The Ketchum Report (Continued)

[Guest Tyler H]

I have another, and seemingly final update from my PhD contact who would prefer not to post here:

"What my analysis says is that the bear sequences are real bear and not just primate sequences that are homologous to bear. That means that a bear was involved. This is consistent with your and Bart's reports. It also points to the fact that it's inclusion in the publication was inappropriate, because it adds more confusion than clarity.

I think this is gonna be it for me on this sample. It's fairly tedious work. It’s pointing to an artifactual mosaic due to the combining of human and bear sequences along with poor quality control of the output. Here is something that you can post:

______________________________________

Further analysis of the sequence associated with Sample 26 indicates that it is 2.7 million nucleotides in length., which is only about 0.1% of the human genome. It tracks from beginning to end with sequences associated with Chromosome 11. Chromosome 11 is 134 million nucleotides in length. So, this would correspond to roughly 0.2% of the content of Chromosome 11. There are segments that identify with very high significance to Ursid (bear) sequences. One major limitation related to the analysis is that Genbank is fairly limited in Ursid sequences. Most of the Ursid sequences identify with Panda, but Panda is fairly well represented in Genbank compared to black bear and other bears. The Ursid sequences appear to be mainly from coding regions, rather than structural regions. It is really impossible to compare the Chromosome 11 structural sequences (non-coding) to similar sequences for bears. So, it is possible that there are bear non-coding sequences, as well. This makes it very hard to decipher what might be going on in terms of the source of the sample vs. the contaminant.

So, are the bear sequences real bear or are they primate sequences that identify closely with bear. A distance tree analysis in BLAST using several sequences that identified as phosphatidylserine synthase-2 like coding regions indicates that the Sample 26 version aligns most closely with Ursid sequence (Panda). A primate cluster (human, gorilla, baboon, chimp, rhesus monkey) are highly homologous but more distantly related. Additional sequences from dog, mouse and galigo (Otolemur) were included as outliers and branch further away as might be expected.

So, the upshot here is that the 2.7 million nucleotide data set for Sample 26 is highly flawed, and, therefore, would be nearly impossible to use to determine whether a non-human primate contributed DNA to the sample."

Edited February 22, 2013 by Tyler H

[Guest]

Exactly, if the peer reviewers are anonymous and the public and scientists are blinded to the reviews (question mark), then there would have been no need for her to go to an online purchase...... she could have sold the thing as a monograph on any homegrown site her little heart desired to settle on. Same impact, same effect without the smoke and mirrors, perhaps.

Now, if real scientists, with real academic affiliations or credentials say they read the reviews of the suspect journal and they did pass review...... what does that really mean since nobody knows what this prior journal engaged in from a scientific publication standpoint?

More conundrum, lack of transparency and fishiness if you ask me. Maybe the independent review, if posted up in proper format, can redeem something, maybe not.

If she can't release that information, why isn't she getting sued for sending it to George Knapp?

[bipedalist]

Hard to be sued by a corporate entity that no longer exists that you turned into a personal business? No?

Why would a reviewer really give a flip (if they really existed that is?) if it was a throw away journal (if it really existed in the first place?) fronted by folks with pseudonyms? Just a thought, no evidence?

Edited February 22, 2013 by bipedalist

[Guest]

Exactly. She sends all this documentation to Knapp, then turns around and says she can't release it or she'll be sued.She's about as transparent as a brick wall.

[Guest njjohn]

Public release is different than showing it to people that won't release it publicly. Also, it wouldn't be the journal that sued her, it would be the reviewers. Most major journals don't release reviews for that very reason. So while it might look fishy, it's pretty common in the industry.

[bipedalist]

Well the independent review will tell us whether there were reviewers for her first effort OR NOT.

If it is the OR NOT option, then the circle game continues unless they have effusive praise for her scientific acumen and results of her study.

Any journal worth their salt would support their editors, assoc. editors and reviewers if for no other reason than to simply send a cease and desist letter re: release of names. Many reviewers are friends of friends (i.e., assoc. editors).

Edited February 22, 2013 by bipedalist

[Guest slimwitless]

Please, no Sasquai. Richard Stubstad tried to get this going, and though I liked the guy, I believe I succeeded in talking him out of trying this nonsense.

Did you speak with him on the phone? I always thought he sounded a bit like Jimmy Stewart, actor and Yeti enthusiast.

[Guest]

PB,

I am going to try and answer succinctly point by point

This is a terrible question, but just to gauge the degree of difficulty in determining these questions about species dna recognition:

1. If one submitted a sizable tissue sample, thick enough to take undisturbed tissue from within to test. And this tissue was taken from a horse. Would, or could there be questionable results that would indicate anything other than horse?

see above

2. If the tissue had been in contact with zebra tissue or donkey or burro tissue, could that contamination be identified exclusive of knowing of it beforehand?

I would think that blasting the individual sequences would identify the contaminating species, as the majority of sequence would come back horse with a sprinkling of zebra (depending on level of contamination). No prior knowledge would be required.

3. I understand the horse, zebra and donkey/burro are known species and have documented genomes. would the identifying of the actual tissue and the components of the contaminates be just a matter of plugging in the algorithms or running it through the BLAST protocol or program?

The type of the tissue could not be determined by genomic sequencing but the species certainly. And the contaminants should be likewise identifiable. If you had some totally new contaminant, with no reference in the data base, it should still have some homology to something related. When you have millions of reads (sequences), you would process the data using a similar protocol.

I have never personally processed this sort of sequencing data, so this may not be possible with the shear volume generated. But theoretically, it should be possible.

Thank You RidgeRunner for taking the time and effort to reply to my questions. I need to cram a whole lot more to really keep reading this thread, or have any chance of understanding the basics of the debate here lol Thanks PB

[MIB]

Exactly. She sends all this documentation to Knapp, then turns around and says she can't release it or she'll be sued.She's about as transparent as a brick wall.

She and Knapp both stated that he was under NDA.

MIB

[Guest]

I was listening to the C2C interview in which she laments the lower standard of proof required by the recent announcement of a new monkey species. I got curious what real science is supposed to look like so looked up the paper and was surprised by what I found.

Oddly, it did not cost $30 to view, it is free. It must therefore be worthless.

Also, their photographic evidence is totally lacking. They included easily faked hi-res images of not only the face but also the critter's naughty bits. Who wants to see that? Nowhere to be seen is a video of just the monkey's rib-cage so we can determine once and for all if it is a living, breathing animal. Those wacky scientists!

Link below...

http://www.plosone.o...al.pone.0044271

On a serious note, can someone with knowledge of genetics see if that appears to be all of the DNA data? Is is presented better than Ketchum's for investigation by researchers?

That's a great question, TwilightZone! I wish I had thought of comparing this paper to MK's manuscript as a way of highlighting the qualities of a bona fide scientific paper.

First, The PLoS One paper has BODIES! Seven bodies - with bones, blood, skin, muscle and organs (and DNA). There is no question of provenance with the PLoS One samples. There is no guessing what brushed against a tree or fence, what left blood on a board with nails, slobbered on a car window, or pooped along a trail. They have bodies with detailed descriptions and photos (ewwww, monkey genitals), and they can get more any time they want for you and me to study.

Second, using standard techniques and primers they amplified regions of nuclear DNA and created a contig with homology to known Cercopithecini and partial homology to human Xq13.3. There were no regions that were non-homologous to ANY KNOWN ORGANISM in GeneBank. There was nothing in their DNA studies that would make a geneticist wonder "did they do something wrong, or are there artifacts?"

Third, they have detailed morphological analysis of the bodies/skeletons and behavior that showed that various morphological and behavioral parameters differed significantly from those of known Cercopithecini. In other words, they have both DNA and physical/behavioral data to support each other.

Forth, the manuscript was published in an established (not one issue with one paper), highly-respected journal (that has a pretty good impact factor, a senior editor, an editorial board, and published instructions for authors and reviewers) after peer-review. You don't have to pay to read the paper.

Yup, there's a clear difference in quality. Hope that helps.

Genes

[Guest]

That's a great question, TwilightZone! I wish I had thought of comparing this paper to MK's manuscript as a way of highlighting the qualities of a bona fide scientific paper.

First, The PLoS One paper has BODIES! Seven bodies - with bones, blood, skin, muscle and organs (and DNA). There is no question of provenance with the PLoS One samples. There is no guessing what brushed against a tree or fence, what left blood on a board with nails, slobbered on a car window, or pooped along a trail. They have bodies with detailed descriptions and photos (ewwww, monkey genitals), and they can get more any time they want for you and me to study.

Second, using standard techniques and primers they amplified regions of nuclear DNA and created a contig with homology to known Cercopithecini and partial homology to human Xq13.3. There were no regions that were non-homologous to ANY KNOWN ORGANISM in GeneBank. There was nothing in their DNA studies that would make a geneticist wonder "did they do something wrong, or are there artifacts?"

Third, they have detailed morphological analysis of the bodies/skeletons and behavior that showed that various morphological and behavioral parameters differed significantly from those of known Cercopithecini. In other words, they have both DNA and physical/behavioral data to support each other.

Forth, the manuscript was published in an established (not one issue with one paper), highly-respected journal (that has a pretty good impact factor, a senior editor, an editorial board, and published instructions for authors and reviewers) after peer-review. You don't have to pay to read the paper.

Yup, there's a clear difference in quality. Hope that helps.

Genes

Thanks Genes! I downloaded the article a couple of days ago and immediately thought "they have bodies" too. Then I thought it might be cruel to really compare the two - soooo different. But you did a good job (without being too snarky).

[Guest Silent Sam]

Does she specifically address the storage or the preservation of the samples? Does she discuss the chain of evidence?

In her paper Ketchum states the samples were treated as forensic samples and cataloged to maintain a chain of custody. However when I spoke with Family Tree DNA (one of the "For Hire" labs listed in the paper) they informed me that the samples they received had no chain of custody.

[Guest]

... the samples they received had no chain of custody.

Did they mean no chain of custody "at all"? She did say that she could not vouch for the chain of custody before it came to her, but once she received the samples she established a chain of custody. If the lab did not have a chain of custody at all this contradicts what she stated on C2C.

[Guest njjohn]

She sent the samples to family tree and the other labs blind, so they didn't know what they were testing. Do blind samples usually include chain of custody? Or to keep it blind, the samples alone are sent?

[southernyahoo]

Given that both Bart's lab and my lab came up with very clear evidence that the human DNA present is from the human handlers of the samples (ie, Justin, in our case), I have to disagree with this. And, given that both of our labs identified his samples as bear tissue, I think using bear as the reference to assemble the contigs would have been more logical.

Since this paper is not just about your sample, I disagree and say that all sample ID's start with mtDNA and single contributors. then progresses to nuDNA. Human DNA is primate DNA and is a logical reference. It has been pointed out the other apes would still score high.

Sample 26 may be bear, and if so I wouldn't mind at all removing it from the results. I'm more than ready to evaluate the rest.