Continuation Part Eight: Discussion of the Amanda Knox/Raffaele Sollecito case

[Planigale]

I've asked for months that some of the people here that are experts get together and write a guide for use in court.

I wish that Planigale, Diocletus, Kaosium, and Chris (use little common words ) would write a paper on this case focusing on just three or four pieces of evidence and some basic mistakes in the lab.

I would describe a proper LCN lab, compare it to the Rome lab and explain how easy LCN is to spread by mere touch. IIRC the NY lab used for one of the first LCN used in court and how it was set up, tested and how long that took.

It is important for people to understand that LCN isn't the same as regular DNA in that it can't be seen first. Whereas anyone can understand that if semen, blood or saliva is found in a suspicious location identifying its owner is powerful evidence, finding a microscopic particle of DNA isn't necessarily part of any crime.

So an over view of DNA and LCN. An overview of procedure and protocol in all aspects of collection, transport and testing (storage).

An explanation of how forensic DNA has developed and the value of it in crime work as well as the dangers.

Then go into some detail (footnotes for scientists) of what was wrong with the specific pieces of evidence in this case and showing pieces that were done correctly.

If I forgot an expert it was not intentional.

Thanks for pointing me to this. I'll take a longer look. I once did a PCR training course but this is not my area of expertise. The thing I have been suspicious about is the negative controls for when the RT qtPCR was out of action and the knife was run having got an undetectable from the Qubit.

When the RT qtPCR was in action I have no problems with not typing the negative controls when a count of 50 was returned indicating no detectable DNA so long as no samples with a count of 50 were typed.

The problem arises when the knife was run. Assuming the negative controls returned 'undetectable' on the Qubit one had to decide whether no samples with an 'undetectable' would be further processed. If you decided to process samples with undetectable then you needed to process the negative controls in the same way. My guess is this never happened. So although there were negative controls, and they have returned an 'undetectable', this was not an adequate control process because the sensitivity of the Qubit was not sufficient to detect low level contamination, exactly what might explain the DNA of MK on the blade. The suppression of the typing results is exactly what one would have expected if this happened. I do not blame Stefanoni, the obvious thing was not to type the negative controls with undetectable DNA. The unclear thing is why she processed a sample with undetectable DNA, once she did that she was obligated to do the same for the negative controls, a slip but not I think deliberate. She probably never realised the problem until the defence insisted on the controls. So she was able to stand up and say the negative controls were negative (true), but using an insensitive assay and not treating them the same as the samples invalidates the controls.

I may be wrong, she may have put the negative controls through for typing, in which case why not provide the results, unless of course they showed DNA, even then one could make an argument that as long as they were not showing alleles that might match MK it is irrelevant to the result on the blade. I am very resistant to the idea that Stefanoni deliberately concealed exculpatory results; others may have a different view and think this is a more likely explanation.

When one is working in a lab with serial samples, and there is a sample you decide not to process, let us say in this instance a negative control you decide not to put through for typing, you may leave an empty well. The reason for doing this is variable, but if you are putting multiple samples in to an e.g. 8x8 block you know that sample 8,16,24 are at the end of a row and it is a check that you have not put two samples into the same well. So there may be an innocent explanation for there being some gaps e.g. leaving the wells empty for samples with too low a result that you decide not to process. But then one would just say no sample processed or something similar.

 

[Chris_Halkides]

paper copies of the negative controls have limited value

I agree, he seems to be regarding them as defence experts rather than independent court appointed experts. The view of the Scientific Police on their own practice overrides the defence experts, the independent experts and international standards.

I find the discussion about controls is odd. In science we now tend to do large data dumps to accessible on line stores so people can examine the 'raw' data rather than your own interpretation in papers. If someone contacted me asking for the data my first response might be why didn't you download it. But there are sometimes issues, I have downloaded a completely different dataset from the one I wanted, because the link was wrong. So we would just email the dataset. As a matter of courtesy if someone is reviewing your work rather than saying look at the paper copies filed in the court proceedings (or even on a disc) it is reasonable to send directly an electronic file. I find this worryingly obstructive, why did not the police just say we filed the controls but here is a copy. Why be obstructive to the agents of the court?

Finally I note there is no statement from Nencini saying he has seen the control data with his own eyes only that Stefanoni says she provided the data. Nencini may not realise that just saying the controls were negative is sufficient. We need to know the details, were they run to same sensitivity as the sample? The same number of replication cycles? Did this include environmental controls?

Planigale,

I agree with almost everything you have written, but I would like to make two additional points. DNA forensic scientists prefer to receive the data in its rawest form, which is the fluorescence versus time information prior to the color separation that gives us the four familiar traces in an egram. They can look for pull-up that way, and they can also zoom in on small peaks once the data have been processed. Even if Stefanoni provided paper copies, negative controls in this form would have very limited utility. However, if Stefanoni provided paper copies, then where are they?!

As Mehul Anjaria pointed out, one would have to run the negative control through the same procedures (including the concentration step in the speed vac) as were used to generate the knife profile. I doubt she did any controls, but I very much doubt that she thought about doing controls that included this critical step. I happen to think that this step is as likely as any step for where contamination occurred.

 

[Diocletus]

The thing I have been suspicious about is the negative controls for when the RT qtPCR was out of action and the knife was run having got an undetectable from the Qubit.

I'm not so sure that the RT was out of action. It was definitely used for some purpose during the time period in question, as we have a gap in RT run numbers from Nos. 545 through 548. At most, she decided for some undisclosed reason not to use RT-qPCR for this particular case. I think that I see the reason for this as to the samples that were processed late on 6 November, and the reason is that the RT was probably in use and she needed urgently to process samples from this case, including reference samples for the newly-arrested folks (BTW, it seems to me that this testing was illegal). It is not clear why she continued to use the Qubit on November 13-14.

When the RT qtPCR was in action I have no problems with not typing the negative controls when a count of 50 was returned indicating no detectable DNA so long as no samples with a count of 50 were typed.

Agreed, but that's not what she did. It appears that she passed the negative controls (even 50/50's) from RTq-PCR on to amplification, as the RTq-PCR negative controls were assigned an amplification-specific ID number. Presumably, she is using them as some sort of batch control.

No results of controls from the amplification process were disclosed. We can only see that they match up (positionally) to the NTCs from RT-qPCR.

The problem arises when the knife was run. Assuming the negative controls returned 'undetectable' on the Qubit one had to decide whether no samples with an 'undetectable' would be further processed. If you decided to process samples with undetectable then you needed to process the negative controls in the same way. My guess is this never happened. So although there were negative controls, and they have returned an 'undetectable', this was not an adequate control process because the sensitivity of the Qubit was not sufficient to detect low level contamination, exactly what might explain the DNA of MK on the blade. The suppression of the typing results is exactly what one would have expected if this happened.

Spot on. It is not clear that she ran any controls whatsoever for the Qubit.

I do not blame Stefanoni, the obvious thing was not to type the negative controls with undetectable DNA. The unclear thing is why she processed a sample with undetectable DNA, once she did that she was obligated to do the same for the negative controls, a slip but not I think deliberate. She probably never realised the problem until the defence insisted on the controls. So she was able to stand up and say the negative controls were negative (true), but using an insensitive assay and not treating them the same as the samples invalidates the controls.

If this were the case, then what she should have done is produce her results and then testify honestly about the issues that you just mentioned. You are describing a situation in which there is no check against contamination.

When one is working in a lab with serial samples, and there is a sample you decide not to process, let us say in this instance a negative control you decide not to put through for typing, you may leave an empty well. The reason for doing this is variable, but if you are putting multiple samples in to an e.g. 8x8 block you know that sample 8,16,24 are at the end of a row and it is a check that you have not put two samples into the same well. So there may be an innocent explanation for there being some gaps e.g. leaving the wells empty for samples with too low a result that you decide not to process. But then one would just say no sample processed or something similar.

That might be true in a few cases, but what is clear is that she amplified hundreds of samples, but disclosed few profiles and no controls (from quantification or amplification).

 

[anglolawyer]

Without any knowledge of DNA at all the knife stinks. It's selection is suspect and the testing equally so. What are the odds? Stefanoni's known results, which she suppressed and lied about (is that a clue, do you think?) justified her tossing the knife away as meaningless. We are asked to believe her intuition, on top of the cop's, led her to this non-repeatable, decisive result. And that's without the bra clasp panto. It's no surprise when the big picture reeks that the details should fail to reassure.

 

[Grinder]

No, Because Rudy has no real reason to lie about the time Grinder. It's not like some of the other things he says which seem couched in ways trying to absolve himself from the crime. Rudy admits being there...so why does the time matter to him? Seriously? Give me one logical reason that Rudy lies about the time? The gastric evidence, the clothes in the washer, the outer clothes, Rudy's statements they all confirm around a 9:30 TOD.

The clothes mean nothing. Her dress means nothing. As I said and you ignored the 9:20-30 scream fit with the Patrick meetup as described by Amanda.

Who says Rudy waits 45 minutes? I can see a real possibility that Rudy ran down into the ravine and spent that time trying to calm down and think clearly before heading back to his apartment. And out of curiosity, I thought we resolved the spelling of Rudy's name?

Yes he ran into the ravine. Sure. Did he steal things first or return. When did he drink? Never mind.

Rudi is how the vast majority of the documents I read from Italian spell it. But I think either way it is known to whom I refer.

 

[codyjuneau]

Andrea Vogt on Nencini on C-V

<snip>

He categorically rules out contamination Sollecito's involvment in the murder. Even in the case of the bra clasp catalogued by police but only collected 46 days later, had Sollecito’s DNA been fruit of contamination been due to involvement in the murder it would have been found elsewhere and on other objects taken from the scene that day, such as on Meredith’s blue sweatshirt where Guede’s DNA was found, he wrote, not just on the tiny hook of the bra clasp.
<snip>

.
I fixed Nencini's logic.
.

 

[Grinder]

DNA people thanks but I hope these recent posts weren't meant for us rabble.

One of the things I think was discussed is that if a device produces a too low reading and the technician decides to do more (more amplification ?) then it needs to be done on all the controls (negative?) in order to assure that the DNA isn't a remnant or residual. This would also mean that all tested items should be tested in the same manner, whether control or not.

If any of you do decide to write an article and I think a real publication might be interested, I would volunteer to be a rabble that reads for the dummy factor. I don't think you all get how stilted the science language is.

Thank you all.

 

[acbytesla]

The clothes mean nothing. Her dress means nothing. As I said and you ignored the 9:20-30 scream fit with the Patrick meetup as described by Amanda.

Yes he ran into the ravine. Sure. Did he steal things first or return. When did he drink? Never mind.

Rudi is how the vast majority of the documents I read from Italian spell it. But I think either way it is known to whom I refer.

When does Amanda describe a 9:20 to 9:30 scream fit? Amanda says she wasn't there. I'll grant you that the clothes and calling her mother is minimal anecdotal evidence. Never the less they are INDICATORS. The real evidence is the gastric evidence and Rudy's own words. You still haven't addressed why Rudy would lie about the time of death. We'll never prove whether it was 9:15 or 9:45...and I really DON'T think it makes that much difference, although obviously 9:15 to 9:30 would have made it a physical impossibility and the 9:45 to 10:00 PM TOD seems incredibly improbable unless it was a premeditated conspiracy.

 

[Grinder]

Without any knowledge of DNA at all the knife stinks. It's selection is suspect and the testing equally so. What are the odds? Stefanoni's known results, which she suppressed and lied about (is that a clue, do you think?) justified her tossing the knife away as meaningless. We are asked to believe her intuition, on top of the cop's, led her to this non-repeatable, decisive result. And that's without the bra clasp panto. It's no surprise when the big picture reeks that the details should fail to reassure.

I completely agree. As we have discussed it isn't just the selection of this knife but the lack of selection of the other knives in the drawer and at the time multiple knives at Patrick's bar and home. In addition one would think that the knives in the cottage would be confiscated and at least checked for blood if not prints and DNA.

My god man 4-1 say it isn't so.

 

[Diocletus]

DNA people thanks but I hope these recent posts weren't meant for us rabble.

One of the things I think was discussed is that if a device produces a too low reading and the technician decides to do more (more amplification ?) then it needs to be done on all the controls (negative?) in order to assure that the DNA isn't a remnant or residual. This would also mean that all tested items should be tested in the same manner, whether control or not.

If any of you do decide to write an article and I think a real publication might be interested, I would volunteer to be a rabble that reads for the dummy factor. I don't think you all get how stilted the science language is.

Thank you all.

If you spent as much time reading up on dna as gastric emptying, you'd be able to follow right along.

In all seriousness, though, C&V summarize most of this stuff.

 

[codyjuneau]

Wasn't Lana's house broken into a few times, recently before the Nov 1, 2007? Seem's I recall reading that.
I agree, likely Rudy knew the roads well at night.
Rudy was in that "profession" of burglarizing and who knows what else... it's logical he knew all the dark roads and paths out of sight, most thugs and burglars would.

.
I have not heard that before. If there is a link to it I would sure be interested.
.

 

[codyjuneau]

DNA people thanks but I hope these recent posts weren't meant for us rabble.

One of the things I think was discussed is that if a device produces a too low reading and the technician decides to do more (more amplification ?) then it needs to be done on all the controls (negative?) in order to assure that the DNA isn't a remnant or residual. This would also mean that all tested items should be tested in the same manner, whether control or not.

If any of you do decide to write an article and I think a real publication might be interested, I would volunteer to be a rabble that reads for the dummy factor. I don't think you all get how stilted the science language is.

Thank you all.

.
Maybe it would work better if you write it, they correct it, you rewrite it, rinse and repeat.
.

 

[Grinder]

When does Amanda describe a 9:20 to 9:30 scream fit? Amanda says she wasn't there. I'll grant you that the clothes and calling her mother is minimal anecdotal evidence. Never the less they are INDICATORS. The real evidence is the gastric evidence and Rudy's own words. You still haven't addressed why Rudy would lie about the time of death. We'll never prove whether it was 9:15 or 9:45...and I really DON'T think it makes that much difference, although obviously 9:15 to 9:30 would have made it a physical impossibility and the 9:45 to 10:00 PM TOD seems incredibly improbable unless it was a premeditated conspiracy.

I would bet that if Meredith wasn't attacked immediately she would have tossed the book on the bed and laid down to digest and maybe catch 40 winks. It was not freezing but cool and her dress was not heavy and was loose fitting. Rudi's words are not evidence.

I did discuss why he might give the wrong time for the scream. One reason is that he didn't know the time which is almost 100% certain for obvious reasons. The second reason just given to you is that he wanted the story to fit with Amanda's story. He had read that she arranged to meet Patrick in her text sent before 9 and they met at the plaza and she took him to the cottage and he raped and killed Meredith and she screamed. He figured that was about 9:20-30.

When did he say he arrived at the cottage?

 

[Grinder]

.
Maybe it would work better if you write it, they correct it, you rewrite it, rinse and repeat.
.

On the off chance that wasn't mocking, I'm not well enough versed on DNA to write the first version.

I would like a basic description on collecting and transport protocol with examples of how easy secondary and tertiary transfer is. Then a brief description of the machines and how the evidence is handled. What controls should be run and how the records are kept both machine and notes by operators.

The actual reading of the results would come next. This would include some idea of how easy or hard this aspect is including how reliable any reading is. I've been confused as to how important each allele is. How many sure matches are needed.

Remember I said I'd do a review to get the DNA dummy view addressed.

 

[Grinder]

If you spent as much time reading up on dna as gastric emptying, you'd be able to follow right along.

In all seriousness, though, C&V summarize most of this stuff.

DNA gives me a stomach ache.

If you'd give the link I should read, I will.

 

[carbonjam72]

Stefanoni; Methods - deliberate or accidental

Thanks for pointing me to this. I'll take a longer look. I once did a PCR training course but this is not my area of expertise. The thing I have been suspicious about is the negative controls for when the RT qtPCR was out of action and the knife was run having got an undetectable from the Qubit.

When the RT qtPCR was in action I have no problems with not typing the negative controls when a count of 50 was returned indicating no detectable DNA so long as no samples with a count of 50 were typed.

The problem arises when the knife was run. Assuming the negative controls returned 'undetectable' on the Qubit one had to decide whether no samples with an 'undetectable' would be further processed. If you decided to process samples with undetectable then you needed to process the negative controls in the same way. My guess is this never happened. So although there were negative controls, and they have returned an 'undetectable', this was not an adequate control process because the sensitivity of the Qubit was not sufficient to detect low level contamination, exactly what might explain the DNA of MK on the blade. The suppression of the typing results is exactly what one would have expected if this happened. I do not blame Stefanoni, the obvious thing was not to type the negative controls with undetectable DNA. The unclear thing is why she processed a sample with undetectable DNA, once she did that she was obligated to do the same for the negative controls, a slip but not I think deliberate. She probably never realised the problem until the defence insisted on the controls. So she was able to stand up and say the negative controls were negative (true), but using an insensitive assay and not treating them the same as the samples invalidates the controls.

I may be wrong, she may have put the negative controls through for typing, in which case why not provide the results, unless of course they showed DNA, even then one could make an argument that as long as they were not showing alleles that might match MK it is irrelevant to the result on the blade. I am very resistant to the idea that Stefanoni deliberately concealed exculpatory results; others may have a different view and think this is a more likely explanation. When one is working in a lab with serial samples, and there is a sample you decide not to process, let us say in this instance a negative control you decide not to put through for typing, you may leave an empty well. The reason for doing this is variable, but if you are putting multiple samples in to an e.g. 8x8 block you know that sample 8,16,24 are at the end of a row and it is a check that you have not put two samples into the same well. So there may be an innocent explanation for there being some gaps e.g. leaving the wells empty for samples with too low a result that you decide not to process. But then one would just say no sample processed or something similar.

(First off, Dr Waterbury has a detailed review of the forensics in his book, "The Framing of Amanda Knox". Trying to read through it now, maybe I'll find some help. But much appreciation to the DNA experts here for helpiong to sort through.)

It's a fair point to be cautious about accusing people of dishonesty. In fairness though, she did commit perjury on the witness stand; when she said she hadn't run blood tests on footprints revealed by luminol, when in fact she had and the tests were negative.

Also, she is on the crime scene video from the first day asking urgently about getting a 'presumable semen' stain tested (MMK.com), and that tested stain never surfaced.

The result of Meredith's DNA on the blade may have been a lucky accident. But look at the pressure she was under to find a match, to provide some evidence in support of the prosecution's claims. She just happens to match Raf to the Bra strap, and Meredith to a knife that Amanda had used (did they know she used that knife?).

Getting a 'hole in one' is one thing. Doing it blindfolded is something else entirely.

When you go to a kids magic show, or even David Copperfield, you can be amazed at the tricks, but you're aware that they're tricks. Magic isn't real. Stefanoni's results seem too good to be true, at least for the prosecution. They haven't done anything honest from the beginning. Doesn't prove guilty of tampering, but these people are playing it straight.

I don't think outright planted (because the presentation of the findings would be stronger if they were actively planted) but something in the neighborhood.

 

[codyjuneau]

On the off chance that wasn't mocking, I'm not well enough versed on DNA to write the first version.

I would like a basic description on collecting and transport protocol with examples of how easy secondary and tertiary transfer is. Then a brief description of the machines and how the evidence is handled. What controls should be run and how the records are kept both machine and notes by operators.

The actual reading of the results would come next. This would include some idea of how easy or hard this aspect is including how reliable any reading is. I've been confused as to how important each allele is. How many sure matches are needed.

Remember I said I'd do a review to get the DNA dummy view addressed.

.
I was not mocking at all Grinder.

I think that if a non-expert writes it, with factual errors corrected by the experts, there is a very good chance it will be more understandable to other non-experts. I will volunteer to be the judge of that.
.

 

[Bill Williams]

Andrew Gumbel's English Translation
"The police officer who actually removed the knife from the kitchen drawer stated during the first trial that he had been struck by the fact that knife seemed much cleaner than the other items in the drawer, leading him to believe that it had been recently cleaned. This observation, which might seem irrelevant or a matter of perception, raised important questions. Forensic examination of the knife, its blade, and a series of streaks, barely perceptible to the naked eye, resulted in identification of the DNA of both Meredith Kercher and Raffaele Sollecito."​

First highlight - this is a complete misrepresentation of what even Judge Massei at the 2009 trial at least deals with. It is Stefanoni who says there was a striation on the knife in which 36b was lodged, presumably to protect it from cleaning. (The issues with that claim are legion, but will not stop me here.

Suffice it to say that even Massei acknowledges that even the preliminary judge's expert did not see those striations or grooves. Peter Quennell on TJMK claims to have a picture of the knife showing those grooves... begging the question of even Nencini's claim of them being "barely perceptible", or why the first prosecution expert did not see them at all.

Below is what one biochemist says about the cleaning of that knife with bleach... IF you believe the cops who said that upon entering Raffaele's place the apartment "smelled of bleach." (Why it needed to "smell of bleach" for the cleaning of a knife, where a cap-full in a sink-full of water would have sufficed....)

Striation or no striation, groove or no groove... (one poster to this thread even said that the blood over-top of the lone speck of alleged DNA in a groove no one by Stefanoni can see protected the DNA from cleaning, but that will not delay us here.... here's the biochemist....)

If the knife had been cleaned with bleach any DNA on its surface would have dissolved in the liquid and been washed away. If a minute quantity had somehow attached itself to the knife (by some novel biochemical mechanism unknown to the science of Molecular Biology) it would have been chemically modified by the bleach so that it would not be detectable by PCR. Thus, this scenario is biochemically impossible.​

 

[Grinder]

.
I was not mocking at all Grinder.

I think that if a non-expert writes it, with factual errors corrected by the experts, there is a very good chance it will be more understandable to other non-experts. I will volunteer to be the judge of that.
.

 

[Grinder]

It is a pity that Amanda and Raf weren't quizzed at their first hearing after being arrested as to their knowledge of destroying DNA with bleach. Had they acknowledged this knowledge which can be 100% implied by the MySpace picture of Raf with the cleaver and bottle, then it is 100% certain they would have cleaned the knife in bleach. My micro biologist buddy says that in household bleach destruction would take a little time.

The defense should contend that the smell of bleach indicates the kids cleaned the knife with the substance and therefore the DNA found has to be contamination.

This I believe would resonate with the Italian mind.