FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2016754384> ?p ?o ?g. }

• W2016754384 endingPage "332" @default.
• W2016754384 startingPage "327" @default.
• W2016754384 abstract "The Journal of Hepatology has a long history of publishing studies on genetically complex diseases. Two recent studies have addressed the genetic basis of primary sclerosing cholangitis (PSC) [1Yang X. Cullen S.N. Li J.H. Chapman R.W. Jewell D.P. Susceptibility to primary sclerosing cholangitis is associated with polymorphisms of intercellular ashesion molecule-1.J Hepatol. 2004; 40: 375-379Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar, 2Wiencke K. Louka A.S. Spurkland A. Vatn M. the IBSEN study group Schrumpf E. Boberg K.M. Association of matrix metalloproteinase-1 and -3 promoter polymorphisms with clinical subsets of Norwegian primary sclerosing cholangitis patients.J Hepatol. 2004; 41: 209-214Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar]. This Editorial examines the subject of complex disease genetics, using these two examples to illustrate the wider issues facing editors, reviewers, readers and investigators alike, and aims to provide general guidance for all interested parties. In genetics ‘complex traits’ are defined as those where the mode of inheritance is unknown, that is; where a Mendelian autosomal dominant, autosomal recessive or sex-linked pattern inheritance attributable to a single gene locus does not apply. Previously complex traits were called ‘polygenic’ (involving more than one gene), multifactorial (involving the interaction of the host genome and one or more environmental factors) or oligogenic (where individual mutations in several different genes lead to the same clinical syndrome, but each patient or family with the disease may possess a single disease causing mutation only). Over 90% of all human diseases can be placed under this heading including autoimmune and viral liver disease. Perhaps the best definition of a complex trait is: where ‘one or more genes acting alone or in concert increase or reduce the risk of that trait’ [[3]Haines J.L. Pericak-Vance M.A. Overview of mapping common and genetically complex disease genes.in: Haines J.L. Pericak-Vance M.A. Approaches to gene mapping in complex diseases. John Wiley and Sons, New York1998: 1-16Google Scholar]. This definition allows for all of the possibilities (oligogenic, polygenic and multifactorial) and includes two essential details. First, genetic variations (mutations or polymorphisms) result in differences in the risk of disease, the disease causing mutations (DCMs) do not by themselves confer disease. Second, it uses the term ‘trait’ as opposed to ‘disease’. This is important where the ‘disease’ under consideration is clinically heterogeneous. A ‘trait’ may refer to either the disease itself, a group of related diseases (or syndrome) or to a particular clinical sub-group (phenotype) within a disease or syndrome. Thus, using the term ‘trait’ allows for the prospect that DCMs may not only determine which diseases individuals develop, but also the clinical severity of the resulting disease. Complex traits represent a diverse spectrum of human diseases. The heritable component of complex traits varies. Some, especially those which we categorise as ‘oligogenic’ (for example, Alzheimer's disease), behave more like Mendelian diseases than complex traits. These diseases have large heritable components, consequently multiplex families are common. Measuring the heritable component of such diseases is relatively easy. Comparisons of the disease incidence in monozygotic (identical) and dizygotic (non-identical) twins provides a useful indicator of the balance between genes and environment in disease genesis. Whereas, comparing the disease incidence in siblings (or other family members) enables investigators to determine the sibling (offspring or first-degree relative) relative risk—a statistic referred to as λ (λs—for sibling relative risk). At the other end of the spectrum, many complex traits have a very small heritable component, with few of the hallmarks of genetic diseases. Here measuring the heritable component of the disease may be much more difficult. Currently, the autoimmune liver diseases autoimmune hepatitis and primary sclerosing cholangitis fall into this latter category. However, the ‘absence of evidence’ is not quite the same as ‘evidence of absence’ and these data have yet to be established for AIH and PSC. Contrast this situation with that of the sister disease PBC, where the heritable component is quite marked and quantifiable, for example, the λs value for PBC is 10.4 [[4]Jones D.E. Watt F.E. Metcalf J.V. Bassendine M.F. James O.F.W. Familial primary biliary cirrhosis reassessed: a geographically-based population study.J Hepatol. 1999; 30: 402-407Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar]. The absence of large numbers of multiplex families, in autoimmune and viral liver disease, creates a second problem for investigators. Traditionally disease genes have been identified and mapped by ‘linkage analysis’ which has become the gold standard for gene mapping. When such families are not available for analysis, as in AIH, PSC and also PBC, genetic ‘mapping’ may be performed by association analysis based, on either simplex families as in transmission disequilibrium testing (TDT), or upon collection of unrelated patients and healthy controls as in case-control association studies. The advantage of intra-familial association studies (TDT) is they negate the need for matched controls, because the controls in TDT analysis are the unaffected siblings of the index case. This also resolves the potential problem of population stratification which can arise when the patients are drawn from a different ancestral pool to the controls [[5]Cardon L.R. Palmer L.J. Population stratification and spurious allelic association.Lancet. 2003; 361: 598-604Abstract Full Text Full Text PDF PubMed Scopus (990) Google Scholar]. However, TDT requires large numbers of simplex families and is best suited to investigations of diseases where the majority of cases present as children. TDT does not suit studies of medium-to-late-onset disease. Practically therefore most genetic studies in autoimmune liver disease have been (and will continue to be) based on case-control association studies. Case-control studies have received a mixed coverage in the scientific and medical press [6Colhoun H. McKeigue P.M. Smith G.D. Problems of reporting genetic associations with complex outcomes.Lancet. 2003; 361: 865-872Abstract Full Text Full Text PDF PubMed Scopus (1011) Google Scholar, 7Cardon L.R. Bell J.I. Association study designs for complex disease.Nat Rev. 2001; 2: 91-99Crossref Scopus (1173) Google Scholar]. Case control association analysis has the advantage of taking into account sporadic (non-familial) cases and can, provided the study groups are large enough, produce very powerful results. It is also the only practical option for the study of genes in diseases with a small heritable component and also in infectious diseases where transmission is horizontal rather than vertical. The major disadvantage of this approach compared with linkage analysis is that an association does not necessarily imply a primary relationship between the polymorphism and disease, but may be due to linkage disequilibrium between the polymorphism and DCM elsewhere in the same gene or a neighbouring gene. Despite this much of our current knowledge of the genetic basis of complex disease comes from association analysis and increasingly, large scale, high-throughput case-control studies are seen as one of the most viable and informative options for studying even relatively common complex diseases like insulin dependent diabetes mellitus (IDDM) and autoimmune thyroid disease. However, evaluation of such studies can be very difficult for editors, reviewers and readers alike. Guidelines for association studies have been published [8Todd J.A. Tackling common disease.Nature. 2001; 411: 537-539Crossref PubMed Scopus (36) Google Scholar, 9AnonymousFreely associating.Nat Genet. 1999; 22: 1-2Crossref PubMed Scopus (333) Google Scholar] and the strengths and weaknesses of these studies reviewed extensively [6Colhoun H. McKeigue P.M. Smith G.D. Problems of reporting genetic associations with complex outcomes.Lancet. 2003; 361: 865-872Abstract Full Text Full Text PDF PubMed Scopus (1011) Google Scholar, 7Cardon L.R. Bell J.I. Association study designs for complex disease.Nat Rev. 2001; 2: 91-99Crossref Scopus (1173) Google Scholar, 8Todd J.A. Tackling common disease.Nature. 2001; 411: 537-539Crossref PubMed Scopus (36) Google Scholar, 9AnonymousFreely associating.Nat Genet. 1999; 22: 1-2Crossref PubMed Scopus (333) Google Scholar]. The principle considerations are candidate selection and statistical power. The best advice suggests that candidates should be ‘biologically plausible’ and the polymorphisms studied should be ‘functional’. Plausibility and function are related issues. We should consider two other factors in study design; [a] most genes have multiple polymorphisms and these are inherited as blocks or haplotypes and [b] biological pathways are often redundant. Biological plausibility goes right to the heart of the planning process. The extent of polymorphism in the human genome is staggering. In the post genome era we have the information and bioinformatics tools to consider almost any of the 33,000 human genes. However, we do not yet know the function of all of them. Even for those genes where the biological function is known it is often difficult to relate that knowledge to the pathogenesis of a particular disease. The best clues often come from related diseases or syndromes. So it is that Yang et al., [[1]Yang X. Cullen S.N. Li J.H. Chapman R.W. Jewell D.P. Susceptibility to primary sclerosing cholangitis is associated with polymorphisms of intercellular ashesion molecule-1.J Hepatol. 2004; 40: 375-379Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar] chose to investigate the ICAM1 gene in PSC. The polymorphism they studied has been linked with susceptibility to a number of other inflammatory diseases, including inflammatory bowel disease and Bechet's disease. The plausibility of ICAM1 as a candidate being strengthened by the observation of ICAM (intra-cellular adhesion molecules) on bile ductules and interlobular bile ducts in advanced PSC, and by finding of increased levels of soluble ICAM in serum of PSC patients. In the second study by Wiencke et al., [[2]Wiencke K. Louka A.S. Spurkland A. Vatn M. the IBSEN study group Schrumpf E. Boberg K.M. Association of matrix metalloproteinase-1 and -3 promoter polymorphisms with clinical subsets of Norwegian primary sclerosing cholangitis patients.J Hepatol. 2004; 41: 209-214Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar] the candidates are the MMP1 and MMP3 genes both located on chromosome 11q. The matrix metalloproteins (MMP's) they encode are important in extra cellular matrix (ECM) metabolism and regulation and hence may be implicated in fibrogenesis, which is the key process in the pathogenesis of PSC. Yet, there are many MMPs and the inter-relation between MMPs and their tissue inhibitors (tissue inhibitors of metalloproteinases or TIMPs) is both complex and redundant. Redundancy results in partial deficiencies in biological pathways being compensated by the activity of other pathway components. In addition, many pathways are characterised by interacting networks, receptor–ligand interaction is often complicated by the existence of duplicate receptors with different consequences in terms of cell signalling. For example, antagonists and decoy-receptors exist in the cytokine pathways. As a result, the base-line hypotheses of both Yang et al., and Wiencke et al., [1Yang X. Cullen S.N. Li J.H. Chapman R.W. Jewell D.P. Susceptibility to primary sclerosing cholangitis is associated with polymorphisms of intercellular ashesion molecule-1.J Hepatol. 2004; 40: 375-379Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar, 2Wiencke K. Louka A.S. Spurkland A. Vatn M. the IBSEN study group Schrumpf E. Boberg K.M. Association of matrix metalloproteinase-1 and -3 promoter polymorphisms with clinical subsets of Norwegian primary sclerosing cholangitis patients.J Hepatol. 2004; 41: 209-214Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar] which are commonly used by all who investigate complex disease, i.e. that a genetic deficit in a key protein predisposes to increased disease risk, are at best over simplified. Furthermore, in both of the above studies one could make a reasonable case for investigating almost any candidate gene from either the immunoregulatory or ECM pathways, respectively. Therefore as reviewers and readers how strict should we be in considering ‘biological plausibility’ of these candidates? One extreme suggestion, given the extent of polymorphism in the human genome, is that the only way to identify the major genetic determinants of a complex disease is to perform case-control studies of single nucleotide polymorphisms (SNPs) scanning the entire genome. This ‘non-hypothesis constrained’ approach sets aside the issues of biological plausibility and functionality in favour of mass coverage of the genome. Just as the whole genome linkage scans of the 1990's used microsatellites to identify ‘linkage regions’, this approach would identify areas where there are significant genetic associations for further analysis. The SNPs tested would not be selected based on their function, but simply upon position in the genome. Later, rounds of fine mapping in each ‘associated region’ would be used to identify the specific candidate genes and DCMs and at this stage biological plausibility and function would be investigated. In practice, only a minority of SNPs in the genome have been tested for functional correlates and functionality has not been a primary consideration in most studies of autoimmune and viral liver disease. Interestingly, both of the recent studies in PSC selected SNPs for which there is data indicating a functional relationship; the first examined SNPs in exon 4 (leading to a glycine for arginine exchange) and exon 6 (leading to a lysine for glutamic acid exchange) of ICAM1, the second considered insertion/deletion polymorphisms in the promoters of the MMP1 and MMP3 genes. Finally, neither of these studies can be considered to have fully explored these three selected genes. Most genes encode multiple polymorphisms. Currently, the best strategy for investigating a gene is to identify the informative SNPs and investigate SNP-haplotypes using these informative SNPs as tags—so called ‘haplotype tagging’ [[10]Johnson G.C.L. Esposito L. Barratt B.J. Smith A.N. Heward J. Genova G.D. et al.Haplotype tagging for the identification of common disease genes.Nat Genet. 2001; 29: 233-237Crossref PubMed Scopus (1025) Google Scholar]. There are two reasons why this approach should be considered. First, unless all of the informative polymorphisms in a specific candidate are tested one cannot rule out the possibility of an association with an untested SNP. This is most important where the tested SNPs are negative for genetic association with the disease. Second, the polypeptides that genes encode and their levels of expression relate to the sum total of the haplotype sequence and subtle effects may be missed if the hypothesis is predicated on a single SNP only. To date few studies have used this approach. However, we cannot consider these issues in isolation. In determining the validity of a study and its relevance we must also consider the statistical analysis. The quality of any case-control association study is directly proportional to the numbers studied. Statistical power is a measure of the ability of a study to confidently detect an association. The strength of an association is measured in terms of the risk ratio (generally authors use the odds ratio OR) and assessed in terms of the confidence interval (CI) quantified as the probability value (p). When authors report a positive association the major issue is whether the observation is a ‘true’ or ‘false’ positive. The probability ‘p’ is a measure of our confidence in the ‘truth’ of the observation. Generally, we accept ‘p’ values <0.05 as significant (5% level). However, it is common for case-control association studies to first report significant associations and then for all subsequent investigators to refute the findings of the original studies. This has lead to a call for more stringent confidence intervals to be set, perhaps even as low as 10−5 [[11]Dahlman I. Eaves I.A. Kosoy R. Morrison V.A. Heward J. Gough S.C. et al.Parameters for reliable results in genetic association studies in common disease.Nat Genet. 2002; 30: 149-150Crossref PubMed Scopus (201) Google Scholar]. Whilst this may seem extreme it would protect against ‘false positive’ observations. The most pressing problem in complex disease genetics has become how to assess negative (null) data. This is where the issue of statistical power comes into play. The statistical power of a study is proportional to the numbers studied, the frequency of the candidate polymorphism in the healthy population and the size of effect that is being measured. In complex diseases genes play a lesser role than they do in Mendelian disease, and the size of effect can be expected to be comparatively smaller for diseases with a small heritable component versus those with a large heritable component. Studies indicate that some candidate genes have large effects—reflecting the importance of their biological role, whereas others have relatively small effects. This latter difference may also reflect the influence of sets or blocks of genes inherited as conserved haplotypes for example: the extended HLA A1-B8-DR3 (8.1) haplotype on chromosome 6p, or the cytokine haplotype on chromosome 5q. The reasons for the greater influence of specific HLA haplotypes for example in autoimmune liver disease [12Donaldson P.T. Recent advances in clinical practice: genetics of liver disease: immunogenetics and disease pathogenesis.Gut. 2004; 53: 599-608Crossref PubMed Scopus (133) Google Scholar, 13Donaldson P.T. Norris S. Evaluation of the role of MHC class II alleles, haplotypes and selected amino acid sequences in primary sclerosing cholangitis.Autoimmunity. 2002; 35: 555-564Crossref PubMed Scopus (77) Google Scholar], versus other immunoregulatory genes (for example, CTLA4 in AIH) lies either with the dominant role played by key MHC alleles in the immune response and/or with the potential for each haplotype to encode multiple susceptibility alleles which may have an additive contribution to disease risk. In the context of the current debate it is essential to recognise this difference and to plan studies, accordingly. In order to detect genes which have a relatively small influence on disease risk we need much larger sample sizes than for genes with large effects. The question which always arises when reviewing or reading negative studies is—was the study adequately powered? (i.e. was the sample size large enough to detect a small effect and to what level?). The trick for authors to grasp is that their hypothesis should include some estimate of the size of effect they are aiming to detect, and the manuscript should include a calculation of the statistical power. In practice this is often omitted (as with Yang et al., [[1]Yang X. Cullen S.N. Li J.H. Chapman R.W. Jewell D.P. Susceptibility to primary sclerosing cholangitis is associated with polymorphisms of intercellular ashesion molecule-1.J Hepatol. 2004; 40: 375-379Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar]) or the calculation performed post hoc (as with Wiencke et al., [[2]Wiencke K. Louka A.S. Spurkland A. Vatn M. the IBSEN study group Schrumpf E. Boberg K.M. Association of matrix metalloproteinase-1 and -3 promoter polymorphisms with clinical subsets of Norwegian primary sclerosing cholangitis patients.J Hepatol. 2004; 41: 209-214Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar]). Both of the studies discussed here include positive as well as negative findings but, how valid are these findings? Positive results first: Yang et al., [[1]Yang X. Cullen S.N. Li J.H. Chapman R.W. Jewell D.P. Susceptibility to primary sclerosing cholangitis is associated with polymorphisms of intercellular ashesion molecule-1.J Hepatol. 2004; 40: 375-379Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar] studied 104 PSC patients and 213 healthy controls. They found a positive association with the ICAM1 E469K SNP in PSC patients, whereby the patients were less likely to be homozygous for the E variant (EE) compared to healthy controls (12% in PSC versus 24% in controls, OR=0.41). There was also a strong protective effect associated with homozygous carriage of the G241-K469 haplotype in PSC though no individual association with any particular R241G genotype was found. Overall, this indicates that E469 homozgosity protects from PSC and, based on an odds ratio of 0.41, E469 homozgotes are 2.44 times less likely to develop PSC than patients with other ICAM1 genotypes. The probability that these not true positive findings, but may be due to chance is set at 0.02. In the study by Wiencke et al., [[2]Wiencke K. Louka A.S. Spurkland A. Vatn M. the IBSEN study group Schrumpf E. Boberg K.M. Association of matrix metalloproteinase-1 and -3 promoter polymorphisms with clinical subsets of Norwegian primary sclerosing cholangitis patients.J Hepatol. 2004; 41: 209-214Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar] there was no overall association with either of the tested SNPs and PSC compared to healthy controls, but there were two associations with clinical subgroups. The major observation was an increase in the MMP3 5A/5A status in PSC patients with concurrent ulcerative colitis (UC) versus those without UC (38% versus 21%, OR=2.4). The probability that this does not represent a true positive genetic association, but is simply a random observation occurring by chance is set at 0.03. The second observation was that all of the PSC patients with cholangiocarcinoma carried the MMP1−1G allele (versus 70% of healthy controls), but it was based on only 15 patients. This is too few, as the authors themselves realise, to get excited about yet, but the observation has great potential. In both of the studies correction for multiple testing (Bonferroni's correction) has been applied. The need to correct for multiple testing has been debated by many authors on the basis that it is too stringent or not necessary [[14]Perneger T.V. What's wrong with Bonferroni adjustments.BMJ. 1998; 316: 1236-1238Crossref PubMed Scopus (4460) Google Scholar] and leads to rejection of weak positive associations. However, correction is appropriate where multiple sub-groups are tested except in circumstance where either a second (confirmatory) set is being studied or the authors were testing an a priori hypothesis. It may not apply in all cases where multiple questions are being asked of the same patient series. In these two examples, the corrected p values; 0.02 and 0.03, are not striking and fall close to the 0.05 minimum acceptable level of confidence and a minimum level of correction has been applied in both cases. Therefore there is strong risk that both major observations are false positive findings. In the case of Yang et al., [[1]Yang X. Cullen S.N. Li J.H. Chapman R.W. Jewell D.P. Susceptibility to primary sclerosing cholangitis is associated with polymorphisms of intercellular ashesion molecule-1.J Hepatol. 2004; 40: 375-379Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar] there are no prior data to validate the authors findings, and replication in a second set would have been advisable before publication. Whereas, in the case of Wiencke et al., [[2]Wiencke K. Louka A.S. Spurkland A. Vatn M. the IBSEN study group Schrumpf E. Boberg K.M. Association of matrix metalloproteinase-1 and -3 promoter polymorphisms with clinical subsets of Norwegian primary sclerosing cholangitis patients.J Hepatol. 2004; 41: 209-214Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar] there are prior published data for MMP3, but not for MMP1 though the MMP3 data only partly agree with those in the original report. Why is there a difference between the findings of these two studies? The original report [[15]Satsangi J. Chapman R.W.G. Haldar N. Donaldson P. Mitchell S. Simmons J. et al.A functional polymorphism of the stromelysin gene (MMP-3) influences susceptibility to primary sclerosing cholangitis.Gastroenterology. 2001; 121: 124-130Abstract Full Text PDF PubMed Scopus (86) Google Scholar] was based on a total of 111 PSC patients and 171 healthy controls, versus 165 patients and 346 controls in the recent Norwegian study [[2]Wiencke K. Louka A.S. Spurkland A. Vatn M. the IBSEN study group Schrumpf E. Boberg K.M. Association of matrix metalloproteinase-1 and -3 promoter polymorphisms with clinical subsets of Norwegian primary sclerosing cholangitis patients.J Hepatol. 2004; 41: 209-214Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar]. The first report found the MMP3-5A allele to be increased in the PSC patients versus controls, whist the second found this association to be restricted to those PSC patients with concurrent UC only. The idea that PSC+UC may have a different genetic background to PSC−UC is not without precedent [[13]Donaldson P.T. Norris S. Evaluation of the role of MHC class II alleles, haplotypes and selected amino acid sequences in primary sclerosing cholangitis.Autoimmunity. 2002; 35: 555-564Crossref PubMed Scopus (77) Google Scholar]. The Norwegian study has marginally more statistical power than the study of Satsangi et al., [[15]Satsangi J. Chapman R.W.G. Haldar N. Donaldson P. Mitchell S. Simmons J. et al.A functional polymorphism of the stromelysin gene (MMP-3) influences susceptibility to primary sclerosing cholangitis.Gastroenterology. 2001; 121: 124-130Abstract Full Text PDF PubMed Scopus (86) Google Scholar], but this does not account for these differences. Close scrutiny of the subgroup composition in the two studies may however provide the explanation. In the study of Satsangi et al., [[15]Satsangi J. Chapman R.W.G. Haldar N. Donaldson P. Mitchell S. Simmons J. et al.A functional polymorphism of the stromelysin gene (MMP-3) influences susceptibility to primary sclerosing cholangitis.Gastroenterology. 2001; 121: 124-130Abstract Full Text PDF PubMed Scopus (86) Google Scholar], the majority of the patients (92/111 or 83%) had concurrent IBD compared with only 56% of those studied by Wiencke et al., [[2]Wiencke K. Louka A.S. Spurkland A. Vatn M. the IBSEN study group Schrumpf E. Boberg K.M. Association of matrix metalloproteinase-1 and -3 promoter polymorphisms with clinical subsets of Norwegian primary sclerosing cholangitis patients.J Hepatol. 2004; 41: 209-214Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar]. Therefore, perhaps it is not surprising that the original study did not differentiate between PSC with and without UC in describing the influence of the MMP3-5A allele on disease susceptibility. There were simply too few patients with PSC alone in the study for the analysis of these subgroups to be statistically valid. The reported lack of a difference in Satsangi's report may be a type II error and it underpins the importance of statistical power even in subgroups. In general, when the findings of studies based on different populations vary they pose the question—should we always expect studies in different populations to concur? The answer to this question depends on the degree of racial or ethnic separation or the degree of (historical) geographic isolation of the two tested populations. For example, we would not expect the UK and Norwegian populations to vary as much as the UK and Japan. In this respect, recent studies of the CARD15 gene in Crohn's disease tell an interesting tale. Out of the three major CARD15 SNP linked with Crohn's disease, the major SNP responsible for disease susceptibility differs between Jewish and non-Jewish populations [16Zhou Z. Lin X.Y. Akolkar P.N. Gulwani-Akolkar B. Levine J. Katz S. et al.Variation at NOD2/CARD15 in familial and sporadic cases of Crohn's disease in Ashkenazi Jewish population.Am J Gastroenterol. 2002; 97: 3095-3101Crossref PubMed Google Scholar, 17Cuthbert A.P. Fisher S.A. Mirza M.M. King K. Hampe J. Croucher P.J. et al.The contribution of NOD2 gene mutations to the risk and site of inflammatory bowel disease.Gastroenterology. 2002; 122: 867-874Abstract Full Text Full Text PDF PubMed Scopus (623) Google Scholar], and a Jewish- specific CARD15 SNP (JW-1), has recently been detected [[18]Sugimura K. Taylor K.D. Lin Y.C. Hang T. Wang D. Tang X.M. et al.A novel NOD2/CARD15 haplotype conferring risk for Crohn disease in Ashkenazi Jews.Am J Human Genet. 2003; 72: 509-518Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar]. Similar differences exist between MHC-encoded susceptibility to type 1 AIH in Brazilian versus North American and European white patients [[19]Donaldson P.T. Genetics in autoimmune hepatitis.Semin Liver Dis. 2002; 22: 353-363Crossref PubMed Scopus (95) Google Scholar]. The major difficulties in assessing studies of genes in complex disease, relate to negative data. In our two examples; Yang et al., [[1]Yang X. Cullen S.N. Li J.H. Chapman R.W. Jewell D.P. Susceptibility to primary sclerosing cholangitis is associated with polymorphisms of intercellular ashesion molecule-1.J Hepatol. 2004; 40: 375-379Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar] did not find an association with the ICAM1 R241G SNP and Wiencke et al., [[2]Wiencke K. Louka A.S. Spurkland A. Vatn M. the IBSEN study group Schru" @default.
• W2016754384 created "2016-06-24" @default.
• W2016754384 creator A5042932407 @default.
• W2016754384 date "2004-08-01" @default.
• W2016754384 modified "2023-09-25" @default.
• W2016754384 title "Genetics of autoimmune and viral liver diseases; understanding the issues" @default.
• W2016754384 cites W1519521580 @default.
• W2016754384 cites W1522239197 @default.
• W2016754384 cites W1583476286 @default.
• W2016754384 cites W1639453419 @default.
• W2016754384 cites W1967649993 @default.
• W2016754384 cites W1969767037 @default.
• W2016754384 cites W1976943968 @default.
• W2016754384 cites W1997154421 @default.
• W2016754384 cites W2012546852 @default.
• W2016754384 cites W2026881541 @default.
• W2016754384 cites W2035126508 @default.
• W2016754384 cites W2040814562 @default.
• W2016754384 cites W2044480628 @default.
• W2016754384 cites W2082797463 @default.
• W2016754384 cites W2101199845 @default.
• W2016754384 cites W2106631668 @default.
• W2016754384 cites W2112227290 @default.
• W2016754384 cites W2120483845 @default.
• W2016754384 cites W2141214289 @default.
• W2016754384 cites W2155399146 @default.
• W2016754384 cites W2164017889 @default.
• W2016754384 doi "https://doi.org/10.1016/j.jhep.2004.06.001" @default.
• W2016754384 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/15288484" @default.
• W2016754384 hasPublicationYear "2004" @default.
• W2016754384 type Work @default.
• W2016754384 sameAs 2016754384 @default.
• W2016754384 citedByCount "31" @default.
• W2016754384 countsByYear W20167543842012 @default.
• W2016754384 countsByYear W20167543842013 @default.
• W2016754384 countsByYear W20167543842015 @default.
• W2016754384 countsByYear W20167543842016 @default.
• W2016754384 countsByYear W20167543842022 @default.
• W2016754384 crossrefType "journal-article" @default.
• W2016754384 hasAuthorship W2016754384A5042932407 @default.
• W2016754384 hasConcept C159047783 @default.
• W2016754384 hasConcept C203014093 @default.
• W2016754384 hasConcept C2776029263 @default.
• W2016754384 hasConcept C2777203481 @default.
• W2016754384 hasConcept C54355233 @default.
• W2016754384 hasConcept C70721500 @default.
• W2016754384 hasConcept C71924100 @default.
• W2016754384 hasConcept C86803240 @default.
• W2016754384 hasConceptScore W2016754384C159047783 @default.
• W2016754384 hasConceptScore W2016754384C203014093 @default.
• W2016754384 hasConceptScore W2016754384C2776029263 @default.
• W2016754384 hasConceptScore W2016754384C2777203481 @default.
• W2016754384 hasConceptScore W2016754384C54355233 @default.
• W2016754384 hasConceptScore W2016754384C70721500 @default.
• W2016754384 hasConceptScore W2016754384C71924100 @default.
• W2016754384 hasConceptScore W2016754384C86803240 @default.
• W2016754384 hasIssue "2" @default.
• W2016754384 hasLocation W20167543841 @default.
• W2016754384 hasLocation W20167543842 @default.
• W2016754384 hasOpenAccess W2016754384 @default.
• W2016754384 hasPrimaryLocation W20167543841 @default.
• W2016754384 hasRelatedWork W1991523530 @default.
• W2016754384 hasRelatedWork W2002128513 @default.
• W2016754384 hasRelatedWork W2020824267 @default.
• W2016754384 hasRelatedWork W2031436818 @default.
• W2016754384 hasRelatedWork W2057739827 @default.
• W2016754384 hasRelatedWork W2075354549 @default.
• W2016754384 hasRelatedWork W31773153 @default.
• W2016754384 hasRelatedWork W4231664593 @default.
• W2016754384 hasRelatedWork W4248404839 @default.
• W2016754384 hasRelatedWork W2092874662 @default.
• W2016754384 hasVolume "41" @default.
• W2016754384 isParatext "false" @default.
• W2016754384 isRetracted "false" @default.
• W2016754384 magId "2016754384" @default.
• W2016754384 workType "article" @default.