A problem with the "Eddowes Shawl" DNA match

[Pcdunn]

I find it totally wrong to have sold all those books on the back of dodgy d.n.a evidence at its best it's just incompetent to the level of laurel and hardy and the worst fraudulent I would like to see the police look at this and I don't mean pc Amos.

Well, I agree with you. I am on another forum where we discuss a variety of things, and when the link to the story about the shawl and the "DNA" was posted, one person wrote that it was just amazing what science could do. A few others stated that it wouldn't be definite until it was published in a peer-reviewed journal. They were right, of course, but the first person's comment is understandable, thanks to popular television shows about crime-solving. We expect science to do more than it is capable of-- and we expect people to use it correctly. (Which is our first mistake, I suppose...)

[pinkmoon]

I have finally read through all of this thread, and would like to thank everyone who patiently contributed explanations that helped this former English Lit major to understand the science of genetics and the controversy. I believe that the author and publisher rushed to press, and the speed contributed to a mistake in the scientific analysis. Does anyone know if there will be an attempt at a peer review?
Pat D.

I find it totally wrong to have sold all those books on the back of dodgy d.n.a evidence at its best it's just incompetent to the level of laurel and hardy and the worst fraudulent I would like to see the police look at this and I don't mean pc Amos.

[Robert]

Robert Anderson, who was at the recent conference, reported Jari as saying that his new technique required peer review. I don't know if anything else is in the pipeline.

[Pcdunn]

I have finally read through all of this thread, and would like to thank everyone who patiently contributed explanations that helped this former English Lit major to understand the science of genetics and the controversy. I believe that the author and publisher rushed to press, and the speed contributed to a mistake in the scientific analysis. Does anyone know if there will be an attempt at a peer review?
Pat D.

[Pcdunn]

The final words in the book 'Aaron Kosminski is Jack the Ripper' remain just HIS opinion. There is NO proof. He has not provided ANY proof. He's simply told us he has proof and he expects us to believe him.

He had a preconceived suspect and a preconceived agenda which instantly negates ANY of his 'proven' claims. A vested interest and proper scientific analysis are not good bed-fellows if you want to get at the truth.

They are, however, exactly what we can expect from a book using pseudoscientific analysis (or, rather, using an established science to produce "support" for a preconceived conclusion)... And that is NOT good science.

[Batman]

As a biologist I have would like to point out that scientists punlish their findings in peer-review magazines, not in pop.culture books. If its in pop culture without peer-review then chances are you are reading pseudscience.

[Chris]

There has been an interesting discussion on jtrforums about the "Eddowes" DNA match. Further revelations are promised about Dr Louhelainen's response to the one person at the Salisbury meeting who asked him about the 314.1C error in the DNA analysis:

http://www.jtrforums.com/showthread.php?t=22016

But I do feel that there's a danger of getting away from the central point, what with all the emphasis on who said what to whom, what they might have meant by it, and what responsibility people might or might not have if other people quoted their mistaken work in a book without their knowledge - most of it highly hypothetical.

I think the thing we mustn't lose sight of as Ripperologists is the information we need to know before we can begin to discuss whether there's anything significant about the match with the DNA of Catherine Eddowes's descendant:

(1) If the 314.1C error is corrected, can the other people who handled the shawl still be eliminated as a possible source of the DNA that was matched?

According to the book, this was done on the basis that the control samples from the other people who handled the shawl didn't have 314.1C: "One of these amplified mtDNA segments had a sequence variation which gave a match between one of the shawl samples and Karen Miller's DNA only; i.e. the DNA sequence retrieved from the shawl did not match with control reference sequences. This DNA alteration is known as global private mutation (314.1C) ..." [quoted from Dr Louhelainen's summary of the analysis]

But now we know 314.1C is not rare, but a misnomer for 315.1C, which is found in more than 99% of the population, it's almost certain that at least one of the control sequences would have had it. So can they still be eliminated?

(2) Without 314.1C/315.1C (mistakenly thought to be rare) what percentage of the population would agree with the "small segment" that the "Eddowes" match was based on?

It was one of seven segments from the hypervariable regions, and therefore probably contained only 100-200 base positions. The available data suggest that such a short sequence might well have been shared with quite a large percentage of the population.

[Chris]

People who have been following the discussion of the Ripper DNA analysis may be interested to see ...

And also to know that the sequence variations found in the mtDNA from the bones included 315.1C ...

[Chris]

People who have been following the discussion of the Ripper DNA analysis may be interested to see the scientific paper which has just been published on the DNA analysis relating to the bones identified as those of Richard III:

http://www.nature.com/ncomms/2014/141202/ncomms6631/full/ncomms6631.html

When the results were first announced, there was some criticism on the same lines as the criticism of the Edwards-Louhelainen work on the Ripper case, namely that the announcement was made before the work had been peer-reviewed, that the match might be the result of contamination and that it might not be conclusive if it was limited to short segments of DNA:

http://www.livescience.com/26839-richard-iii-announcement-reactions.html

But as published, the match is based on the whole mitochondrial DNA sequence (sixteen and a half thousand base positions long). A perfect match was found with one living relation in the female line, and a single difference in the sequence for another relation.

The researchers searched the EMPOP database containing 26,127 European sequences (the same one used by Dr Louhelainen) and another database of 1,832 British sequences, and didn't find a match to the sequence from the bones. On that basis, given the uniform frequencies of mitochondrial haplotypes across Europe, and the historical mobility of noble women, they suggested that an estimated frequency of [less than] 1 in 10,000 would be justified, but they chose to err on the side of caution by basing the estimate instead on the smaller British database, giving 1 in 1,832.

[Chris]

Many thanks for all your interesting and informative posts on this topic. I have been away from the boards lately (and meetings and conferences too), so am struggling to catch up. What struck me about the above quote was the alleged 99.2 per cent match. I recalled that exact percentage appearing in your initial post to this thread (quoted again below), but in another context entirely (in fact it could hardly have been more different: one implying a suspect's involvement; the other undermining the case regarding the victim), so I have to presume it's just one of those odd little coincidences, although it still makes me very slightly uncomfortable. I have underlined the percentage in each quote for the sake of clarity.

Thank you. I hadn't seen this when you first posted it, nor had I noticed the coincidence. But as you say, the contexts in which the figure appears are completely different, so I think it can only be a coincidence.

[caz]

No, they were both mitochondrial DNA.

... he [Louhelainen] started work on comparing M’s mitochondrial DNA with that of the cells extracted from the semen stain on the shawl.
...
What Jari had found was a 99.2 per cent match when he ran the alignment in one direction, and going the other way it was a 100 per cent perfect match.

Hi Chris,

Many thanks for all your interesting and informative posts on this topic. I have been away from the boards lately (and meetings and conferences too), so am struggling to catch up. What struck me about the above quote was the alleged 99.2 per cent match. I recalled that exact percentage appearing in your initial post to this thread (quoted again below), but in another context entirely (in fact it could hardly have been more different: one implying a suspect's involvement; the other undermining the case regarding the victim), so I have to presume it's just one of those odd little coincidences, although it still makes me very slightly uncomfortable. I have underlined the percentage in each quote for the sake of clarity.

Anyway, it all makes a nice change from the blessed diary shenanigans!

In "Naming Jack the Ripper", Russell Edwards quotes Dr Jari Louhelainen's comments about the match between a segment of mitochondrial DNA obtained from the area of a possible blood stain on the "shawl" and the corresponding segment from a female-line descendant of Catherine Eddowes:
"One of these amplified mtDNA segments had a sequence variation which have a match between one of the shawl samples and Karen Miller’s DNA only; i.e. the DNA sequence retrieved from the shawl did not match with control reference sequences. This DNA alteration is known as global private mutation (314.1C) and it is not very common in worldwide population, as it has frequency estimate of 0.000003506, i.e. approximately 1/ 290,000. This figure has been calculated using the database at Institute of Legal Medicine, GMI, based on the latest available information."

Just over a week ago, Tracy I'anson posted an excerpt from a paper describing software designed to identify missing persons, which discusses the conventions for describing variations in the mitochondrial DNA sequence:
"An insert, such as the common extra “C” after base position 315 is listed as “315.1 C”. For matching purposes, the program tolerates errors in nomenclature for equivalent variants such as the extra C in the poly-cytosine region being reported as “314.1 C”."

http://forum.casebook.org/showpost.php?p=309837&postcount=2968

The paper she quoted can be found here:

http://www.genecodesforensics.com/news/Promega%202004%20M-FISys%20for%20Missing%20Persons%20paper%20revised.pdf

The authors refer to 314.1C as an "error in nomenclature". This is because 314.1C indicates that the measured sequence differs from the standard reference sequence for mitochondrial DNA by the insertion of one additional C (cytosine) after position 314 in the sequence. But the reference sequence actually has a string of five Cs in a row around here, in positions 311 to 315. The additional C could be inserted at any point in this string, and the resulting sequence - which is all that can be measured - would be exactly the same. The convention in forensic genetics is to describe the insertion as having occurred after the last possible position - that is, in this case, after position 315. So the conventional description for this mutation is 315.1C, not 314.1C.

The problem is that 315.1C is not a rare mutation, as the authors quoted by Tracy indicate. In fact the presence of an extra C in this position is much more common than its absence, because this is a case in which the reference sequence itself contains an uncommon mutation. The database referred to in the book can be found at http://empop.org/ and it indicates that 315.1C is present in 99.2% of the sequences which have information for this position.

It appears that something has gone badly wrong with the analysis here, and obviously the quoted figure of 1 in 290,000 can't be accepted without further explanation.

Love,

Observant Caz
X

[Archaic]

Article: 'Jack the Ripper, One of Us? No, Thanks.'

Hi guys.

I saw an article this morning in an online journal called The Jewish Daily Forward which discusses the controversy over 'Naming Jack the Ripper', the shawl DNA, and the claimed scientific identification of Aaron Kosminski as Jack the Ripper.

The article is well written and more thorough than most. Its author is Lenny Picker, described as an attorney who has lectured on the Ripper case.

According to the article, an Israeli paper reported that Kosminski family descendents feared "reprisals", which I don't believe I've heard before.

'Jack the Ripper, One of Us? No, Thanks. 125 Years of Accusations and Anti-Semitism': http://m.forward.com/articles/209121...-us-no-thanks/

Best regards,
Archaic

[Chris]

The confusion here seems to be almost total. If anyone else who attended the public sessions at the conference has taken proper notes they may tell us something, but otherwise I can't see the point of trying to decipher these Chinese whispers.

[mickreed]

why is it being discussed in the case of Kates MTDNA?

I'm a bit mystified as well.

[Panderoona]

Hi Panderoona,

Well, really your guess is a good as mine. From everything I know about mtDNA, it is unlikely that they can prove a direct descent between the DNA on the shawl and the DNA of the Eddowes descendant. The only thing that claimed to make it highly likely was the extreme rarity of 314.1C. That fell over when it was found to be an error of nomenclature, and not rare at all. As I said somewhere else, mtDNA can exclude somebody where there's no match, but cannot prove. In court it's may used as a building block with loads of other evidence.

For example, you might have a witness that places someone who resembles Freda Bloggs at the scene of a crime, Freda may have no alibi for the occasion, Proceeds of the crime might be found in her bottom drawer, and mtDNA that matches hers may have been found at the scene. All in all strongly suggestive.

We don't have anything like that in this case.

Your comments on Y-DNA seem apt, but irrelevant in the case of the Eddowes match, since we are talking about women. It may be relevant if the so-called 'Kosminski' DNA is ever made public.

many thanks for your reply. The main reason I mentioned the YDNA is that it appears to be YDNA that requires these 12/13 markers in say, a paternity test, where one person could be excluded rather than a certain identification. If this 12/13 marker thing is only relevant to YDna, why is it being discussed in the case of Kates MTDNA?