Aligned rank transform tests in linear models

Non-Additive Effects of Environmental Factors on Growth and Physiology of Invasive Solidago canadensis and a Co-Occurring Native Species (Artemisia argyi)

Changes in environmental factors, such as temperature and UV, have significant impacts on the growth and development of both native and invasive plant species. However, few studies examine the combined effects of warming and enhanced UV on plant growth and performance in invasive species. Here, we investigated single and combined effects of warming and UV radiation on growth, leaf functional and photosynthesis traits, and nutrient content (i.e., total organic carbon, nitrogen and phosphorous) of invasive Solidago canadensis and its co-occurring native species, Artemisia argyi, when grown in culture racks in the greenhouse. The species were grown in monoculture and together in a mixed community, with and without warming, and with and without increased UV in a full factorial design. We found that growth in S. canadensis and A. argyi were inhibited and more affected by warming than UV-B radiation. Additionally, there were both antagonistic and synergistic interactions between warming and UV-B on growth and performance in both species. Overall, our results suggested that S. canadensis was more tolerant to elevated temperatures and high UV radiation compared to the native species. Therefore, substantial increases in temperature and UV-B may favour invasive S. canadensis over native A. argyi. Research focusing on the effects of a wider range of temperatures and UV levels is required to improve our understanding of the responses of these two species to greater environmental variability and the impacts of climate change.

Measuring short distance dispersal of Alliaria petiolata and determining potential long distance dispersal mechanisms

Introduction

Alliaria petiolata, an herbaceous plant, has invaded woodlands in North America. Its ecology has been thoroughly studied, but an overlooked aspect of its biology is seed dispersal distances and mechanisms. We measured seed dispersal distances in the field and tested if epizoochory is a potential mechanism for long-distance seed dispersal.

Methods

Dispersal distances were measured by placing seed traps in a sector design around three seed point sources, which consisted of 15 second-year plants transplanted within a 0.25 m radius circle. Traps were placed at intervals ranging from 0.25–3.25 m from the point source. Traps remained in the field until a majority of seeds were dispersed. Eight probability density functions were fitted to seed trap counts via maximum likelihood. Epizoochory was tested as a potential seed dispersal mechanism for A. petiolata through a combination of field and laboratory experiments. To test if small mammals transport A. petiolata seeds in their fur, experimental blocks were placed around dense A. petiolata patches. Each block contained a mammal inclusion treatment (MIT) and control. The MIT consisted of a wood-frame (31 × 61× 31 cm) covered in wire mesh, except for the two 31 × 31 cm ends, placed over a germination tray filled with potting soil. A pan filled with bait was placed in the center of the tray. The control frame (11 × 31 × 61 cm) was placed over a germination tray and completely covered in wire mesh to exclude animal activity. Treatments were in the field for peak seed dispersal. In March, trays were moved to a greenhouse and A. petiolata seedlings were counted and then compared between treatments. To determine if A. petiolata seeds attach to raccoon (Procyon lotor) and white-tailed deer (Odocoileus virginianus) fur, wet and dry seeds were dropped onto wet and dry fur. Furs were rotated 180 degrees and the seeds that remained attached were counted. To measure seed retention, seeds were dropped on furs and rotated as before, then the furs were agitated for one hour. The seeds retained in the fur were counted.

Results

For the seed dispersal experiment, the 2Dt function provided the best fit and was the most biologically meaningful. It predicted that seed density rapidly declined with distance from the point source. Mean dispersal distance was 0.52 m and 95% of seeds dispersed within 1.14 m. The epizoochory field experiment showed increased mammal activity and A. petiolata seedlings in germination trays of the MIT compared to control. Laboratory studies showed 3–26% of seeds were attached and retained by raccoon and deer fur. Retention significantly increased if either seed or fur were wet (57–98%).

Discussion

Without animal seed vectors, most seeds fall within a short distance of the seed source; however, long distance dispersal may be accomplished by epizoochory. Our data are consistent with A. petiolata’s widespread distribution and development of dense clusters of the species in invaded areas.

Nonparametric tests for differential gene expression and interaction effects in multi-factorial microarray experiments

Background

Numerous nonparametric approaches have been proposed in literature to detect differential gene expression in the setting of two user-defined groups. However, there is a lack of nonparametric procedures to analyze microarray data with multiple factors attributing to the gene expression. Furthermore, incorporating interaction effects in the analysis of microarray data has long been of great interest to biological scientists, little of which has been investigated in the nonparametric framework.

Results

In this paper, we propose a set of nonparametric tests to detect treatment effects, clinical covariate effects, and interaction effects for multifactorial microarray data. When the distribution of expression data is skewed or heavy-tailed, the rank tests are substantially more powerful than the competing parametric F tests. On the other hand, in the case of light or medium-tailed distributions, the rank tests appear to be marginally less powerful than the parametric competitors.

Conclusion

The proposed rank tests enable us to detect differential gene expression and establish interaction effects for microarray data with various non-normally distributed expression measurements across genome. In the presence of outliers, they are advantageous alternative approaches to the existing parametric F tests due to the robustness feature.