Full Article

The ‘Junk DNA’ non-coding patents have provoked unusual interest for reasons including;

1. the widespread impression that patents’ owner Australian Biotechnology Company Genetic Technologies Limited (GTG) interprets the patent claims to cover any test involving amplification of non-coding DNA;

4. reactions to the scope of the patents encompassing all eukaryotic organisms (how can a company patent DNA in all species?); and to whether society should permit such perceived monopolies.

The Intron Diagnostic patent (filed August 1989) describes a method for practice of my discovery of the general utility of non-coding DNA sequence variation for Linkage Disequilibrium (LD)-based surrogate gene allele/mutation typing in single individuals (without the need for pedigree information)(*2).

The Genome Mapping Patent (filed July 1990) describes a method for population-based (case-control) gene fine mapping by genome-wide LD haplotype identification in individuals with a trait, compared with individuals lacking the trait (*3). The method is in three parts:

(2) testing of polymorphic sites in a plurality of patients and comparison subjects;

(3) analysis of ‘Haplotype Heterogeneity Restriction’, a term I gave to the concept that trait bearers would have restricted diversity of haplotypes in regions containing genes of interest. [A more common phrase for this goal of all current single nucleotide polymorphism (SNP) and short tandem repeat (STR)-based LD mapping is ‘Excess Haplotype Sharing’].

It has become obvious to me that the disbelief in, and reactions to, the patents mainly arise from imprecision in the wording of the patent claims, and from a failure to clearly distinguish my discoveries and inventions from prior art. In particular, the use in the Intron Diagnostic patent of the word ‘linkage’, and the absence of the terms ‘linkage disequilibrium’ and ‘haplotype’, directs attention towards more than a decade of prior art, the latter four years employing polymerase chain reaction (PCR). An interpretation that non-coding DNA sequences occurring within genes (introns) are in genetic ‘linkage’ with the genetic locus is non-sense to many skilled in genetic art. Thus, confusion and controversy is understandable.

The invention on which both families of ‘Junk DNA’ patents is based is the utility of the polymorphic information content of any amplifiable non-coding sequence to sufficiently mark coding locus alleles/mutations and haplotypes by linkage disequilibrium (LD) in single individuals. I discovered that non-coding sequence variation was sufficiently structured according to coding locus allele and multi-locus haplotypes to substitute for, and to dispense with, the linkage requirement for pedigree chromosome transmission.

Prior to patent filings in 1989/90, genetic science mainly involved the study of family members for identification of linked loci by chromosome transmission. Southern blotting, using cDNA probes for detection of restriction fragment length polymorphisms (RFLP), was the main diagnostic and gene discovery DNA technology. RFLP site variations were recognized to be useful for gene mapping and for mutation diagnosis by pedigree linkage analysis from the time of publications in 1978 by Kan and Dozy (1,2), and in 1980 by Botstein et al. (3) and by Little et al. (4). Some restriction enzyme sites, located up to tens of kilobases from the gene, exhibited LD with the disease gene locus, and enabled the assignment of ‘haplotypes’ (5-9).

From end-1985, with the description of sequence amplification by polymerase chain reaction (PCR) (10), RFLP analysis by Southern blotting began to be replaced by PCR. Beginning around 1987 through to the early 1990’s, the obvious application of PCR for amplification of RFLP sites to simplify linkage analysis was reported (11-14). With the advent of PCR, research focus shifted rapidly and near-exclusively to exon amplification for detection of coding mutations and alleles. Non-coding DNA was viewed as irrelevant to the main matter of studying coding sequence responsible for translated gene products.

The 10^th International Histocompatibility Workshop (10IHW), culminating in 1987, applied Southern blotting for RFLP study of HLA loci distributed over 3 megabases in the major histocompatibility complex (MHC - 6p21.1-3). Now that RFLP has been largely superseded as a DNA technology, this RFLP study, involving 13 loci, 12 restriction enzymes, and 107 human cell lines, is likely to remain the largest ever performed.

My analysis in 1988/89 of the Workshop RFLP DNA fragment patterns revealed that RFLP sites were non-randomly ordered according to coding HLA locus alleles, and to combinations of alleles as multi-locus haplotypes (15,16). Since some 95% of genomic DNA was known to be non-coding, the great majority of the sequence variations responsible for the RFLP patterns were likely to be located in non-coding regions.

I found that RFLP allele and haplotype associations were conserved between peoples of the same HLA type but of different ethnic backgrounds. Some haplotypes extended over hundreds of kilobases. The existence of extended haplotypes occurring as supratypic HLA multi-gene blocks had been known since at least 1983 (17) so, contrary to a current view, haplotype block structure in the genome is not a recent discovery.

In 1989 this extent of the non-random ordering of non-coding polymorphisms as sequence information reflecting coding locus allele and multi-locus haplotypic block patterns was not known. As late as 1997, when sequencing HLA introns, Blasczyk and associates wrote: “Against all expectations, this (highly polymorphic variability of introns) is not characterised by random point mutations but by a highly systematic diversity reflecting the ancestral relationship of the HLA alleles” (18).

In 1986-88 I was working in collaboration with Dr. Henry Erlich at Cetus Corporation, Emeryville, CA. Dr. Erlich was an inventor of patents for application of RFLP to HLA typing, and a co-inventor of PCR with Dr. Mullis and others. More than anyone else in the world, Dr. Erlich was in a position to imagine that application of PCR for analysis of non-coding sequence polymorphism revealed by RFLP might be useful in diagnosis and gene mapping (19). The Intron Diagnostic patent prosecution history records that Dr. Erlich actually taught away from the concept, taking the accepted view of the time that there was no value in analysis of any DNA sequence other than coding exons.

I performed the 10IHW HLA RFLP analyses under the guidance of Dr. Jean-Marc Lalouel at the Howard Hughes Institute in Salt Lake City, a pre-eminent world center for gene discovery by RFLP-based family linkage analysis. Over the 9 months that I spent with Dr. Lalouel, no one imagined that the noncoding polymorphic patterns my analyses were revealing would mark haplotypes sufficient to dispense with the requirement for families in gene mapping.

There was no prior art (*4) known to me that described a method for utilizing this polymorphic information by sequence analysis of any non-coding DNA sufficient for LD-based surrogate coding allele / mutation typing, and for haplotyping, in single individuals such as those comprising the subjects of the HLA RFLP Workshop.

One study that had a prospect of anticipating my claimed discovery concerned immunoglobulin genes. In the early 1980s, Nottenburg, St John and Weissman discovered that SNPs in the JH region of murine immunoglobulin genes occurred at a frequency of approx. 1/200 bases, and correlated with Ig alleles. In a publication in 1987 (20) they used one of the SNPs, detected by DNA sequence analysis, to identify the allele from which clones of VDJ rearrangements were derived. Nottenberg, now a patent attorney who has written an analysis of the ‘Junk DNA’ patents (21: see this web site), acknowledges that she did not imagine the extent of non-coding polymorphic information content, the structure of the polymorphic patterns, or the utility of SNPs for allele and haplotype identification.

The earliest reference involving population LD mapping, for better gene localization of a disease-associated gene that had been chromosome region-assigned by linkage, was published in 1992 (22,23). The first report of LD mapping for genome-wide searches, without prior knowledge of chromosome gene location, appeared two years later, in 1994 (24). The patent cited as prior art the 1989 publication of discovery of the gene for Cystic Fibrosis by final-stage localization of pedigree-assigned haplotypes (25).

Suggestions that the general utility of LD-based noncoding polymorphisms was ‘obvious’ in 1989 need to consider that it was 7 years from the discovery of PCR before the first application to LD mapping was reported, and 9 years before application to genome-wide mapping.

Short tandem repeats (STRs – 26,27) and methods for large-scale DNA analysis (28) had been described, but gave no indication of prior imagination to apply STRs to LD-based gene typing or mapping, including in publications in the early 1990’s, subsequent to patent filing in 1990.

With increasing use of STRs and SNPs in the past 15 years, it is understandable that it is difficult for some to now imagine that LD-based gene typing fine-scale mapping was not so obvious in 1989.

Previous art had involved extensive searches for informative RFLPs. My discovery was that it was not necessary to undertake further searches. Analysis of the sequence of small (amplifiable) non-coding regions was all that was required. Any region would suffice.

My expectation that the non-random, allele and haplotypic structure of non-coding sequence polymorphism would be a universal characteristic of all eukaryotic organisms, rather than being unique to HLA genes, was supported by declarations to the US Patent Office from experts in blood genetics, mouse genomics, and soybean plant genetics.

While non-coding DNA may be ‘junk’ with respect to protein coding, the conservation of noncoding sequences over thousands of years in humans, and over tens of millions of years between humans and other animals, and in plants, must have reflected important, then as now largely unknown, functions. Perhaps the first review of some functional possibilities for introns appeared in 1994 (29).

Patent Legal Considerations

Patent legal aspects will be detailed elsewhere by Nottenberg (21). Here I will comment only on issues of novelty / anticipation, and of inventive step / non-obviousness. The terms differ in different countries, as does the law. Concerning novelty (anticipation), the prior art information must disclose all of the essential features of an invention. In some jurisdictions, patent rejection requires identification of a single reference that discloses to one of ordinary skill in the art, either explicitly or by implication, all elements of a particular patent claim. In others, disclosures of components in two or more documents are considered, but only if the connection between the documents would be viewed by a person skilled in the art as a single source of information.

Patent rejection on grounds of lack of inventive step, or obviousness, is considered when two or more references are proposed that, in combination, disclose or suggest the modification in the prior art that is required for the invention. There has to be a motivation to combine the references to achieve all elements of a patent other than from the existence of the patent. In other words, the combined prior art must disclose or suggest the modification in the prior art process that is required for the invention, without reference to the patent. Failure to fulfill this requirement is deemed to be “impermissible hindsight”. Many proposed articles are in this category (see, for example, duplicative references 30,31).

Other considerations include whether the publications are enabling, by providing detailed methodology for practicing the claimed invention, and whether one skilled in the art would expect a modification to be successful.

I will remain appreciative of receiving any candidate prior scientific publications that might challenge the patents for lack of novelty, or for lack of inventive step. Literature that I have examined is listed following the references to this article. To date, I am unaware of any prior art that warrants disclaiming the inventions.

Scientific experimentation and patent claims are separate matters in that an inventor may claim a scope of the invention that extends beyond that of the performed scientific experiments. The objective of the Inventor’s Attorney is to compose the claims to achieve the widest possible scope, while avoiding interpretations that encompass prior art. Interpretation of the scope of claims includes recognition of method usages that are not explicitly excluded. Thus, interpretation of the patent claims for evaluation of proposed infringement is a complex matter. Patent language and style is difficult to understand without specific training. Complex phrasing and wording (in part because a claim must be written as a single sentence), variable usage of Definition terms, unclear distinction from prior art, and unclear and possibly inaccurate statements in the Specification (text) of a patent provide for different interpretations. In the event of dispute, claim meaning will be adjudicated by a court of law.

Concerning definitions in the Intron Diagnostic patent, six terms are of special relevance:

(2) locus: refers to a coding gene locus, from 5’untranslated upstream to 3’untranslated downstream;

LD. The population genetic concept of LD was well understood to refer to polymorphisms at RFLP sites and alleles at coding loci exhibiting co-occurrence frequencies that varied from equilibrium expectation. However, the term is a misnomer because allele associations having frequencies significantly different from expected can occur between unlinked loci. Linkage is a within-family phenomenon. LD is a population phenomenon. LD refers to non-random association of alleles at different loci in the gametes of the population. Thus, the term gametic (phase) disequilibrium is more appropriate. Allele association has come into more recent use.

Haplotype. The term ‘haplotype’, a contraction of haploid genotype, was coined by HLA geneticist Ruggero Ceppellini in 1967 to apply to haploid (single chromosome) combinations of HLA coding locus alleles (32). This use of ‘haplotype’ is incorporated in the definition in the Patents. In addition, the unit of inheritance is not an entire chromosome, but combinations of each of the two parental chromosomes. Each single chromosome is a mosaic of the two parental chromosomes. The Patent definition incorporates this fact of inheritance that a single haploid element is the length of chromosome between sites of crossing over (recombination) between one parental chromosome and the other.

The Patent definition of Haplotype is: “As used herein, “haplotype” is a region of genomic DNA on a chromosome which is bounded by recombination sites such that genetic [coding] loci within a haplotypic region are usually inherited as a unit. However, occasionally genetic rearrangements may occur within a haplotype. Thus, the term haplotype is an operational term that refers to the occurrence on a chromosome of linked [coding] loci”. (NB – my insertion of [coding] is to qualify the term ‘locus’ according to the Patent’s definition). Each of the coding loci within the haplotype are represented as a single allele, so a haplotype is taken to mean the combination of alleles at two or more linked loci on a single chromosome.

Subsequently, non-HLA geneticists have adopted the term ‘haplotype’ to apply to combinations of SNPs at two or more linked single nucleotide polymorphic loci, including non-coding loci. By this usage, the shortest ‘haplotype’ is any pairwise combination of two adjacent nucleotides at contiguous polymorphic loci. If the two single nucleotide polymorphic (SNP) loci are separated by a distance that allows amplification of the sequence length then the amplified sequence of any of the four combinations of the di-nucleotide loci has also been termed a ‘haplotype’. Ruano and Kidd (33) employ the terms 'locus' and 'haplotype' in these ways when they termed the amplicons generated by primers at two non-coding polymorphic (single nucleotide) loci as ‘haplotypes’. Multiple SNPs occurring in non-coding DNA, particularly in intergenic sequences, are now employed for genome-wide gene discovery by LD (Allele association) mapping, involving the assignment of ‘haplotypes’. Definitions of ‘locus’ and ‘haplotype’ in the patents differ from these usages.

When two or more SNPs are identified at a coding locus, it has become common practice to term combinations of those SNPs, inferred or observed, as 'haplotypes'. Sequence variations at coding loci define alleles. Not all alleles at a coding locus are specified by a single SNP (i.e. an allele-specific SNP). More usually, as best exemplified by HLA genes, alleles are defined by unique combinations of SNPs distributed over coding elements of the locus. Maximum statistical power for detection of gene-trait associations can be expected to require assignment of coding locus variation as locus alleles. Thus, for investigation of coding loci for their role in disease / function, priority should be given to defining the locus alleles present in a population, rather than to group two or more SNPs together as so-called ‘haplotypes’.

The terms ‘linkage disequilibrium’ and ‘haplotype’ do not occur in the Intron Diagnostic patent claims. Instead, the independent claims speak of ‘linkage’, and of ‘alleles’. The term ‘linkage’ is correctly used. However, as mentioned earlier, an interpretation that non-coding DNA sequences occurring within genes (introns) are in genetic ‘linkage’ with the genetic locus for the gene being considered is non-sense to many skilled in genetic art. Since linkage is lost by recombination, it is useful to ask: “Can recombination occur between the two entities being regarded as in linkage?” If not, to describe two entities as exhibiting ‘linkage’ lacks meaning.

An informative non-coding sequence polymorphism and the coding locus having the allele are in linkage if they are co-inherited. Linkage applies to loci, not to alleles. The patented invention concerns the phenomenon of LD association between a non-coding polymorphism and an allele of a linked coding locus, not linkage between non-coding polymorphic sites and coding loci. Since gene mutation / carrier state diagnosis and genome-wide gene mapping by linkage was established practice long before 1989, it is understandable that the patents have prompted incredulity.

The situation is further complicated by the definition of LD as “..the co-occurrence of two alleles at linked loci such that the frequency of the co-occurrence of the alleles is greater than would be expected from the separate frequencies of occurrence of each allele”. Since ‘allele’ is previously defined as “a genetic variation associated with a coding region”, and ‘locus’ refers to a coding locus, then it would not be possible to use the term ‘LD’ to apply to the association between a non-coding polymorphism and a coding locus allele.

Understanding the scope and limitations of patent claims is a complex process. Nonetheless, these complexities notwithstanding, key phrases in the independent claims, interpreted using the definitions specified in the patent, apply to:

(1) amplification by an intron-spanning primer pair of a non-coding sequence that is in linkage with an allele associated with a coding locus;

(2) the sequence variability of the amplified non-coding sequence being sufficient to assign at least one allele at the adjacent coding locus.

firstly: a non-coding region-spanning amplified sequence, which is in linkage with and has a sequence which is characteristic of,

The intention of the Intron Diagnostic patent, to describe those situations where a non-coding sequence variant is in LD with a coding locus allele such that the former is characteristic of the latter, is clearly illustrated in the proof of principle experiment of HLA Class II typing in the 4^th Asia-Oceania Histocompatibility Workshop (4AOHW). Non-coding SNPs were shown to surrogately mark HLA-DQA1 coding locus alleles (as assigned at 1994) (34,35), and to define HLA-DR/DQ haplotypes (36). In genetic disease an example involves the CFTR gene, mutations in which are responsible for Cystic Fibrosis. An variant of the 4-base GATT repeat in intron 6 is in absolute LD with the commonest coding mutation delta F508 (37). Similarly, in intron 8 of the CFTR gene there is a polythymidine (polyT) tract sequence associated with the acceptor splice site (IVS8-polyT) preceding exon 9. The (polyT) tract is a tri-variant polymorphic site, occurring as 5T, 7T and 9T variants. The 9T variant of IVS8-polyT is, like the GATT repeat variant, also in absolute LD with the delta F508 mutation in Caucasians (38), while the 7T repeat occurs with delta F508 in some 2/3rds of Arab CF patients (39). Testing for the GATT repeat or polyT intron motifs as a surrogate for the delta F508 coding mutation is taught by the patent.

The Patent applies to those situations in which a non-coding amplicon is analysed to provide information on a coding allele, or on a combination of coding locus alleles, as haplotypes.

Aware that HLA coding alleles were being assigned by direct sequence analysis, a major use of the patent was expected to be in identifying and assigning HLA haplotypes as haploid combinations of multiple locus alleles. Haplotypes could only be marked by non-coding sequence polymorphisms since variants in coding sequences defined alleles. This utility of the patent for non-coding markers required extensive sequence information on family-confirmed haplotypes, and undertaking which is only recently begun. It can be expected that current research on SNP-haplotype characterization will enable the realization of this strategy, first described in 1993 (36).

Before 1989 and since, some biotechnology companies have considered using patented inventions for monopolistic control of coding exon allele/mutation typing. An intent of the Intron Diagnostic patent was that non-coding typing would provide a surrogate alternative in such commercial circumstances. Specific examples include HLA typing, mutation testing at the Hfe locus for haemochromatosis diagnosis, and BRCA1 locus mutation analysis for breast cancer risk. However, the requirement for a surrogate testing system based on non-coding polymorphisms has not eventuated.

The Patent was not intended to apply, and in my opinion does not apply, to amplification of non-coding sequence variants for information inherent to the amplicon sequence. These include promoter variants, non-coding sites serving as alternate splice mutants, or other non-coding sites, which have been deemed to be disease-causing mutations but where the mechanism remains to be revealed.

A specific instance of such non-coding sites is in the CFTR gene. Three intron sites, all involved in alternative splicing, can be tested in commercial kits, such as those vendored by Applied Biosystems, Foster City, CA). The unusual nucleotide, or nucleotide (polyT) tract, at these sites predisposes to alternative splicing of the RNA transcript. When this results either in ‘skipping’ of one or more exons, or in shortening (‘truncating’) of the normal gene product by interruption of transcription, or by the inclusion of intron sequence within the transcript due to a mutation in the usual splice site, then the resulting protein may be non-functional. The site responsible for the faulty protein is regarded as a ‘mutation’. One of the three non-coding intron sites is a polyT tract, in which the shortest (5T) variant predisposes to exon skipping by alternative splicing. The test is performed to assess the presence or absence of the 5T variant ‘mutation’. The 5T variant is itself a mutation, not a marker of a mutation. Similar examples have been described for the beta-globin (40) and phenylalanine hydroxylase (PAH) gene loci (41). These circumstances of ‘stand alone’ intron sites are not encompassed by the claims because each alternative splice site is not in linkage with the coding locus. Each is within the locus. Each site is not characteristic of a second sequence comprising an allele associated with, and at another location within, a coding locus. Each site is itself the ‘mutation’. In these and similar instances, there is no second element of an allele associated with a coding locus for which the non-coding variant is characteristic.

To my knowledge, all current DNA Diagnostic test performed on non-coding sequences can be considered in two categories:

(1) mutation (alternative splice site) testing: where the site is investigated for its inherent information, as in the CFTR, beta-globin and PAH genes;

(2) marker (haplotype) tagging: where non-coding sequence analysis is used to characterize haplotypes in pedigree analysis, by linkage.

The first category is not taught by the Patent since there is no coding sequence allele in linkage with the site being tested. The non-coding site cannot be in linkage with itself. Furthermore, the PAH gene report was recognized as prior art (*5) in the Intron Diagnostic patent.

The second category was argued in the patent prosecution history to be prior art. The earliest examples are the reports in 1978 by Kan and Dozy (1,2). Amplification of prior-known RFLP sites as an obvious simplification of previous linkage carrier detection was also recognized in the Intron Diagnostic patent as prior art (11,12).

I am unaware of any current DNA molecular laboratory test for disease-associated gene diagnostics, or for HLA typing, that utilizes the method of the Intron Diagnostic patent.

By contrast, all uses of population-based LD / Allele association fine-mapping seem to me to be encompassed by the Genome Mapping patent.

However, patent legal considerations, particularly pertaining to definitions, such as uses of the term ‘haplotype’ that do not involve coding loci, may well be critical in court adjudication of claims interpretation, and of proposed practice of the invention as claimed (infringement) (*5).

Considering patent law rules for ‘anticipation / novelty’ and for ‘obviousness’, no prior art scientific literature has been brought to my attention that warrants disclaiming the inventions.

------------------------------------------------------------------------------------------------------------

(*1) Dr. Simons has no association with GTG, and derives no benefit from any aspect of the Company’s business.

(*2) US Patent No: 5,192,659. Intron Sequence Analysis Method for Detection of Adjacent and Remote Locus Alleles as Haplotypes [Intron Diagnostic patent].

(*3) US Patent No: 5,851,762. Genomic Mapping Method by Direct Haplotyping using Intron Sequence Analysis [Genomic Mapping patent].

(*4) Prior art is a term used in patent law to indicate printed and oral disclosures (e.g. publications, meeting posters and public talks) of subject matter related to the patents, here concerning amplifiable non-coding sequences for surrogate LD-based gene typing and gene discovery.

(*5) Confusion concerning the definition of ‘haplotype’ is similar to that now arising with the term ‘epigenetic’. To contrast with genetics, in 1942 Waddington defined ‘epigenetics’ as the study of the processes by which genotype gives rise to the phenotype (1). As noted by Lederburg, in 1942 there were few ideas about processes, and none about the underlying chemical mechanisms (2). In 1987, Holliday applied ‘epigenetic’ to the changes in gene activity during development (3). Seven years later, in 1994, Holliday provided a supplementary definition of epigenetics to refer to nuclear inheritance which is not based on differences in DNA sequence (4). This definition would be included in the process of development of the phenotype, and so is a sub-definition of the original usage. The concurrent usage of two meanings is likely to give rise to misunderstanding. That there are others (epigenetics concerns those forms of inheritance that do not follow the rules of Mendel; the existence of genetic phenomena beyond the familiar –[5]) can be expected to compound any confusion.

Similarly, when Ceppellini defined ‘haplotype’ in 1967 as combinations of coding locus alleles (6), he would have had no idea that non-coding DNA sequence non-randomness would also reflect coding haplotypes. Here, too, there has now been a shift of usage meaning from Haplotype C (coding; Ceppellini) to Haplotype NC (non-coding). As explained in this article, the patents define ‘haplotype’ as Haplotype C.

Ceppellini R, Curtoni ES, Mattiuz PL, Miggiano V, Scudeller G & Serra A. In:Histocompatibility Testing 1967 (Eds. Curtoni ES, Mattiuz PL, Tosi RM). The Williams and Wilkins Company. Pp. 149-187

1. Kan YW, Dozy AM. Antenatal diagnosis of sickle-cell anaemia by D.N.A. analysis of amniotic-fluid cells. Lancet 1978 Oct. 28; 2(8096): 910-12.

2. Kan YW, Dozy AM. Polymorphism of DNA sequence adjacent to human beta-globin structural gene: relationship to sickle mutation. Proc. Natl. Acad. Sci. USA 1978 Nov; 75(11): 5631-5.

3. Botstein D, White RL, Skolnick M, and Davis RW. Construction of a genetic linkage map in man using restriction fragment length polymorphism. Am J Hum Genet. 1980 May; 43(3): 314-31.

4. Little PFR, Annison G, Darling S, Williamson R. Model for antenatal diagnosis of beta-thalassaemia and other monogenic disorders by molecular analysis of linked DNA polymorphisms. Nature 1980 May 15; 285: 144-47.

5. Rees A, Stocks J, Paul H, Ohuchi Y, Galton D. Haplotypes identified by DNA polymorphisms at the apolipoprotein A-1 and C-III loci and hypertriglyceridaemia. A study in a Japanese population. Hum Genet. 1986 Feb; 72(2): 168-71.

6. Antonarakis SE, Oettgen P, Chakravarti A, Halloran SL, Hudson RR, Feisee L, Karathanasis SK. DNA polymorphism haplotypes of the human apolipoprotein APOOA1-APOC3-APOA4 gene cluster. Hum Genet. 1988 Nov; 80(3): 265-73.

7. Estivill X, Scambler PJ, Wainwright BJ, Hawley K, Frederick P, Schwartz M, Baiget M, Kere J, Williamson R, Farrall M. Patterns of polymorphism and linkage disequilibrium for cystic fibrosis. Genomics 1987 Nov; 1(3): 257-63.

8. Chebloune Y, Pagnier J, Trabuchet G, Faure C, Verdier G, Labie D, Nigon V. Structural analysis of the 5’ flanking region of the beta-globin in African sickle cell anemia patients: further evidence for three origins of the sickle cell mutation in Africa. Proc Natl Acad Sci U.S.A. 1988 Jun; 85(12): 4431-5.

9. Leitersdorf E, Chakravarti A, Hobbs HH. Polymorphic DNA haplotypes at the LDL receptor locus. Am J Hum Genet. 1989 Mar; 44(3): 409-21.

10. Saiki RK., Scharf S., Faloona F., Mullis KB., Horn GT., Erlich HA., Arnheim N. Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science. 1985 Dec. 20; 230(4732): 1350-4.

11. Kogan SC, Doherty M, Gitschier J. An improved method for prenatal diagnosis of genetic diseases by analysis of amplified DNA sequences. New Engl. J. Med. 1987 Oct. 15; 317(16): 985-90.

12. DiLella AG, Huang WM, Woo SL. Screening for Phenylketonuria Mutations by DNA amplification with the Polymerase Chain Reaction. Lancet. 1988 Mar. 5; I(8584): 497-9.

13. McIntosh I, Curtis A, Millan FA, Brock DJ. Prenatal exclusion testing for Huntington disease using the polymerase chain reaction. Am J Med Genet. 1989 Feb; 32(2): 274-6.

14. Graham JB, Kunkel GR, Tennyson GS, Lord ST, Fowlkes DM. The Malmo polymorphism of factor IX: establishing the genotypes by rapid analysis of DNA. Blood 1989 Jun; 73(8): 2104-7.

15. Simons MJ, Wheeler R, Cohen D, Lalouel-JM, Dupont B. Restriction fragment length polymorphism of HLA genes: Summary of the 10^th International Histocompatibility Workshop Southern Blot analysis. In Immunobiology of HLA: Histocompatibility Testing 1987 (Ed. Dupont B). Springer-Verlag, New York. Vol.1, 959-1023.

16. Simons M.J, Erlich H.A. RFLP - sequence interrelations at the DPA and DPB loci. In: Immunobiology of HLA: Histocompatibility testing 1987 (Ed. Dupont B). Springer-Verlag, New York. Vol.1, 953-958.

17. McCluskey J, Kay PH, Dawkins RL, Komori KA, Christiansen FT, McCann VJ. Association of Specific MHC Supratypes with Rheumatoid Arthritis and Insulin-Dependent Diabetes. Disease Markers. 1983; 1: 197-212.

18. Blasczyk R, Kotsch K, Wehling J. The nature of polymorphism of the HLA class I non-coding regions and their contribution to the diversification of HLA. Hereditas. 1997; 127(1-2): 7-9.

19. Erlich HA, Higuchi R, Lichtenwalter K, Reynolds R, Sensabaugh G. Reliability of the HLA-DQ alpha PCR-based oligonucleotide typing system. J Forensic Sci 1990 Sep; 35(5): 1017-19.

20. Nottenberg C, St. John T, Weissman IL. Unusual immunoglobulin DNA sequences from the nonexpressed chromosome of mouse normal B lymphocytes: implications for allelic exclusion and the DNA rearrangement process. J. Immunol. 1987 Sep 1; 139(5): 1718-26.

21. Nottenburg C, Sharples J. Analysis of ‘Junk DNA’ Patents. In preparation.

22. Hastbacka J, de la Chappelle A, Kaitila I, Sistonen P, Weaver A, Lander E. Linkage disequilibrium mapping in isolated founder populations: diastrophic dysplasia in Finland. Nature Genetics 1992; 2: 204-211.

23. Lazzeroni LC. A chronology of fine-scale gene mapping by linkage disequilibrium. Statistical Methods in Medical Research 2001; 10: 57-76.

24. Houwen RH, Baharloo S, Blankenship K, Raemaekers P, Juyn J, Sandkujil LA, Freimer NB. Genome screening by searching for shared segments: mapping a gene for benign recurrent intrahepatic cholestasis. Nat Genet. 1994 Dec.; 8(4); 380-6.

25. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, Buchwald M, Tsui LC. Identification of cystic fibrosis gene: genetic analysis. Science 1989 Sep; 245(4922): 1073-80.

26. Weber JL, May PE. Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction. Am J Hum Genet. 1989 Mar; 44(3): 388-96.

27. Litt M, Luty JA. A hypervariable microsatellite revealed by in vitro amplification of a dinucleotide repeat within the cardiac muscle actin gene. Am J Hum Genet. 1989 Mar; 44(3): 397-401.

28. Skolnick MH, Wallace RB. Simultaneous analysis of multiple polymorphic loci using amplified sequence polymorphisms (ASPs). Genomics. 1988 May; 2(4): 273-9.

29. Mattick JS. Introns: evolution and function. Curr Opin Genet Dev. 1994 Dec; 4(6): 823-31.

30. Tsui LC, Buetow K, Buchwald M. Genetic analysis of cystic fibrosis using linked DNA markers. Am J Hum Genet. 1986 Dec; 39(6): 720-8.

31. Murray JC, Buetow KH, Donovan M, Hirnung S, Motulsky AG, Disteche C, Dyer K, Swisshelm K, Anderson J, Giblett E, et al. Linkage disequilibrium of plasminogen polymorphisms and assignment of the gene to human chromosome 6q26-6q27. Am J Hum Genet. 1987 Apr; 40(4): 338-50.

32. Ceppellini R, Curtoni ES, Mattiuz PL, Miggiano V, Scudeller G, Serra A. Genetics of Leukocyte Antigens: A Family Study of Segregation and Linkage. In: Histocompatibility Testing 1967 (Eds. Curtoni ES, Mattiuz PL, Tosi RM). The Williams and Wilkins Company. 149-187.

33. Ruano G, Kidd KK. Direct Haplotyping of chromosomal segments from multiple heterozygotes via allele-specific PCR amplification. Nucleic Acids Research 1989 Oct. 25; 17(20): 8392.

34. Limm TM, Ashdown ML, Naughton MJ, McGinnis MD, Simons MJ. HLA-DQA1 allele and suballele typing using noncoding sequence polymorphisms. Application to 4AOHW cell panel typing. Human Immunology 1993 Sep; 38(1): 57-68.

35. Amos DB, van Rood, JJ. Editorial. (“An approach pioneered by Malcolm Simons is presented in this issue. ..although simple and inexpensive, it performed perfectly in the preceding fourth AOH workshop”). Human Immunology 1993 Sep; 38(1): 1-2.

36. Simons M.J, Limm T.M, Naughton M.J, Quinn D.L, McGinnis M.D, Ashdown M.L. Strategy for Definition of DR/DQ Haplotypes in the 4AOHW Cell Panel Using Noncoding Sequence Polymorphisms. Human Immunology 1993 Sep; 38(1): 69-74.

37. Gaitskhoki VS, Voronina OV, Potapova Oyu, Kirjukhina LV, Gembitskaya TE, Kapranov NI, Petrova NV, Khafizova ZA, Schwartz EI. Linkage disequilibrium between cystic fibrosis mutations and polymorphic 4-bp repeat within CFTR gene. Biochem. Med. Metab. Biol. 1993 Oct; 50(2): 186-9.

38. Dork T, Fislage R, Neumann T, Wulf B, Tummler B. Exon 9 of the CFTR gene: splice site haplotypes and cystic fibrosis mutations. Hum Genet. 1994 Jan: 93(1): 67-73.

39. Desgeorges M, Megarbane A, Guittard C, Carles S, Loiselet J, Demaille J, Claustres M. Cystic fibrosis in Lebanon: distribution of CFTR mutations among Arab communities. Hum Genet. 1997 Aug: 100(2): 279-83

40. Amselem S, Nunes V, Vidaud M, Estivill X, Wong C, d’Auriol L, Vidaud D, Galibert F, Baiget M, Goossens M. Determination of the spectrum of beta-thalassaemia gene in Spain by use fo the dot-blot analysis of amplified beta-globin DNA. Am J Hum Genet. 1988 Jul; 43(1): 95-100.

41. DiLella AG, Marvit J, Lidsky AS, Guettler F, Woo SLC. Tight linkage between a splicing mutation and a specific DNA haplotype in phenylketonuria. Nature 1986 Aug. 28; 322: 799-803.

---------------------------------------------------------------------------------------------------------------------