overlappingCGM: Automatic detection and analysis of overlapping co-expressed gene modules

When it comes to the co-expressed gene module detection, its typical challenges consist of overlap between identified modules and local co-expression in a subset of biological samples. A recent study have reported that the decomposition methods are the most appropriate ones for solving these challenges. In this study, we represent a R tool, termed overlapping co-expressed gene module (overlappingCGM), which possesses those methods with a wholly automatic analysis framework to help non-technical users to easily perform complicated statistical analyses and then gain robust results. We also develop a novel auxiliary statistical approach to select the optimal number of principle components using a permutation procedure. Two example datasets are used, related to human breast cancer and mouse metabolic syndrome, to enable the illustration of the straightforward use of the tool. Computational experiment results show that overlappingCGM outperforms state-of-the-art techniques. The R scripts used in the study, including all information on the tool and its usage are made publicly available at https://github.com/huynguyen250896/overlappingCGM.

Asymptotic permutation tests for coefficients of variation and standardised means in general one-way ANOVA models

Coefficients of variations are unit-free measures that can, for example, be used to compare the variability of different samples. To this end, we study inference methods for them as well as their reciprocal given by standardised means in general heterogeneous one-way ANOVA designs. As no specific model assumptions are made, a permutation method is proposed to guarantee good finite sample performance. Building on recent limit theorems for randomisation techniques, we prove that the permutation procedure is asymptotically correct in general and finitely exact when data is exchangeable. These results are fostered in extensive simulation studies and two illustrative data analyses.

Phylogenetic Diversity of Plant Metacommunity of the Dnieper River Arena Terrace Within the ‘Dnieper-Orilskiy ’ Nature Reserve

This article presents the features of the phylogenetic organization of the plant communities of the Dnieper River terrace within the ‘Dnieper-Orilskiy’ Nature Reserve and the patterns of its spatial variation involving remote sensing data of the Earth’s surface. The research materials were collected in the period 2012−2016 from within the nature reserve. The research polygon is within the first terrace (arena) of the Dnieper valley. Sandy steppe, meadow, forest and marsh communities within the Protoch river floodplain and the Orlova ravine, as well as artificial pine plantations were the habitats present within the research polygon. The vegetation description was carried out on 10×10 m (100 m2) plots. A total of 94 geobotanical descriptions were made. Data on plant phylogeny was obtained by the Phylomatic service. Phylogenetic diversity of the communities was assessed by the Faith, Simpson and Shannon indices. Phylogenetic analysis was performed by means of a double principal coordinate analysis (DPCoA). The vegetation cover within the investigated polygon was represented by 189 species. Abundance Phylogenetic Deviation (APD) for the investigated metacommunity was evaluated to −0.53, which is statistically significantly different from random alternatives (p = 0.001). The APD negative value indicates that phylogenetic organization of the investigated metacommunity is overdispersed. The permutation procedure allowed us to establish that the eigenvalues of the DPCoA-axes obtained as a result of the real phylogenetic tree were significantly higher than their own number for the random phylogenetic trees for the first seven axes. This indicates that the first seven axes are useful for additional information on the ordination structure of the metacommunity.

Evidence of an Epistatic Effect Between Dysbindin-1 and Neuritin-1 Genes on the Risk for Schizophrenia Spectrum Disorders

BackgroundThe interest in studying gene–gene interactions is increasing for psychiatric diseases such as schizophrenia-spectrum disorders (SSD), where multiple genes are involved. Dysbindin-1 (DTNBP1) and Neuritin-1 (NRN1) genes have been previously associated with SSD and both are involved in synaptic plasticity. We aimed to study whether these genes show an epistatic effect on the risk for SSD.MethodsThe sample comprised 388 SSD patients and 397 healthy subjects. Interaction was tested between: (i) three DTNBP1 SNPs (rs2619537, rs2743864, rs1047631) related to changes in gene expression; and (ii) an haplotype in NRN1 previously associated with the risk for SSD (rs645649-rs582262: HAP-risk C-C).ResultsAn interaction between DTNBP1 rs2743864 and NRN1 HAP-risk was detected by using the model based multifactor dimensionality reduction (MB-MDR) approach (P = 0.0049, after permutation procedure), meaning that the risk for SSD is significantly higher in those subjects carrying both the A allele of rs2743864 and the HAP-risk C-C. This interaction was confirmed by using a logistic regression model (P = 0.033, OR (95%CI) = 2.699 (1.08–6.71), R2 = 0.162).DiscussionOur results suggest that DTNBP1 and NRN1 genes show a joint effect on the risk for SSD. Although the precise mechanism underlying this effect is unclear, the fact that these genes have been involved in synaptic maturation, connectivity and glutamate signalling suggests that our findings could be of value as a link to the schizophrenia aetiology.