The geographic structure of a population, which is modelled as a network of several small random-mating populations or demes exchanging migrants between them, limits the rapid spread of a beneficial allele under strong directional selection to the entire population. This weakens or modifies the hitchhiking effect of the beneficial allele on the pattern of genetic variation at linked neutral loci. Previous studies suggested that the characteristic patterns of polymorphism arise with selective sweeps in such a subdivided population. However, they did not fully address the stochastic pattern, as expected in an actual sample of DNA sequence, of such patterns. This study uses a novel method of individual-based forward-in-time simulation to generate multi-locus neutral polymorphism after a selective sweep in a moderately subdivided population. Population subdivision is shown to cause frequency spectrum to shift slightly such that Tajima's D becomes less negative than expected under a panmictic population. Similarly, the pattern of linkage disequilibrium showed very small change due to population subdivision. On the other hand, the value of Wright's FST at closely linked neutral loci relative to that at unlinked loci greatly increased by population subdivision as predicted by previous studies. Finally, the distribution of the gradient of heterozygosity along the migration path of beneficial mutation, previously suggested to allow the inference of the direction of spread, was investigated. The variance of difference in heterozygosity was much larger than the mean, suggesting that such an inference may not be practical.

Intronic microsatellites repeats were implicated in the pathogenic mechanisms of several diseases. SLC26A4 gene, involved in the genetic susceptibility of autoimmune thyroid disease (AITD), harbours large non-coding introns. Using the tandem repeat finder (TRF) Software, two new polymorphic microsatellite markers, rs59736472 and rs57250751, located at introns 10 and 20, respectively, were identified. A case-control design including 308 patients affected with AITD (146 GD, 90 HT and 72 PIM) and 212 unmatched healthy controls were performed for each marker (rs59736472, D7S2459 and rs57250751). Furthermore, we used PHASE 2.0 version to reconstruct haplotypes, Kolmogorov–Smirnov (KS) and the Clump analysis program for multivariate analysis. The fluorescent genotyping revealed three alleles (106,112 and 115 bp) for rs57250751 and 12 alleles for both D7S2459 and rs59736472 ranging from 134 to 156 bp and from 144 to 168 bp, respectively. The case-control analysis confirmed the positive association of D7S2459 with Hashimoto thyroiditis (HT) disease previously reported. Moreover, a significant association was found only with rs59736472 and HT disease. Haplotype-specific analysis showed that the 140-148-115 haplotype may increase the risk of HT (χ2=9·8, 1 df, P=0·0017, OR=2·07, IC [1·27–3·36]). Consequently, considering the number of repetitions of both D7S2459 and rs59736472, we found 15 alleles ranging from 45 to 59 repetitions. The case-control analysis showed a significant association of the 55 repetition with HT disease (χ2=6·32, 1 df, pc=0·012, OR=1·74, IC [1·1–2·76]). In conclusion, we suggest the association of longer alleles of intron 10 of SLC26A4 gene with HT disease.

In cancer research, high-throughput profiling studies have been extensively conducted, searching for markers associated with prognosis. Owing to the ‘large d, small n’ characteristic, results generated from the analysis of a single dataset can be unsatisfactory. Recent studies have shown that integrative analysis, which simultaneously analyses multiple datasets, can be more effective than single-dataset analysis and classic meta-analysis. In most of existing integrative analysis, the homogeneity model has been assumed, which postulates that different datasets share the same set of markers. Several approaches have been designed to reinforce this assumption. In practice, different datasets may differ in terms of patient selection criteria, profiling techniques, and many other aspects. Such differences may make the homogeneity model too restricted. In this study, we assume the heterogeneity model, under which different datasets are allowed to have different sets of markers. With multiple cancer prognosis datasets, we adopt the accelerated failure time model to describe survival. This model may have the lowest computational cost among popular semiparametric survival models. For marker selection, we adopt a sparse group minimax concave penalty approach. This approach has an intuitive formulation and can be computed using an effective group coordinate descent algorithm. Simulation study shows that it outperforms the existing approaches under both the homogeneity and heterogeneity models. Data analysis further demonstrates the merit of heterogeneity model and proposed approach.

Gene expression profiling of peroxisome-proliferator-activated receptor α (PPARα) has been used in several studies, but there were no consistent results on gene expression patterns involved in PPARα activation in genome-wide due to different sample sizes or platforms. Here, we employed two published microarray datasets both PPARα dependent in mouse liver and applied meta-analysis on them to increase the power of the identification of differentially expressed genes and significantly enriched pathways. As a result, we have improved the concordance in identifying many biological mechanisms involved in PPARα activation. We suggest that our analysis not only leads to more identified genes by combining datasets from different resources together, but also provides some novel hepatic tissue-specific marker genes related to PPARα according to our re-analysis.

In order to formulate the Fundamental Theorem of Natural Selection, Fisher defined the average excess and average effect of a gene substitution. Finding these notions to be somewhat opaque, some authors have recommended reformulating Fisher's ideas in terms of covariance and regression, which are classical concepts of statistics. We argue that Fisher intended his two averages to express a distinction between correlation and causation. On this view, the average effect is a specific weighted average of the actual phenotypic changes that result from physically changing the allelic states of homologous genes. We show that the statistical and causal conceptions of the average effect, perceived as inconsistent by Falconer, can be reconciled if certain relationships between the genotype frequencies and non-additive residuals are conserved. There are certain theory-internal considerations favouring Fisher's original formulation in terms of causality; for example, the frequency-weighted mean of the average effects equaling zero at each locus becomes a derivable consequence rather than an arbitrary constraint. More broadly, Fisher's distinction between correlation and causation is of critical importance to gene-trait mapping studies and the foundations of evolutionary biology.