- Thread starter
- #601
Tracking the alanine-to-valine mutation, the Italian coeliac reference in post #600 is here:
’....causing an alanine311valine variation....linkage disequilibrium between two sites, if not differential fitness is involved, depends mainly on factors that influence the recombination frequency. These factors are, fundamentally, the physical distance between the sites and their ‘age,’ i.e. the number of generations since the most recent variation was introduced. The observation that all four possible haplotypic combinations between all the adjacent variation sites are present suggests that several intragenic recombination events occurred.
Alternatively repeated mutations at the same site can be hypothesized especially considering that two of the polymorphisms (C378T and C956T) involve a CpG dinucleotide, but this mechanism is not favoured by the fact that all variations are dimorphic. Thus, the observed haplotypic combinations were probably generated by recombination events that took place within very short intervals (97-30,000bp). This could be due to ancient origin of the APN variations and/or/ to a particularly high recombination frequency.
Assuming that the APN locus is subject to frequent recombination events, it is likely that a putative undetected causal variation, if it has approximately the same age as the other polymorphisms, underwent the same recombination events and its degree of linkage disequilibrium with the tested variants falls in the range between 0.30 and 0.80 detected in the APN gene. With these D’ values, the probability to detect an association with 200 families is low.
In fact, according to Abel and Myhsok (1998) an association in these conditions could be detected with a power of 80% only by increasing the number of tested families to 400/800 in the favourable hypothesis that both the deleterious and the marker alleles have a frequency of about 0.5.
In conclusion, with the families utilized in the present study, it would have been possible to detect an unknown CD associated variant in the APN gene only if it had a very high D’ (~/=1) with the markers. Therefore we can exclude a direct involvement in CD of all the tested polymorphisms and of any undetected variation which is in complete linkage disequilibrium with those analysed.’
Linkage disequilibrium between intra-locus variants in the aminopeptidase n gene and test of their association with coeliac disease - PubMed
Coeliac disease (CD) is a multigenic and multifactorial enteropathy triggered by gluten-composing proteins. A possible involvement of the intestinal Aminopeptidase N (APN) was investigated by an association analysis. SSCP analysis detected four variants at position 281, 378, 956 and 2957...
pubmed.ncbi.nlm.nih.gov
Alternatively repeated mutations at the same site can be hypothesized especially considering that two of the polymorphisms (C378T and C956T) involve a CpG dinucleotide, but this mechanism is not favoured by the fact that all variations are dimorphic. Thus, the observed haplotypic combinations were probably generated by recombination events that took place within very short intervals (97-30,000bp). This could be due to ancient origin of the APN variations and/or/ to a particularly high recombination frequency.
Assuming that the APN locus is subject to frequent recombination events, it is likely that a putative undetected causal variation, if it has approximately the same age as the other polymorphisms, underwent the same recombination events and its degree of linkage disequilibrium with the tested variants falls in the range between 0.30 and 0.80 detected in the APN gene. With these D’ values, the probability to detect an association with 200 families is low.
In fact, according to Abel and Myhsok (1998) an association in these conditions could be detected with a power of 80% only by increasing the number of tested families to 400/800 in the favourable hypothesis that both the deleterious and the marker alleles have a frequency of about 0.5.
In conclusion, with the families utilized in the present study, it would have been possible to detect an unknown CD associated variant in the APN gene only if it had a very high D’ (~/=1) with the markers. Therefore we can exclude a direct involvement in CD of all the tested polymorphisms and of any undetected variation which is in complete linkage disequilibrium with those analysed.’