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.
Anybody noticed how far NITE trading activity has fallen in the past two months. NITE has dominated trading in DNAP as long as I have been invested in the company, averaging approximately 40% of the overall share volume. Only one month in all of 2003 did their monthly volume fall below 40% (Max. month was 52%), and they did 30%.
Thus far, January and February 2004, NITE has accounted for only 16% of the total trading volume. Schwab and Roth Capital seem to have stepped in as the big players:
http://www.otcbb.com/asp/tradeact_mv.asp?SearchBy=issue&Issue=DNAP&SortBy=volume&Month=2...
Even GVRC, whose volume generally accounted for 20-25% of the monthly total, has fallen in 2004. Though some of their normal volume is probably also being taken up by the other two mentioned above.
Interesting.....
Later,
W2P
mingwan0...See, I KNEW you knew...lol Interesting bunch of fellows that seem to share common interests.
Later,
W2P
cosmic...Sorry. Here's a little more if you're going to be up anyway...lol
Murray Brilliant is also interested in pigmentation genes and has noted some rather significant disease associations, particularly neurological disorders (I seem to remember something about that in the ancestry patent claims):
http://bmcb.biology.arizona.edu/brilliant.html
Here it looks as if he's found a link between a mutated gene located on the same chromosome as some of the pigmentation genes, and SIDS (Sudden Infant Death Syndrome):
http://www.elks4kids.org/newsletter/mar2002.asp
And here he's working with Dr. Richard King, University of Minnesota exploring the connection to OCA1, OCA2 and albinism. This work sounds eerily familiar:
http://www.ihg.med.umn.edu/people/king.html
Later,
W2P
Complete text of "Sequences Associated With Human Iris Pigmentation", published in Genetics magazine, December 2003:
http://www.genetics.org/cgi/content/full/165/4/2071
Note the acknowledgment:
"...We thank D. C. Rao, Director of the Division of Biostatistics, Washington University, St. Louis, for help preparing this manuscript; Mark Shriver, Department of Anthropology and Human Genetics at The Pennsylvania State University for his help with the biogeographical ancestry admixture aspect of the project; and Murray Brilliant, professor of Pediatrics and Molecular and Cellular Biology at the University of Arizona for their kind advice and support of our work. We also thank Robert White for his help with sample collection. We sincerely thank the referees for their valuable suggestions for improvements on the earlier version of this article. Last, we thank the reviewers of this manuscript who suggested a number of important improvements."
A name that I'm certain our own mingwan0 is familiar with...
Later,
W2P
Mike...Oops, the correct post is 12883. Sorry. eom
Later,
W2P
Mike...This text is taken from the Description for DNAPrint's Ancestry Informative Markers Patent Application. There is a link to the complete description in my post #12882.
That's the only reference I have seen and haven't spent any time researching the significance of PSCA in general.
The figures will be in the complete application which may not be available for some time. I'd look for them to be published at ESpace.net at some point in the future:
http://gb.espacenet.com/espacenet/gb/en/e_net.htm?search5
Later,
W2P
ming...You mean THESE claims?
12. The method of claim 1, wherein the trait comprises responsiveness of the individual to a drug.
13. The method of claim 12, wherein the drug is a cancer chemotherapeutic agent.
14. The method of claim 12, wherein the drug is a statin.
15. The method of claim [1,] wherein the trait comprises susceptibility to a disease.
16. The method of claim 15, wherein the disease has an ethnic predisposition.
17. The method of claim 16, wherein the disease is a cancer, diabetes, or hypertension.
18. The method of claim 17, wherein the cancer is prostate cancer.
19. The method of claim 15, wherein the disease is a neurological disorder.
20. The method of claim 19, wherein the disease is schizophrenia or Parkinson's disease.
21. The method of claim 15, wherein the disease is alcoholism.
22. The method of claim 1, wherein the trait comprises a pigmentation trait.
23. The method of claim 22, wherein the pigmentation trait comprises eye color, skin color, hair color, or a combination thereof.
Oh well, you get the point. Ancestry Patent Application Description and Claims:
http://www.wipo.int/cgi-pct/guest/ifetch5?ENG+PCT-ALL.vdb+14+1091049-SCORE+256+3+22978+DECL-ENG+2+7+...
Later,
W2P
Ancestry 3.0?
The [ANCESTRYBYDNA] 1.0 test (DNAPrint [GENOMICS,] Inc. ) is a first version of the BGA test that was specifically designed to provide information on the proportions of ancestry at the continental level. As such, the [ANCESTRYBYDNATM] 1.0 test allowed information to be obtained as to levels of Native American, European, and African ancestry, as three component groups. The [ANCESTRYBYDNA] 2.0 test, in comparison, provides information on the proportions of ancestry at the continental level for most continents, including Native American, Indo-European (includes European, Middle Eastern and South Asian groups such as Indians), African, and East Asian (which includes Pacific Islanders, and can distinguish ancestries within Asia and the Pacific Rim. The [ANCESTRYBYDNATM] 3.0 test can further define the levels of ancestry within continents, for example, by distinguishing Japanese from Chinese, or Northern European from Middle Eastern, thus providing greater insight into where within a particular continent a person's ancestors were derived.
[[0168]] For the [ANCESTRYBYDNATM] 2.0 test, a logical grouping into four BGA delineations was made, wherein South Asian, Middle Eastern and European are grouped into a single group called IndoEuropean (see Example 2). This grouping was based on anthropological evidence and cultural connections between these groups (e. g. , their languages are derived from a common base). The results disclosed herein demonstrate that these groups are far more similar to one another in genetic sequence content than to other groups. The [ANCESTRYBYDNA] 2.0 test also performs more accurately when Pacific Islanders are grouped with East Asians. As such, the four groupings used in the [ANCESTRYBYDNA] 2.0 test include 1) Native American (i. e. , those who migrated to inhabit South and North America); 2) IndoEuropean (Europeans, Middle Easterners and South Asians such as Indians; 3) East Asians (Japanese, Chinese, Koreans, Pacific Islanders); and 4) Africans (sub-Saharan). The [ANCESTRYBYDNATM] 3.0 test can further distinguish between South Asian and European, and between Pacific Islander and East Asian, thus providing 6 proportions (Native American, European, African, South Asian, East Asian and Pacific Islander), although the confidence intervals are larger than those obtained with the [ANCESTRYBYDNA] 2.0 test. Further improvement to the tests are provided, wherein the confidence intervals are reduced. Confidence intervals around a point estimate can be reduced, thus increasing the accuracy of the test, by analyzing a complementary panel, thereby improving the confidence intervals by about 50%.
Later,
W2P
Ancestry 3.0?
The [ANCESTRYBYDNA] 1.0 test (DNAPrint [GENOMICS,] Inc. ) is a first version of the BGA test that was specifically designed to provide information on the proportions of ancestry at the continental level. As such, the [ANCESTRYBYDNATM] 1.0 test allowed information to be obtained as to levels of Native American, European, and African ancestry, as three component groups. The [ANCESTRYBYDNA] 2.0 test, in comparison, provides information on the proportions of ancestry at the continental level for most continents, including Native American, Indo-European (includes European, Middle Eastern and South Asian groups such as Indians), African, and East Asian (which includes Pacific Islanders, and can distinguish ancestries within Asia and the Pacific Rim. The [ANCESTRYBYDNATM] 3.0 test can further define the levels of ancestry within continents, for example, by distinguishing Japanese from Chinese, or Northern European from Middle Eastern, thus providing greater insight into where within a particular continent a person's ancestors were derived.
[[0168]] For the [ANCESTRYBYDNATM] 2.0 test, a logical grouping into four BGA delineations was made, wherein South Asian, Middle Eastern and European are grouped into a single group called IndoEuropean (see Example 2). This grouping was based on anthropological evidence and cultural connections between these groups (e. g. , their languages are derived from a common base). The results disclosed herein demonstrate that these groups are far more similar to one another in genetic sequence content than to other groups. The [ANCESTRYBYDNA] 2.0 test also performs more accurately when Pacific Islanders are grouped with East Asians. As such, the four groupings used in the [ANCESTRYBYDNA] 2.0 test include 1) Native American (i. e. , those who migrated to inhabit South and North America); 2) IndoEuropean (Europeans, Middle Easterners and South Asians such as Indians; 3) East Asians (Japanese, Chinese, Koreans, Pacific Islanders); and 4) Africans (sub-Saharan). The [ANCESTRYBYDNATM] 3.0 test can further distinguish between South Asian and European, and between Pacific Islander and East Asian, thus providing 6 proportions (Native American, European, African, South Asian, East Asian and Pacific Islander), although the confidence intervals are larger than those obtained with the [ANCESTRYBYDNA] 2.0 test. Further improvement to the tests are provided, wherein the confidence intervals are reduced. Confidence intervals around a point estimate can be reduced, thus increasing the accuracy of the test, by analyzing a complementary panel, thereby improving the confidence intervals by about 50%.
Later,
W2P
How do we differ from McKeigue's work:
There is growing evidence that admixture mapping will be an effective means of gene identification. At least three independent groups have reported strong admixture linkage disequilibrium (ALD) between linked markers spaced at substantial distances (see, e. g. , Parra et al., supra, 1998 and 2001; Pfaff et al. , supra, 2001 ; McKeigue et al., supra, 2000). Given the very high levels of association that have been observed over long genetic distances, it is expected that phenotypes dramatically different between parental populations because of some genetic difference will also show associations with linked AIMs. However, as promising as that MALD approach appears, until the present disclosure, no systematic screen has been reported identifying SNP based versions of the AIMs required. McKeigue and others have identified panels of STR AIMs for use with this approach, but the use of STRs for this purpose is problematic because of the allelic complexity of STRs and the massive databases required in order to accurately estimate allele frequencies. Even small errors or faulty assumptions on the frequencies of unobserved alleles can amplify to cripple the statistical power of a study.
Later,
W2P
For those concerned about the HapMap Project, this section describes the difference between DNAPrint's work and that of the HapMap Project Team:
Admixture mapping as disclosed herein is complementary to, but distinct from, the HMP. First, the primary focus of the HMP is to understand the fine scale structure of individual genomic regions throughout the genome, whereas the present methods allows an understanding of the LD that results specifically from admixture. The level [OF LD] from admixture is on the order of millions of bases (Mb; megabases) and tens of Mb, while the HMP is focused on the level of [10'S] to [100'S OF KILOBASES (LIB),] and genomic and population features that affect the results from one project may not be noted in the other. Second, admixture mapping require accurate parental allele frequency estimates. As such, a large number of different African, Native American, European, and Asian populations have been typed (see Table 6, below), while the HMP will likely focus on one or two samples of the major population groups.
[[0116]] Third, large samples (n = 500) of African-Americans and Hispanics have been typed, thus providing sufficient statistical power to test the coverage of the admixture map and to compare analytical methods. In addition, several representative populations from different regions of the country were typed so that geographical variation in ancestral proportions and admixture dynamics can be examined. Although some admixed populations will likely be included in the HMP, the numbers of individuals and numbers of different population samples being discussed are fewer than those as disclosed herein and, therefore, will not allow the same comparisons. For example, having a sample of 10 for each of 4 ancestral groups is not adequate for the identification of sequences present preferentially in one or some of those groups ; as disclosed herein, at least 50 individuals were tested for each of several tens of ancestral groups (not just four) in order to comprehensively identify these markers.
[[0117]] Fourth, the focus of current population variation efforts (e. g. , the SNP Consortium allele frequency project) and, very likely, the HMP has been on East Asian, African, and European samples to the exclusion of Native American populations for a number of complex reasons. The exclusion of these populations, however, results in a deficit in an understanding of the genetics of the fastest growing group of US resident populations, i. e. , Hispanics, who have a significant level of Native American ancestry (20% to 40%). With the markers and methods disclosed herein, the disease genetics of Hispanic populations can be examined.
Similarly, several diverse Native American populations may represent important parental populations for the numerous distinct groups often grouped together as Hispanic.
Later,
W2P
Native Americans would be interested in this:
There are clear differences in the patterns of chromosomal segment ancestry (PCSA) among persons with different ancestral histories (see Figure 1). A series of AIMs across the chromosomes can facilitate the estimation of the most likely parental combinations that lead to the profile of sequences observed in a given person. One example of where estimates of PCSA is important is in the discrimination of persons of Hispanic ancestry from those having primarily European ancestry with some proportion of recent Native American ancestry. Indeed, this is an important determination as the political and legal rights claimed by and provided to these two groups can depend on their ancestry. Hispanic populations such as Mexican-Americans (MA) have approximately 30-40% Native American ancestry, while the balance is European with a minor portion (5% or so) African ancestry. A person who is one quarter Native American will have 25% Native American ancestry and, therefore, will overlap with many MA persons in his level of estimated ancestry. It is expected that PCSA patterns will be significantly different for these two cases and may provide some of the only genetic evidence that would facilitate an accurate definition of the ancestry in such a case.
Later,
W2P
This one's pretty potent:
Not withstanding the power of AIMs for disease gene and forensics analysis, no studies have been conducted to elucidate this power. As disclosed herein, 1) SNPs or deletion/insertion polymorphisms (collectively referred to as AIMs) in the human genome that are of potential use for drug response, disease gene or forensics research were identified ; 2) biochemical and genetic test results are provided that demonstrate these AIMs can be useful for disease gene and forensics research ; 3) the usefulness of AIMs derived from systematic screens of the human genome in actual drug response, disease gene or forensics research is demonstrated; 4) the usefulness of AIMs derived from systematic screens of the human genome to make an inference as to whether an individual is susceptible to acquire a disease, or to not respond to a drug, is demonstrated; 5) the usefulness of AIMs derived from systematic screens of the human genome to make an inference as to whether a crime scene DNA specimen was derived from an individual of, for example, an 80% European, 10% African and 10% Asian heritage or some other ratio/mix is demonstrated; 6) the usefulness of AIMs derived from systematic screens of the human genome to infer the ancestral proportions of an individual from their DNA (e. g. , whether the individual is of 80% European, 10% African and 10% Asian heritage, or some other ratio/mix) is demonstrated; and 7) the usefulness of AIMs derived from systematic screens of the human genome to infer the ancestral proportions of a group of individuals from their DNA (for example, whether the group, which can be a population sample, a family, or a clinically defined group of persons, is of [80% EUROPEAN,] 10% African and 10% Asian heritage, or some other ratio/mix) is demonstrated.
The present results demonstrate that AIMs are useful for the applications described above, and the sequences exemplified herein, as well as additional AIMs identified using the methods disclosed herein, enable these applications. The AIMs and methods of the invention are useful for the study of human diseases, drug response, and physical traits and, therefore, provide exceptional commercial potential. For example, in this dawning era of personalized drug prescription and disease risk assessment, the markers and methods of the invention provide the tools needed to proceed in this fledgling industry. As exemplified herein, an individual's response to a particular medication was dependent on the degree to which that individual exhibited a certain population structure (i. e. , was of certain ancestral heritage) in addition to, but irrespective of, the person's genotype for drug target or xenobiotic metabolism gene sequences. As such, the compositions and methods of the invention provide a means to predict an individual's likelihood to respond to a particular drug.
Later,
W2P
I liked this paragraph as well:
There has been increased interest in association studies as a useful approach to map common disease and drug response genes (Risch and Merikangas, Science 273: 1516-1517, 1996; Jorde, Genome Res. 10: [1435-1444,] 2000; Nordborg and Tavare, Trends Genet.
18: 83-90,2002). Until the present disclosure, however, the implication of ancestry for identifying these genes has not been fully appreciated. As such, the methods of the invention provide a previously undescribed platform for the identification of genes associated with disease susceptibility and drug responsiveness, as well as for the development of advanced forensic methods. As such, compositions and methods are provided for inferring an individual's response to commonly used medications, which, remarkably, is a function of individual ancestry; the disclosed markers and methods are, to a differing extent for each drug, useful for the inference of such response. In addition, compositions and methods are provided for inferring individual and/or group ancestral proportions from knowledge of the individual's or group's DNA sequences. Further, compositions and methods are provided for using knowledge of ancestry relevant DNA sequences to identify disease susceptibility and drug response genes through the MALD process. Also, compositions and methods are provided for qualifying and normalizing study groups for more traditional methods of mapping disease genes. Each of these processes requires an accurate knowledge of ancestry, which can be determined using the methods and compositions disclosed herein.
Later,
W2P
Dr. Dolittle, here we come:
A test individual for whom a trait is to be inferred can be any individual for whom it is desired to infer a trait, and generally is a human. However, the methods of the invention also can be used for inferring traits of other mammals, including, for example, domestic animals such as cats, dogs, or horses; farm animals such as cattle, sheep, pigs, or goats; or other animals. The trait to be examined can be any trait of interest, including, as exemplified herein, proportional ancestry (BGA); hair, skin or iris pigmentation; or drug responsiveness.
Later,
W2P
Another excerpt:
The primary element of the disclosed methods is that most human traits can be predicted through a detailed measurement of AIMs associated with population structure at various levels, provided the trait correlates with that element of structure. A secondary element is that classifiers, or collections of SNP markers, and methods for predicting trait value from DNA can be constructed for most human traits through such an appreciation of population structure. As disclosed herein, such applications can be accomplished through correlation, not just through extended LD found in certain admixed groups such as Hispanics or African Americans, but for any sample of subjects, regardless of race or ethnic background, provided that the AIMs used are appropriate for the element of population structure with which the trait is correlated. These results can be attained because the methods of the invention provide a means to mine the genome for good AIMs, qualify their value as AIMs and accurately measure population structure against the backdrop of human phenotypes.
The use of population structure to infer the trait is the phenomenon DNAPrint has referred to and can be used to infer ANY trait. According to my previous post, they have already catalogued 54 drug response traits for 23 different drugs.
Later,
W2P
mingwan0...Don't mind if I do...
When combined with markers from genes that are biologically relevant for response, they augment the ability to make accurate inferences of response from the DNA. Each of these markers also is an excellent AIM, indicating that the linkage of the AIMs to drug response is likely a function of ancestral differences in response proclivity (see Example [5).] As such, ancestral heritage can be predictive of favorable response to [LIPITOR.] This association has been observed for almost every type of response (n = 54) to almost every type of drug (n = 23) examined, thus confirming that the inference of drug response can be accomplished, at least in part, through the inference of ancestral proportions. As such, it appears that the genes truly relevant for drug response are a function, at least in part, of individual ancestry, and that the gene sequences relevant for drug response are statistically linked with markers that are informative as to ancestry (i. e., [AIMS).]
DNAP has catalogued 54 different response types to 23 different drugs and linked them genetically to Ancestry.
Later,
W2P
mingwan0...Will you be doing the honors? lol
Later,
W2P
Cosmic/Theo...And to complete the record, John J. Oskorep, the attorney named in their previous applications was also employed by Gray Cary at one time. He wrote some rather interesting articles as well.
In this one, he presents the legal precedents that demonstrate the patentability of "abstract ideas", as are exemplified by informatics and algorithms. When you get to the site, it may ask you to select a state. If that happens, simply click your state and it WILL take you to the article:
http://articles.corporate.findlaw.com/articles/file/00051/001340/title/Subject/topic/Intellectual%20....
BTW, Mr. Oskorep is also an electrical engineer and filed a patent of his own, along with his brother, in December 2003.
Later,
W2P
By the way...this name has been showing up rather frequently on DNAPrint's Patent Applications. Thought the Board might be interested in her Bio. She is a PhD, and NIH Post Doctoral Fellow herself:
http://www.graycary.com/gcc/GrayCary-C/Attorney-S/D-H/lhaile.doc_cvt.htm
Looks to me like we're in good hands...
Later,
W2P
bag8ger/ebo783...Thanks for the kind words. I keep an eye out when I can, but have a few things on the plate yet. Just had some time this weekend and thought I'd drop in.
Still watching and waiting...have a great week.
Later,
W2P
mingwan0...Fascinating indeed. There is a great deal of discussion in this patent concerning the advantages of SNP based identity testing versus STR testing and discussion of skin color and ancestry classifiers. There is so much more than simply iris color (as if that's not enough..lol).
Your point concerning evolutionary history is a good one. It wouldn't surprise me to learn that the "phenomena" reference in the AIM patent is related to a discovery concerning genetic evolution of the human (or even extending generally to mammalian) species. (JMO) Thus, the work with Gavin Menzes.
The other thing that struck me were the sections that relate their statistical results back to functional biology. These classifiers not only identify markers useful for feature inference, but they inform concerning interaction of specific genes in ways that will allow research biologists to advance their understanding of process.
Finally, the ability to draw conclusions concerning markers for epistatic variance is fascinating to me. In essence, they not only identify markers to infer what IS there, they find markers that infer what ISN'T!
Have a great evening..
Later,
W2P
It's a bit long and bit technical in places, but this represents the meat of the complete classifier developed for iris color.
Just to assist in your reading, there are some genes or markers, referred to as penetrant genes, which in and of themselves can be strongly linked to a particular condition or feature. Others, termed latent genes, can be identified that are not direct indicators themselves, but which along with a penetrant strengthen the inference. What DNAPrint found was that of the genes found to be involved in variable eye color, the majority were, in fact, latent associations. Lastly, where they refer to epistatic features, they are talking about genes that function to SUPPRESS the function of other genes.
If you muddle through, much of this is understandable. Have fun:
Weighted Quadratic Classification Using Only the Penetrant Genetic Features
[0889] To generate a complex model by which to explain more iris color variance, to an extent that accurate inferences could be made, a weighted quadratic classification algorithm was developed based on standard coordinates from a correspondence analysis (see methods). We first used the penetrant genetic features to compute and weight a variance-covariance matrix (see methods) from 330 Caucasian individuals. This matrix was applied for a blind, quadratic discriminate classification of iris colors in 286 other Caucasians of known but concealed iris color. For the first analysis two groups were defined; a light iris shade group defined as individuals of blue, green or hazel irises, and the dark iris shade group defined as individuals of brown or black irises. On the level of the multilocus genotypes (gene-wise genotypes), an overall accuracy of 98% was obtained for this discrimination. The sensitivity for dark iris color shades was 100% and the sensitivity for light eye color shades was 97% (reading along the rows, Table 6a). The light iris classification was 100% accurate and the dark iris classification was 94% accurate (reading down the columns, Table 6b). Using this method at the level of individual SNP alleles, SNP genotypes or individual haplotype alleles produced lower accuracies (with accuracies in increasing order), suggesting that the highest level of intra-genic allele complexity is required for accurate inference of eye color shade and that increasing levels of complexity offer successively greater predictive power. Using the method with multilocus genotypes to infer actual eye colors, rather than just eye color shade, 100% sensitivity was obtained for blue iris classification, 69% sensitivity of brown iris classification, 100% sensitivity of green iris classification and 84% sensitivity of hazel iris classification (reading along rows, Table 6B). The accuracy of blue iris classification was 67%, of brown iris classification 100%, of green iris classification 100% and of hazel iris classification 74% (reading down the columns, Table 6B). Using simulation to estimate the inference power of the quadratic classifier we obtained a log likelihood of r=1.96 (not shown). In effect, the classifier was remarkably accurate and sensitive, with good inference power, but its deficiency was apparent in the misclassification of brown and hazel iris individuals into the blue iris group.
[0890] By adding the latent genetic features to this analysis (latent+penetrant genetic features), the optimal weighting strategy produced a covariance matrix that blindly generalized to the same 286 Caucasians with 100% accuracy and sensitivity for discrimination of light versus dark iris color shades. The optimal model also generalized to this sample with 100% accuracy for the inference of actual iris colors (286/286 correctly classified; along diagonal of Table 7A). Using simulation to estimate inference power of the quadratic classifier, we obtained a log likelihood of r=3.22 for classification into the proper iris color group. Though it is true that markers over-represented in racial groups of average darker iris colors would help the classifier artificially infer eye color in a multi racial sample, it is not true that any such markers would help with the inference of iris colors in Caucasians unless they were functionally relevant for human iris coloration. That these markers contributed towards the classifications within Caucasians suggests that they are functionally related to, or linked to markers functionally related to iris color determination.
C. Discussion
[0892] A complex classifier is presented in this Example for the inference of human iris color from DNA. To our knowledge this is the first such classifier described. Though the pigmentation genes are well documented, until this work, merely a handful of SNP alleles were known to be weakly associated with natural distributions of iris colors in the healthy Caucasian population. The reason for this is that most work attempting to describe natural variation in iris colors has focused on simple genetics approaches, such as single SNP analysis in single genes including the TYR (Sturm et al., Gene 277:49-62, 2001), MC1R (Valverde et al., 1997) and ASIP (Sturm et al., Gene 277:49-62, 2001) genes. By developing new complex genetics methodologies and adopting a systematic approach for identifying and modeling genetic features of variable iris color, we looked at the problem through more of a complex genetics lens than others previously. Nevertheless, most of our results agree with the previous literature. Though the TYR expression product is the rate-limiting step in the catalytic chain leading to the synthesis of eumelanin from tyrosine, previous studies by others have belied the more simple hypothesis that TYR polymorphism is a principle (i.e., penetrant) component underlying normal variation of human pigmentation (Sturm et al., Gene 277:49-62, 2001). The present study also failed to identify penetrant genetic features of variable iris color in the TYR gene. In addition, our systematic approach for identifying penetrant genetic features independently confirmed that the "red hair" SNP alleles described by Valverde et al., Nature Genet. 11:328-330, 1995 and Koppula et al., Hum. Mutat. 9:30-36, 1997 are indeed associated with iris colors. However, our work has extended even these simple gene-wise analyses. While there are no SNPs or haplotypes within the TYR gene associated with iris color, TYR alleles are important within a complex genetics context for the inference of iris colors. While the "red hair" SNPs are indeed associated with natural iris colors (in Irish individuals), they seem to be most strongly associated with Caucasian iris colors within the multilocus context of another coding change in the MC1R gene, and even then, they represent merely one stroke of a larger portrait.
[0893] In fact, one of the most important points to be taken from the work presented herein is that speaking of variable iris color on the level of individual genes is illogical due to the complexity of the trait. The fact of the matter is, neither TYR nor MC1R, nor for that matter any of the other genes we surveyed, are very important for predicting iris colors on their own. This was indicated by the Bayesian conditional probabilities we obtained, which for even the most strongly associated alleles (the penetrant genetic features), were too low for their use as independent classifiers. Since the variance of any complex phenotype is a function of additive, dominance and epistatic genetic variance (in addition to environmental variance) any good complex genetics classifier must capture each of these three components when making inferences, and the classifier we have developed seems to be able to this. The additive component is captured most efficiently through the analysis of multilocus alleles (haplotypes) and the dominance component is captured by expressing individuals as vectors whose components are encodings of multilocus genotypes for each important region. The most innovative advance we have made here is algorithmically capturing the epistatic component. Our work showed that there is a minimal set of 25 penetrant SNPs, of 8 multilocus contexts in 4 genes that are required for minimal inference accuracy. However, a complete set of 57 SNPs, of 19 multilocus contexts (both penetrant and latent), in 7 of the 8 genes is needed for accurate inference. That latent genetic are needed for accurate inference suggests that there is a significant epistatic component to iris color variance in the Caucasian population. The agouti signaling protein (ASIP) harbored four and the silver locus (SILV) harbored three such polymorphisms, each of which was arbitrarily combined into a single latent feature SNP combination. DCT and TYR harbored five and six such polymorphisms, respectively. That no penetrant genetic features were identified in ASIP, SILV or TYR suggests that these genes contribute towards iris color variance largely through epistatic means. The latent features are not equivalently predictive, and to capture the epistatic component during classification, we randomly ascribed weights to different alleles in different contexts and selected the combination that allowed for the most optimal quadratic discrimination. Our results suggest that there is much to be learned about the genetics of iris color from a detailed inspection of this optimal weighting scheme. At present, we do not understand the mechanism by which the features fit together the way they do in the optimal COA-derived quadratic classifier model (we intend to present these data elsewhere), only that they do and that the fit is of maximal practical utility for the inference of iris colors. The results we have obtained suggest that iris color is indeed a complex genetic trait, the "whole" of which was empirically determined to be greater than the sum of it's "parts". On a more general level, our results illustrate a seemingly obvious but interesting concept: simple genetics approaches are useful for ascribing trait associations for individual genes and haplotypes within them, but because most human traits are complex, complex genetics tools are required for their use in the development of accurate classification tests. Given the sources of error for this work, including genotyping errors, errors in self-reported iris color and statistical haplotype inference, it is quite remarkable that perfect classification accuracy was achieved with a combined sample size of 550 for such a complex trait. In terms of feature modeling, almost identical results were obtained using a classification tree (CART-based) method (unpublished data), even though the cost function of the method we used herein relates genotypes (haplotype pairs) to trait values in a more direct way than CART. Thus, it appears that the methods we employed herein are substantiated by other analytical methodologies and may be promising for the generation of other complex genetics classifiers, for example pharmacogenomics or complex disease genetics classifiers.
[0894] Though there are a number of processes, developmental and cellular, that could explain iris color variance, our results suggest that polymorphisms in merely seven genes explain all of the variation in iris colors in the population. This result is surprising. Studies in Drosophila have implicated over 85 genes in iris pigmentation (Ooi et al., EMBO J. 16(15):4508-4518, 1997; Lloyd et al., Trends Cell Biol. 8(7):257-259, 1998) and far more than 8 genes have been implicated in oculocutaneous albinism in model vertebrates. That almost all of iris color variance in human beings can be explained by polymorphisms in 7 of 8 carefully selected genes, given the biological complexity of pigmentation, illustrates that just because a gene is crucial for a process (i.e., its mutation causes loss of function) does not necessarily mean that natural distributions of this process among individuals is related to natural polymorphisms in this gene. By way of analogy, there are many ways to break an automobile engine--removing a water hose for example--but virtually none of the variability in engine performance is caused by variability in hose characteristics. Certain parts of the complex genetics "engine" seem to have become sinks for accumulating functionally relevant polymorphisms during the evolutionary branching of our ancestors.
[0895] In fact, one of the surprising findings of our work was that of all of the genes we tested, the OCA2 gene explained by far the most iris color variance. Five of the 8 feature SNP combinations were from the OCA2 gene and 17 of the 25 SNPs part of these penetrant feature SNP combinations were OCA2 SNPs. To date, no polymorphism screens within OCA2 have yet been described (though they had been called for--see Sturm et al., Gene 277:49-62, 2001) and this work is the first indication of the importance this gene has for natural iris color pigmentation. The OCA2 gene product localizes to the melanosomal membrane and resembles an E. coli Na+/H+anti-porter. Though TYR activity correlates perfectly with eumelanin content in melanosomes (Iozumi et al., J. Invest. Dermatol. 100:806-811, 1993), its activity is thought to be manipulated by the OCA2 gene product through the control of intramelanosomal pH (Ancans et al., J. Invest. Dennatol. 117:158-159, 2001). Tyrosinase taken from dark and light skin functions identically in-vitro, but is highly pH dependent and melanocytes from white skin are more acidic than those from black (Fuller et al., Exp. Cell. Res. 262:97-208, 2001, Ancans et al., Exp. Cell. Res. 268:26-35, 2001). Given these observations, it seems that OCA2 is the primary modifier of TYR activity, which is consistent with our statistical results. It is interesting to note that at the level of the cladogram analysis, four of the five allele associations were obtained for OCA2 feature SNP combinations. It is also interesting to note that the diversity of alleles associated with darker iris colors is significantly greater than that of alleles associated with lighter iris colors. These observations combined suggest that lighter colored irises branched from darker colored irises relatively long ago in human evolutionary time, and that modifications to the OCA2 gene may have been instrumental in this branching. The generally accepted anthropological and molecular view of the origin of modern humans from Africa states that Northern Europeans branched from African founders. Our results suggest that the reason lighter colored irises are almost exclusive to individuals of Northern European ancestry is in large part due to relatively ancient (and numerous) modifications of the OCA2 expression product. The fact that brown classifications were far more accurate relative to blue before, but not after, the addition of the latent genetic features to the classifier model may indicate that blue irises are subject to more epistasis than dark, and that dark eyes tend to be relatively (though not strictly speaking) dominant.
[0896] When applied to a multi-racial sample, the penetrant feature (as well as the combined penetrant+latent feature) classifier performed with substantially better accuracy than when applied only to Caucasians. Since most non-Caucasian ethnic groups exhibit low variability in iris colors (on average of darker shade than Caucasians) this improvement may not seem surprising. However, though an incorrect solution would not necessarily be more accurate when applied to individuals of the world's various populations, notwithstanding genetic heterogeneity, a correct solution would be. The reason for this is that if alleles associated with darker iris color in Caucasians are deterministic, or linked to deterministic alleles for melanin production and iris color, and if we assume the between race component of iris color variance is low, the frequencies of these alleles should be greater in populations of average darker iris color. Because the accuracy of both our models increases when applied pan-ethnically, our results suggest that the penetrant and latent associations we have described are functionally relevant. Since most of the SNPs are intron or silent changes, we infer that the alleles we have described are statistically linked with other unidentified alleles, or are functional in ways other than through amino acid changes (such as RNA transcription, degradation, localization etc.). It is interesting that those that were amino acid changes tend to be changes in polarity, three of four involving an Arginine. Interestingly, the classifier we have generated for iris color does not accurately extend for classification of hair color or skin shade within Caucasians. In fact, this is what one would expect from a good complex genetic model for variable Caucasian iris color, since iris, skin and hair color are known to be independently inherited (and distributed) within this racial group. We have conducted a study similar to the one described herein for hair color and though there is about 33% overlap between the SNP marker sets, the sets are distinct (data be presented elsewhere). We assume that the classifier generated here would be, at least in part, extendable to other racial groups, such as for the discrimination between green, hazel and brown irises in individuals of African descent. Whether or not this is true is a subject for further study.
[0897] As the first genetic solution capable of ascribing qualitative characteristics from anonymously donated DNA, our results represent an important achievement. First, they illustrate one method for modeling complex human traits from high-density genomics data sets. Second, as a forensics tool, our solution could be used to guide criminal or other forensics investigations (in this case, multilocus genotype combinations that are relatively ambiguous could be classified with regard to iris color shade and conditional probability statements offered for specific iris color classifications). Third, as a research tool, the common haplotypes we have identified may help researchers more accurately define the complex genetics risks for pigmentation related diseases such as cataracts and melanoma.
Later,
W2P
slopster...Here's the link to the International Patent publication originally published June 5, 2003. It was republished along with the International Search Report on August 21, 2003:
http://www.wipo.int/ipdl/IPDL-IMAGES/PCTI2C-PDF/images2.html?/2003/342003/US0238345_21082003/+3+1+1+....
US Provisional Application 60/334,310 was originally filed November 28, 2001. It appears to have been converted to 60/410,363 on September 11, 2002 (Note: reference on DNAPrint's website to application 60/140,363 appears to be a typo. Obviously should be 410,363).
I would not expect it to be published by the USPTO for 18 months after the conversion date, 9/11/2002. That would mean we could expect publication around mid-March 2004. If you don't see anything by then, let's chat...lol
Later,
W2P
Slopster...mingwan0 posted the dates of the first "Provisional Applications" to help the board understand the timeframe for certain key associations.
Provisionals are designed to be updated, and at some point must be converted to "Utility Applications". As "Provisional Applications" they would naturally expire when converted. The utility patents for Ovanome and Statnome are still under review at the Patent Office.
Nothing to get concerned about. Just part of the normal process.
Later,
W2P
Ok, I'll say it. Great find mingwan0...this is obviously the patent application referred to in the August 20, 2002 PR concerning the 2,425 SNP's and the "phenomena" discovered by DNAPrint. Here's the relevant excerpt:
As such, DNAPrint believes it is the first to claim markers of this type and elucidate the potential of this new subset of the variable human genome as specifically relevant for predicting drug response.
The new patent application could provide DNAPrint a tremendous advantage towards developing pharmacogenomics classifiers that are specific, sensitive and predictively powerful enough for routine clinical use. "Most of these SNPs have been ignored by the genomics community. Though others may have unwittingly and indirectly linked a very small fraction of these SNPs with variable drug response, technical and conceptual considerations have evidently prevented them from yet recognizing the underlying fundamentals of these associations," said Tony Frudakis, Ph.D., DNAPrint's CEO. "Such recognition would be required to generate a competitive patent application." Indeed, a review of the journal and patent literature reveals no reports describing the phenomena that the Company believes cause the linkages.
...To maintain its competitive advantage, the company will refrain from presenting details of the discovery until the findings are published by the US Patent and Trademark Office.
And of course, just a week later we got this one which has kind of fallen off the radar:
DNAPrint Obtains a Supply Agreement on the LNA Technology From Exiqon
Wednesday August 28, 11:29 am ET
Press Release Source: DNAPrint genomics, Inc.
SARASOTA, Fla.--(BUSINESS WIRE)--Aug. 28, 2002--DNAPrint genomics, Inc. (OTCBB:DNAP - News) announced today that it has obtained a non-exclusive right of supply from the Danish company Exiqon A/S to incorporate Locked Nucleic Acid (LNA(TM)) technology into certain of its pharmacogenomics and forensics tests.
DNAPrint kits use small DNA probes to query specific single nucleotide polymorphisms (SNPs) within an individual's genome. The probes contain molecular "addresses" that targets each particular SNP much like a street address targets a postal letter. Most kits made today contain natural DNA; however there are a number of technical advantages associated with using artificially modified DNA molecules instead. The supply agreement signed today enables DNAPrint to use a specially modified DNA called LNA(TM) in the construction of its test kits. Exiqon holds the patent for the LNA(TM) technology.
The use of LNA(TM) (Locked Nucleic Acid) is expected to provide superior technical performance and robustness for DNAprint kits and services. LNA(TM) is a novel class of DNA analogs that provide outstanding improvements in a number of key DNA hybridization properties. Of paramount importance for diagnostic use, the technology combines exquisite binding affinity towards complementary DNA, but with an unusual ability to discriminate between matching and mismatching target sequences. This quality makes LNA(TM) ideal for SNP (Single Nucleotide Polymorphism) detection, particularly within the context of gene families. The term "Locked Nucleic Acids" was coined by Exiqon to emphasize that the usual conformational freedom of the furanose ring in standard nucleosides is restricted in LNA(TM) due to a methylene linker connecting the 2'-O position to the 4'-C position. This restriction accounts for the enhanced specificity.
Terms of the open-ended supply agreement were not disclosed.
Ahhhh, but I digress...I take it you got that from the IPDL at the WIPO site. All I found was that single page. Is there more? If not, do you have a guess as to when we'll have access to the entire text? Espacenet doesn't even have this much right now.
Oh, and just wanted to mention that this is a prime example of work behind the scenes that had been going on long before we learned of it. It was just July 10, 2002 that we learned of the Penn State alliance and Shriver consulting agreement, yet this patent was filed in BOTH of their names in August 2002. Things don't normally happen QUITE that fast...LOL
Just goes to show that as hard as we look, and as much as we dig up, we still only know parts of the overall picture.
Watching and waiting...the details should prove interesting.
Later,
W2P
P.S. Does this qualify as the company having been mentioned along side of Shriver as a contributor to his research? Are these Paul McKeigue's markers, or ours? Just curious...I get confused you know.
Oops...it's been awhile!
http://www.alexa.com/data/details/traffic_details?q=&url=www.csigizmos.com/
Later,
W2P
Not according to Alexa, at least...but keep an eye on this one..lol
Later,
W2P
Coming to a Theatre Near You:
http://www.csigizmos.com/
Question: What does this have to do with DNAPrint? Answer: Everything! Keep an Eye on this one...110,000 hits per week
Later,
W2P
My Favorite PR from 2002:
DNAPrint Files Patent to Protect 2,425 SNPs Linked to Drug Response
SARASOTA, Fla., Aug. 20 /PRNewswire-FirstCall/ -- DNAPrint genomics, Inc. announced today that it has filed a patent application to protect 2,425 Single Nucleotide Polymorphisms (SNPs) useful for predicting response to a large number of commonly prescribed drugs.
DNAPrint has previously compiled a candidate gene database of several thousand validated drug metabolism and drug target gene SNPs -- collectively known as the "PHENOME" SNP database. However, the new SNPs claimed in the present patent application were identified from a different, more systematic screen of the entire human genome. The sequence of each is useful for explaining variation in drug response to differing extents, depending on the drug, and the Company's data suggests that most are likely linked to genes in the human genome previously not known to be of pharmacological relevance. As such, DNAPrint believes it is the first to claim markers of this type and elucidate the potential of this new subset of the variable human genome as specifically relevant for predicting drug response.
The new patent application could provide DNAPrint a tremendous advantage towards developing pharmacogenomics classifiers that are specific, sensitive and predictively powerful enough for routine clinical use. "Most of these SNPs have been ignored by the genomics community. Though others may have unwittingly and indirectly linked a very small fraction of these SNPs with variable drug response, technical and conceptual considerations have evidently prevented them from yet recognizing the underlying fundamentals of these associations," said Tony Frudakis, Ph.D., DNAPrint's CEO. "Such recognition would be required to generate a competitive patent application." Indeed, a review of the journal and patent literature reveals no reports describing the phenomena that the Company believes cause the linkages.
The discovery that led to the patent application was unexpected, but part of a deliberate and systematic genetic research strategy at DNAPrint. "That we could identify linkages such as this en masse, within the confines of a relatively small research budget, is indicative of the intelligence and objectivity of our systematic study design and algorithmic approach," said Dr. Kondragunta Venkateswarlu, Vice-President of DNAPrint genomics, Inc.
To maintain its competitive advantage, the company will refrain from presenting details of the discovery until the findings are published by the US Patent and Trademark Office.
And at last, that day draws near...
Waiting and watching,
W2P
DougS...Not new. They started around the time Genelex did.
Later,
W2P
Manti...GNET seems to come along once in a while and is only there for a short time. My impression is that they are there to trade for their client(s) and when they complete their order they disappear again.
I saw VERT there for the first time this morning (maybe they were there yesterday as well, but I didn't notice them). If that's the same VERT as the web link I provided (and I couldn't find another one), well, let's just say I find it interesting.
VERT is not there to make a market in the security, like NITE does. They are an "O" Market Participant Type. The "O" stands for "Order Entry Firm".
Order Entry Firms and their relationship to Market Makers is explained here:
http://www.nasdaqtrader.com/trader/tradingservices/productservices/productdescriptions/acesdescripti...
dmceng, it's possible that they are taking a speculative position. If you look through the portfolio of companies listed on their website, you'll see several "development stage" companies. Their stated investment strategy is to place their bets early in the game, based on their perception of the future value of a company's technology. I'd like to think that's why they're here at this time, of course, that is just my own speculation and is presented for discussion purposes only.
Later,
W2P
Miss Scarlet...Scroll down to the bottom of the page:
http://www.otcbb.com/asp/mp_quotes.asp?Quotes=DNAP
Later,
W2P
cosmic...Interesting. Did you happen to see my post on RB this AM concerning VERT?
http://ragingbull.lycos.com/mboard/boards.cgi?board=DNAP&read=291378
I've been wondering for some time what GNET is up to. BTW, it looked like a buy to me. If not, it was a sell at the Ask IMO.
Later,
W2P
mingwan...At least he is placing you in good company...LOL His biggest problem seems to be his envy over the following you enjoy both here and on the RB Board. This "frog" is quite "green" with envy, it appears.
Of course, what he has tried to pass off as "facts" are for the most part gross mistatements of the agreement itself. He also shows that he has very little knowledge of Beckman Coulter or their business, and really has no clue concerning the impact of this change on DNAPrint or the company's opinion one way or the other.
Rather than correct him (which would take a great deal of effort), I'll simply concur with you that his "facts" could use a lot of work, and respectfully suggest that if he were truly concerned with integrity, he would correct the mistatements himself.
Have a great evening,
W2P
mingwan0...WOW, VERY nice...Thanks eom
gunnabeoneday...This stuff was all written by lawyers and bureaucrats! LOL It is confusing by design.
You have the key phrase, i.e., the definition of "Beneficial Ownership". What they're saying is that any shares that either WILL BE or COULD BE acquired within 60 days of the completion of this registration must be counted as "beneficially owned" for this filing. Not necessarily that they are owned now or absolutely WILL be owned. But by definition they must count all the shares that "could" be owned, AS owned, for the purposes of a proper S-2 filing.
The 60 days mentioned in the filing is after the effective date of the registration. The registration is what DNAP and La Jolla expect to take 4 months. And no, La Jolla won't receive any shares until (I believe) 30 days after the registration is completed. Although, DNAPrint should get the other $250K as soon as the registration itself is complete.
At least that's the way I interpreted it when I read it. Admittedly, it's been several weeks now. Someone else please jump in here and clarify if you've got a different understanding.
Hope this helps,
W2P
gunnabeoneday...That is an incorrect interpretation of the table in the S-2. What has been reported is the maximum holdings possible within the first 60 days of the offering.
It does NOT represent current holdings, if any.
Later,
W2P
Chris...No, I don't think it is. Too many with tax implications to shorten the last trading day of the year...
Later,
W2P