://jolt.law.harvard.edu/assets/articlePDFs/v31/31HarvJLTech275.pdf

There is actually some references to some studies that were done dealing with manual interpretation and bias that is interesting, what results are when those that are doing the testing know details and don't know details, I jumped into some rabbit holes when I first read it looking for the other studies because they sounded interesting and it seems that in every trial with any sort of testing, it always comes up whether the expert did or didn't know what they were testing or if they were blind testing. I think very very rarely is the person 'blind testing' in the trials I have watched.

I read that link a couple days ago and it was interesting to me that TrueAllele really doesn't have any validation studies pointing to it's accuracy that have been written by objective users (due to the fact they refuse to release their source code or details about the algorithms they use.)

Furthermore, the court said, TrueAllele had been the subject of validation studies published in peer-reviewed journals, which indicated the contents of the validation studies had been “reviewed by other scholars in the field.”88 The court failed to note, however, that both of the studies it cited for this point had been authored by Perlin and his colleagues, which some critics argue undermines the studies’ validity.8

 

[missy1974]

I watched a youtube video of Dr Perlin trying to explain exactly how his TrueAllele software works. What stood out to me was the very first step in the process. The user of the software first asks questions (such as are there multiple contributors) & begins with a "prior belief." And then the software works on a likelihood of the DNA based on those questions and prior belief (and the data.) To my knowledge, Pujols nor Perlin were asked what questions and prior belief were asked/input at the initial stage in this case. And for those who say TrueAllele *results* are not subjective I say hogwash.

sub·jec·tive
/səbˈjektiv/
adjective

1. 1.
   based on or influenced by personal feelings, tastes, or opinions.
   "his views are highly subjective"
   synonyms: personal, personalized, individual, internal, emotional, instinctive, intuitive, impressionistic; More

Isn't this was Beatriz Pujols testified about? What she put into the computer. Just from looking at my limited notes from her testimony, I have that she set the parameters... she ran it with and without the artificial peaks, she ran it with a different number of assumed contributors, changed the rfu's. She didn't testify for very long, but I think that was the whole point of her testimony. JMO

 

[MrsPC]

Isn't this was Beatriz Pujols testified about? What she put into the computer. Just from looking at my limited notes from her testimony, I have that she set the parameters... she ran it with and without the artificial peaks, she ran it with a different number of assumed contributors, changed the rfu's. She didn't testify for very long, but I think that was the whole point of her testimony. JMO

Her testimony wasn't specific enough though to say what questions she asked and what "prior belief" she input.

 

[citygirl]

Maybe because he is a scientist? They can come across like that because they have so much going on in their heads all the time, IMO.

He actually graduated from med school in 1984. Not sure if he ever practiced. Went on to get his PHD in Computer Science a few years later.

 

[missy1974]

I read that link a couple days ago and it was interesting to me that TrueAllele really doesn't have any validation studies pointing to it's accuracy that have been written by objective users (due to the fact they refuse to release their source code or details about the algorithms they use.)

Furthermore, the court said, TrueAllele had been the subject of validation studies published in peer-reviewed journals, which indicated the contents of the validation studies had been “reviewed by other scholars in the field.”88 The court failed to note, however, that both of the studies it cited for this point had been authored by Perlin and his colleagues, which some critics argue undermines the studies’ validity.8

Regardless of whether you and I think it's reliable, it's being used in the courts, in many States, and countries actually, and not just with TrueAllele's software. I don't know about all the other programs, who uses it most, but I'm going to make an assumption that it's similar to TrueAllele and it's mostly used by prosecutor's. I would agree that someone should probably do some unbiased studies on these programs... it doesn't seem like it would be that hard to do and I wonder why no one has done them? Take degraded DNA that is 'known' and test it in the software... take samples with known contributors and test it... really doesn't seem like it would be that hard lol

What I understood was that he will not release the code but he will allow an attorney (not sure if just anyone can go?) to come and view it at Cybergenetics if they want, and they have to sign a confidentiality agreement.

 

[Mony Mony]

IMO, TrueAllele is not ready for primetime with degraded, low grade, chewed up, cut DNA that has been buried for three years.

With fresh DNA, I am certain his processing is invaluable, but in this case, not so much. I just can't get past the fact there was no McStay match made with the grave DNA.

 

[missy1974]

Her testimony wasn't specific enough though to say what questions she asked and what "prior belief" she input.

What is a 'prior belief' in this circumstance? Isn't that something like 'it was degraded' 'it has X number of contributors', recognizing it could have an artifact so running the sample with and without it.

 

[missy1974]

IMO, TrueAllele is not ready for primetime with degraded, low grade, chewed up, cut DNA that has been buried for three years.

With fresh DNA, I am certain his processing is invaluable, but in this case, not so much. I just can't get past the fact there was no McStay match made with the grave DNA.

From what I understand, most of tests they do and the trials they have been involved in have been from low grade, mixed DNA. Something that can't be interpreted in the lab. That's why people come to them. He made a comment that they can do the "easy cases" as well, but that's not usually why they are used.

 

[Mony Mony]

RSBM

.....I would agree that someone should probably do some unbiased studies on these programs... it doesn't seem like it would be that hard to do and I wonder why no one has done them? Take degraded DNA that is 'known' and test it in the software... take samples with known contributors and test it... really doesn't seem like it would be that hard lol .....

Isn't that what just happened in this case?

 

[MrsPC]

Sorry if I've been beating a dead horse on all the TrueAllele stuff. Honestly, it's my opinion that with all the other circumstantial evidence against Merritt that the TrueAllele testimony won't matter one whit anyway. And I really don't think the Prosecution has a whole heck of a lot from the defense to even rebutt so far. Blake's testimony helped the Prosecution...Jarvis testimony helped the Prosecution...Merritt's sister and daughter's testimonies helped the Prosecution..MM's testimony helped the Prosecution...Bode Lab's testimony helped the prosecution... I think a guilty verdict won't be difficult at all to achieve. IMHO

 

[missy1974]

RSBM



Isn't that what just happened in this case?

no... I'm talking about doing validation studies, independent of Cybergenetics. It just doesn't seem like it would be that hard to do, does it?

 

[Mony Mony]

From what I understand, most of tests they do and the trials they have been involved in have been from low grade, mixed DNA. Something that can't be interpreted in the lab. That's why people come to them. He made a comment that they can do the "easy cases" as well, but that's not usually why they are used.

Suzanna went on and on and on about how it is a great mystery whether anyone could find viable DNA in a sample that has been buried for years. I thought she said it had never been done and boundries of our knowledge of DNA is constantly expanding yadda yadda yadda

 

[Mony Mony]

no... I'm talking about doing validation studies, independent of Cybergenetics. It just doesn't seem like it would be that hard to do, does it?

Why discount this test?

 

[missy1974]

Why discount this test?

I'm not sure what you are talking about LOL

I am talking about ... if there is questions about the validity of these programs as a whole, I am surprised that there are not more studies independent of the said programs... studies use controlled samples...

 

[oceanblueeyes]

He reminded me of Dr. Rudin but he could actually put a sentence together LOL A bit condescending... to both attorney's ... way smarter than all of us, but comes across like an a$$ lol

I still think it's interesting that 90% of his cases are for the prosecution, and it's normally defense attorney's that are crying it's not valid, it's not reliable, not scientific. I found a great research paper online sometime in the last few days, it looked at the pro's and con's of probabilistic genotyping and TrueAllele, it was informative, but even it is looking more at it from a defense attorney/defendant perspective I think.

https://jolt.law.harvard.edu/assets/articlePDFs/v31/31HarvJLTech275.pdf

There is actually some references to some studies that were done dealing with manual interpretation and bias that is interesting, what results are when those that are doing the testing know details and don't know details, I jumped into some rabbit holes when I first read it looking for the other studies because they sounded interesting and it seems that in every trial with any sort of testing, it always comes up whether the expert did or didn't know what they were testing or if they were blind testing. I think very very rarely is the person 'blind testing' in the trials I have watched.

In the other cases where he testified for the prosecution how many times did the DNA come from inside of a many years old graves?

In the other cases were they degraded and had bacteria also?

Since only 10 out of over 350 uses him, I'm not sure the 90 percent is really that many cases or that big of a number.

I can understand the state using him if they thought the DNA found was commingled between more than one individual, and the DNA found was not degraded.

Imo