Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
But what makes you think it doesn't already?
If you have four BGA groups and you find a sufficient number of allele frequencies for each group that are mutually exclusive to the other groups then the test should work as advertised. The MLEs may vary some but the admixture will not. If the SNPs have sufficient depth the MLEs won't vary much at all within one BGA group.
Simply adding more parentals does not ensure greater accuracy because any markers that you find will still have to be inclusive to the whole BGA group. In other words, the method and accuracy of measurement doesn't change.
The real issue here is not the formulation of the test - it's the process of validation. After reading the dnaprint log for over a year now, I'm pretty sure that the accuracy levels of the test are very close to what Dnaprint says they are.
Yes, this is very close to what I've been trying to say. Maybe I'm a poor communicator <g>.
The test begins with a *very* broad and simple hypothesis and it is indeed a very unclean (heuristic) process. Simply adding more samples to the parentals at the early stage of the process is not as easy (or possible) as you might think. But (apoligies to Ming) this is JMO.
Gotta get back to work now. cheers.
You know, Mike, I agree with this post - but it wasn't what you were saying yesterday....
DNAP never made claims that it could distinguish between 1,2 and 3. The postulation that "3" is rare is just that - a postulation. For the purpose of genealogy, it would be really excellent if ADB could distinguish. Frudakis said they are working on it - I assume it has something to do with measuring the number or size of the haplotype blocks.
But mixing and matching AIMs formed in ancient and more recent periods of human evolution is not possible with this test - if you define the test in greater resolution before you get enough feedback from a very broad hypothesis then that's what you're going to wind up doing.
It doesn't make sense to me, Ming.
My whole point about the test being "heuristic" (which was scoffed at and dismissed) is that if you go around the globe and take many samples that are to fit into broader, predetermined populations then you are building assumptions into the test. The ultimate establishment of boundaries was an interplay between the samples and testing. Not just samples and not just testing.
If you haven't seen anything that might revise your opinion then you must not be reading my posts. Maybe if you wanted to get somewhere in the discussion you could respond to some of the things I've actually said instead of restating your opinion.
Besides, DNAPs business model doesn't require that genealogists and anthropologists agree with them. The company itself stated that they are in no hurry to publish their findings until it coincides with their business plan.
But is it real admixture? The test can NOT tell.
Not true. It cannot tell you when the admixture took place. That's a big difference. The test is "calibrated" to detect differences in AIMs at a certain point back in time. Therefore, in the case of NAM, it is not going to differentiate between different waves of migrations that took place over thousands of years and package them into distinct populations. You should expect at least some EA admixture.
The test in the 2.0/2.5 embodiment cannot capture ancient AIMs and compare them to more recent AIMs. The entire test is designed to evolve in layers like strata of earth.
I'm losing interest in this discussion because you keep ignoring my prior posts. You have a lot of holes in your argument, but when someone points them out to you, you pretend not to hear them....carry on with your personal crusade.
I believe that a large part of their alleged “admixture” is simply genetic distance of unmixed populations from the parentals...
1. So....how would genetic distance from unmixed populations show admixture? That doesn't compute. I mean, that would be an *incredible* coincidence.
or, more likely, both the parentals and the other populations have low level admixture, but that this is being over-inflated in the other populations, because of their relative genetic distance from the too-narrow parental group.
2. That can only be answered with a statistical explanation. I don't have any background in biostatistics so I can't give anything but a crude answer. But basically, the test already takes this issue into account. You are right on this issue Admixture is bound to be present in the parental samples and the test had to be designed to address this. That issue was raised on Rootsweb over a year ago but Frudakis had explained how that admixture is statistically pigeonholed. Obviously, it had to do with how the AIMs were selected but it would be better to find the exact post than have me recite it from memory (its long). Think about it this way, if the above concern was at all valid, the test would be horribly inaccurate and specific levels of admixture would be spread over large parts of many populations - not just specific groups (like the Pennsylvania Germans)
The admixture that I've seen on the DNAPrint log is far too particular to be indicative of a systemic problem with the test. Not only that, a lot of the admixture reported seems to make sense against a historical backdrop. Much of the admixture reported has, in fact, confirmed a great deal of history - you conveniently have left that part out.
You still haven't addressed how knowing the particular parental samples would enable you to asses the validity of the science. You don't mean to tell me that you can eyeball it, do you? I don't understand your one track mindedness either. If Ovanome has a successful clinical outcome towards the end of this year, do you really think any of this will matter? It won't.
I don't think I have enough time to debate this much longer but I basically disagree with every single one of your points.
First, I said that you, Mike Tiernan can't verify the parentals - not DNAP. LOL
Second, you say that it isn't important to other researchers - but that's not true either. DNAP isn't the only company in the world to be researching AIMs and they certainly aren't the only ones to have found them. Other researchers don't need the same algorithms that DNAP has developed nor do they have to be competitive. They only need correlate at least some of the AIMs that DNAP is looking for before DNAP does.
Then you say:
DNAP revealing their parentals is not going to give information to competitors that they don't already know. I can sit here and state my opinion of what would be appropriate parentals - I am sure that a population geneticist can do the same.
I think this is the very crux of the disagreement. You somehow think that DNAP developed this test by picking out which parental samples will define which BGA categories. Not true. Nobody has a road map to human migration patterns. You can certainly set up a test to do it that way but it's guaranteed to fail.
The DNAP test itself is establishing its own version of those migration patterns. Its heuristic and so validation is not from replication but by self consistency. (You've never understood that difference). But its sort of important because the alternative is what you propose: picking out parental samples and trying to correlate those to larger populations. The problem with your proposed method is that it doesn't accommodate complexity very well. Humor me for a second and suppose that there is indeed very high levels of admixture across the globe.....how the *hell* are you going to sort through all that with a "candidate parental" approach?
I'll ask you again:
Why would someone who's so concerned about his investment be so preoccupied with getting information that:
1. Is tangential to the success of the company
2. Could possibly hinder the progress of the company
3. Is not verifiable (you (Mike Tiernan) can't assess the legitimacy of the parentals)
Nobody's asking you to believe it - you forget that part. And even if you were given the parentals, there's no way for you to verify it. The example that you've given isn't analogous.
By your reasoning, it's ok to reveal parts of the test to satisfy (literally) one or two investors curiosity at the cost of every other investor. Personally, if they started giving out information that affects their competitive position just because someone out there asked for it, I'd be pretty pissed.
Why would someone who's so concerned about his investment be so preoccupied with getting information that:
1. Is tangential to the success of the company
2. Could possibly hinder the progress of the company
3. Is not verifiable (you can't assess the legitimacy of the parentals)
including potential investors and customers
Get out of here <ggg> There aren't any investors and very few customers who even care.
How can that be a "secret" "proprietary" information, but not the others.
I don't think this means much to DNAP as an investment so to some extent, I'm not as interested in it as you. But if you really want my unscientific opinion, I think DNAP is concerned that if they extend specific information that links parental populations to broader BGA groups then they are giving other researchers a leg up in finding specific AIMs before they do. DNAP isn't done yet and probably won't be for a while.
The Indo-European category is probably the only major category in which none of the traditional political boundaries therein overlap neatly with the sub populations of that category. It's an inherently messy category and we know that DNAP isn't the only one looking for AIMs here. So why should they reveal them? For your own curiosity at the cost of a competitive advantage?
I would bet that if a lot of researchers were to guess which parental populations would define a specific sub population, they would guess wrong on the first try. In fact, there's no way to know beforehand because there are no accurate, detailed migration maps available.
Mike - just so you know, they won't answer to you on this. lol. No amount of posting on Ihub will make them reveal the parentals.
Don't know, Bag. If you look at DNAP's patent applications, they've made a sort of mathematical tool palette with a number of tools that provide some redundancy and flexibility. If one method doesn't work then they try to solve it from more than one angle. Even though we're always talking about the ADMIXMAP, the reality is that they are using a number of tools to get what they want - not just a single method.
My impression is that this is a little more important than the other tools in the toolbox.
DNAPrint genomics, Inc. Announces Collaboration With Famous Genetic Epidemiologist Dr. Paul McKeigue
MONDAY, JUNE 28, 2004 5:37 PM
SARASOTA, Fla., Jun 28, 2004 /PRNewswire-FirstCall via COMTEX/ -- DNAPrint genomics, Inc. ("The Company") announced today that it has entered into strategic algorithm development collaboration with famous genetic epidemiologist Dr. Paul McKeigue. The goal of the collaboration is to develop novel statistical genomics algorithms for a new style of genome screening called Admixture Mapping.
Dr. McKeigue invented the Admixture Mapping (AM) technique in 1998. The technique is distinct from the Mapping by Admixture Linkage Disequilibrium (MALD) technique introduced by Dr. Ranajit Chakraborty in 1986. Both techniques employ knowledge of population structure to identify genes that underlie complex diseases and drug response, but Admixture Mapping is considerably more sophisticated in a mathematical sense, and uses the power of Bayesian Analysis and an "affecters only" study design to dramatically increase the statistical power for detecting these genes.
Soon after its founding, DNAPrint genomics began consulting with Dr. McKeigue and Mark Shriver of the Pennsylvania State University to augment Admixture Mapping methods and build Ancestry Informative Marker (AIM) libraries for efficient commercial-scale screening of the genome. DNAPrint is already using its own patent-pending variation of the Admixture Mapping method to construct predictive tests for tailoring drugs and doses to patients based on their genetic constitution. This platform has resulted in the identification of numerous genes involved in a number of both positive and negative responses to commonly prescribed drugs. DNAPrint intends to employ the new, improved ADMIXMAP algorithms for a variety of disease gene screens, many of which promise to be the first of their kind ever conducted.
"We believe that the mathematics and advanced computational methods we will be refining and developing will allow us to find elusive disease and drug response genes that others have missed using older, less powerful screening technologies," said Dr. Tony Frudakis, DNAPrint's Chief Scientific Officer. He continued, "This deal confers a sort of pole position for the Company with respect to this exciting new genome scanning methodology, because it is not just the maps of AIMs that are necessary for Admixture Mapping, but advanced analytical tools that are capable of accommodating parameter uncertainty."
"The mathematics at work here give DNAPrint an advantage because it allows us to research genetic mediated diseases and to develop or license compounds that work to treat the disorders," said Richard Gabriel, DNAPrint's CEO. "We will extend our reach further into the pharmaceutical industry with advanced mathematical and software programs as well as genomic technologies. We are very pleased to have Dr. McKeigue and his team helping us."
The most common U.S. and European pharmaceutical business model is based on the idea of root genetic causes for disease and small molecule (i.e. drug) targeting of the corresponding defective gene products and their corresponding biological receptors or pathways. DNAP plans to use Admixture Mapping to identify the most likely drug targets, not just the genetic markers of disease or its symptoms. This requires understanding how each individual's genetic ancestry affects treatment or disease progression.
The agreement announced today will provide support for a Bayesian mathematician in Dr. McKeigue's Dublin laboratory, and will enable DNAPrint's computer scientists to improve and augment existing ADMIXMAP code. Paul McKeigue holds joint professorships at the University College London, School of Tropical Medicine and Hygiene and the University of Dublin, Ireland, School of Genetic Epidemiology.
About DNAPrint genomics, Inc.
DNAPrint genomics Inc. uses proprietary human genome research methods to develop genomic-based services and products. The Company introduced ANCESTRYbyDNA in the consumer market in 2001 and DNAWitness in the forensic market in 2003. DNAPrint is developing products in the pharmacogenomic market and has a disease gene discovery program. The Company is traded on the Nasdaq OTC Bulletin Board under the ticker symbol: DNAP. For more information about the company, please visit http://www.dnaprint.com.
All statements in this press release that are not historical are forward- looking statements within the meaning of Section 21E of the Securities Exchange Act as amended. Such statements are subject to risks and uncertainties that could cause actual results to differ materially from those projected, including, but not limited to, uncertainties relating to technologies, product development, manufacturing, market acceptance, cost and pricing of DNAPrint's products, dependence on collaborations and partners, regulatory approvals, competition, intellectual property of others, and patent protection and litigation. DNAPrint genomics, Inc. expressly disclaims any obligation or undertaking to release publicly any updates or revisions to any forward-looking statements contained herein to reflect any change in DNAPrint's expectations with regard thereto or any change in events, conditions, or circumstances on which any such statements are based.
Media and Press Contacts
Richard Gabriel
DNAPrint genomics, Inc.
CEO/President
(941) 366-3400
SOURCE DNAPrint genomics, Inc.
Volume was 2x normal *before* the PR..... Also, that was phase I stuff - maybe a little too far out to be putting a 25% move in one day?
I'm still holding a little but if the close isn't strong I'm out.
Considering that the questions were being gathered by other people for the shareholder meeting its interesting....no, amusing that you've taken it upon yourself to create a new crusade. Such a dedicated shareholder!
edit: Not the "only" recourse. Should have said the "easiest". There are other ways to protect IP but they are reactionary to a hostile bid.
frog - IMO, that's a poison pill. It has been that way since the company was founded and it was certainly one of the first things the management team would have been aware of before joining the company.
For a company that perpetually trades below .05 and is undercapitalized, they have no real protection of their IP. Legal fees alone would destroy the company. As a precaution, the only recourse is to make the company a shell. If a larger entity were interested in acquiring our IP they would be forced to negotiate a deal as there would be nothing to take over.
I've thought about the various possibilities before but this is the one that makes the most sense to me.
Seems it pretty much depends on the presence of moisture and heat that promote microbial activity.
Exactly. It has to be properly stored and saved and then it will last a very long time. DNA is organic and so it will degrade except in very dry or anaerobic conditions.
So, lets say you found a cave man frozen and "in tact" in the Arctic Circle. I would bet there's a good chance that one could run an ADB test on him. There are several instances of centuries old creatures found in near perfect preservation. One that comes to mind was the discovery of a Celt (a Druid some say) that was found almost completely preserved after he had apparently drowned in a bog over 600? years ago. Of course there was moisture, but there was no aeration so a great deal of his body remained in its original state.
Mitochondrial DNA is a different story. For one thing, there's more of it - a lot more. Also, it's not as susceptible to degradation as nuclear dna. (That last part doesn't make sense to me, but that's what they say...)
I didn't do anything. lol. That was Matt who gave him the boot stamp.
he spammed over 30 boards in 10 minutes with an identical message, just changing the symbol.
http://investorshub.com/boards/profile.asp?user=39898
is that supposed to be ok? lol.
I'm all for more information but I'm not so sure about the "creativity" part. Every facet of marketing takes time and money and I think they should be concentrating on growing the core audience before moving to peripheral uses.
DNAP can't analyze Columbus' DNA. DNA goes "bad" if it isn't properly stored (like rotten vegetables). You could perform mitochondrial testing but not a test that covers all 23 chromosome pairs.
ming - I'm not so sure that they will move into the USF incubator space. Odds are that they will but the fact that they were the last ones to sign a letter of intent to lease even though they surely were one of the first to know about the project leads me to believe that they are either looking for something that is more specific to their needs or something that has greater flexibility.
Time will tell I suppose. I wrote IR about this a few weeks back and was told that they want to keep all options open until its time for them to make a decision. Also, if the incubator space was their first choice I suspect they would have already signed a lease and it would have been disclosed in the Q1 filing.
OT - No. Both the market and the economy have been sustained by a massive expansion of credit over the last year, i.e. it has been held up by liquidity rather than improving fundamentals (fundamentals being a healthier trade balance, reduced deficit spending, wage growth, improved participation in the labor force, and the creation of wealth - we have had none of these!). The problem with this is that the amount of credit and/or wealth creation necessary to sustain what recovery we've enjoyed thus far is remarkably steep and, imho, unsustainable over the next 12 months.
Too many "economists" are looking at this economy as if it were a perpetual motion machine that moves of it's own volition...I don't think so. If the market tanks from where we are now then we'll likely have a mid-summer rally but longer term, I think we've seen the end of this mini-bull and a resumption of the bear. IMO, the highs are already in and now it's just a question when "investors" figure out that the line between inflation and deflation is an extremely fine one and has been reduced to what amounts to a high wire act.
maybe you meant 'naked' short? Its perfectly legal to short the otc in the U.S. It's just that most brokers do not include pennies in their short sale eligibility list. The ones that do have astronomical margin requirements.
Thanks folks. Maybe you guys could write my eulogy <g>. Not leaving, I'm just very busy right now. I'll still post from time to time and I might moderate at some capacity later on.
Just too busy to be here all the time. Anyhoo...thanks!
(chris.....you're a cornball <g>)
Straw - I was sort of hoping my prior post asking members not to question other people's intentions as soon as they show up would suffice. Could we just leave it at that? Thanks.
A reminder to the board: let's please not question peoples motivations on the board. Mike Tiernan hasn't done anything wrong. Thanks.
ok - thanks. eom.
Let us say that DNAP could have been more clear.
Agreed. Or, at least, they should publish all relevant info concurrently with releasing a consumer product. Most people won't read it <g> but some people will. At the very least, nobody will be able to accuse them of hiding their science.
In the early days of 2.0, it was sometimes impossible to discern wether minority affiliation was from an ancient source or a more recent source. Once patterns emerged with certain populations, DNAP started posting some of the possibilities and theories and let the customers make their best guess.
IMO, DNAP will always enjoy a sort of love-hate relationship with the genealogy community. Maybe with 3.0 there will be a little more love <g>
Mike, I agree with some of what you say but not the part about "recent" admixture. It was never pitched as recent. I'm not sure who told you that but they need to go back and look at things again. It was just the opposite. For a long time, people were looking for recent events to explain minority affiliation while ignoring some of the broader possibilities raised by the company and yet to be proven migration theories.
I'm not saying that the company was in a position to predict many of the surprises that came out of ADB 2.0 or that it was a useful tool for many who took it - I'm just saying that it was never pitched as an accurate measurement of "recent" admixture. It was labeled as a measurement for ancient DNA from the very beginning.
When did they say that all admixture was recent admixture? I'm not bashing *you* lol. I'm just curious because my understanding of the evolution of Ancestry and the PA Dutch story is quite different from what you're saying.
Also, are you suggesting that there is a way to give constructive criticism merely by knowing the parental population? This is a bit over my head but I don't know how you can do it unless the parental samples are waaaay off. Up to this point all the constructive criticism I've seen has come after the fact.
I don't discount what you say for a second. But IMO, giving constructive criticism off a PR and a few numbers isn't going to be easy. I also don't think you're going to see this on the ABD website anytime soon either. I think it's rather unlikely this will be offered to the genealogy community in it's present form.
But the PR came out just last week. Wouldn't it be more realistic to wait for the supporting data to be published in a journal? That's how the first iteration of ABD was reported.
W2P - I don't know much about the approval process but my understanding was that is takes the structure of a "mentorship" whereby clients and other already-approved labs give feedback and recommendations to DNAP's forensic department.
It would seem to me that, in order for this to happen, DNAP would have to solicit steady business first. So there is a chicken and egg problem for private labs who don't represent a local law enforcement agency. Perhaps this is why they've started pushing DNAWitness and offering STR tests? Just a thought.
Question: or whether an East Asian's major ethnic affiliation is Northern (Chinese/Japanese), Central Asian or South East Asian.
Does anyone know what the ethnic/political boundaries for the above are (well, aside from Chinese/Japanese)? For instance, who would qualify for being 100% Central Asian?
more likely, they're pooling their resources towards the point of greater demand, IMO
speaking of desperate. If you had thought about it longer, I'm sure you could have come up with something better than that.
Very cool. Now if someone could just get GNET off the ask we might be headed somewhere.
Ming, I'd be willing to bet that the "disease gene discovery program" pertains to DNAP's contract service and not a disease gene discovery program for our own use. I'm thinking back to the TWST article were Frudakis said they would begin by finding solutions for other peoples drugs first, and then solutions to our own drugs later.
If that were the case, the good news would be that we were being paid!