Interpreting the results of observational research: chance is not such a fine thing.

In a randomised controlled trial, if the design is not flawed, different outcomes in the study groups must be due to the intervention itself or to chance imbalances between the groups. Because of this tests of statistical significance are used to assess the validity of results from randomised studies. Most published papers in medical research, however, describe observational studies which do not include randomised intervention. This paper argues that the continuing application of tests of significance to such non-randomised investigations is inappropriate. It draws a distinction between bias and chance imbalance on the one hand (both randomised and observational studies can be affected) and confounding on the other (a unique problem for observational investigations). It concludes that neither the P value nor the 95% confidence interval should be used as evidence for the validity of an observational result.     

Resistance gene analogues in sugarcane and sorghum and their association with quantitative trait loci for rust resistance.

Fifty-four different sugarcane resistance gene analogue (RGA) sequences were isolated, characterized, and used to identify molecular markers linked to major disease-resistance loci in sugarcane. Ten RGAs were identified from a sugarcane stem expressed sequence tag (EST) library; the remaining 44 were isolated from sugarcane stem, leaf, and root tissue using primers designed to conserved RGA motifs. The map location of 31 of the RGAs was determined in sugarcane and compared with the location of quantitative trait loci (QTL) for brown rust resistance. After 2 years of phenotyping, 3 RGAs were shown to generate markers that were significantly associated with resistance to this disease. To assist in the understanding of the complex genetic structure of sugarcane, 17 of the 31 RGAs were also mapped in sorghum. Comparative mapping between sugarcane and sorghum revealed syntenic localization of several RGA clusters. The 3 brown rust associated RGAs were shown to map to the same linkage group (LG) in sorghum with 2 mapping to one region and the third to a region previously shown to contain a major rust-resistance QTL in sorghum. These results illustrate the value of using RGAs for the identification of markers linked to disease resistance loci and the value of simultaneous mapping in sugarcane and sorghum.     

Investigation of the het genes that control heterokaryon incompatibility between members of heterokaryon-compatibility (h-c) groups A and G1 of Aspergillus nidulans

A chromosome assay method was used to determine the heterokaryon compatibility relationships between strains belonging to heterokaryon-compatibility (h-c) groups A and G1 of Aspergillus nidulans. A hybrid strain (RD15) was isolated following protoplast fusion of strains 65-5 (h-cA) and 7-141 (h-cG1). The morphology of RD15 was severely abnormal compared to diploid strains of A. nidulans produced from heterokaryon-compatible haploid parents. Inocula of RD15 were induced to haploidize on medium containing Benlate and a parasexual progeny sample of 291 haploid segregants was obtained. The progeny strains were genotyped for standard markers. Allelic ratios and pairwise marker segregations were determined. Pairs of progeny strains that carried different alleles for the standard markers on each linkage group in turn were tested for compatibility. Strain pairs that possessed different alleles for the markers on linkage groups II, III, V, VI and VII were incompatible indicating the presence of heterokaryon-incompatible (het) genes on these linkage groups. Backcrosses to an h-cGl strain showed that two het genes were located on linkage group III and confirmed a total of six het gene differences between the h-cA and h-cGl strains.     

Dispersal between and colonization of apple byMetaseiulus occidentalis andTyphlodromus pyri (Acarina: Phytoseiidae)

Adult femaleMetaseiulus occidentalis Nesbitt immigrated faster and at higher rates than didTyphlodromus pyri Scheuten to mini-orchards of young apple trees at 0, 10 and 100 m downwind from four apple orchards in the Hood River Valley, Oregon. Colonization of mini-orchards also was faster byM. occidentalis. Rates of emigration were similar between species as measured by movement into trees placed within, but not in contact with source orchard trees. Dispersal and colonization may be influenced by intrinsic physiological and morphological attributes, food habits, and responses to prey and habitat requirements. Implications of dispersal and colonization by these mites to biological control and pesticide resistance management are discussed.     

Evolution of pesticide resistance in apple pests and their natural enemies

Evolution of pesticide resistance in 24 apple pest and natural enemy species was simulated with a computer model. Population ecology parameters were varied among species while physiological, biochemical and genetic assumptions were held constant. There was good agreement between the model's predictions and observed historical patterns of azinphosmethyl resistance among pests and natural enemies. Correspondence between predicted and observed was improved by assuming that natural enemies evolved resistance only after their prey/hosts became resistant, but not by assuming greater initial susceptibility in natural enemies. Results suggest that ecological factors may be important in determining rates of resistance evolution. This is paper no. 2843 of the Hawaii Institute of Tropical Agriculture & Human Resources journal series, and no. 7245 of the Oreg. Agric. Exp. Sta. journal series.     

Predicting population trends using citizen science data: do subsampling methods produce reliable estimates for mammals?

Accurate assessment of population trends is invaluable in wildlife management, particularly for identifying species which are of conservation concern, and consequently, reliable cost-effective methods for their determination are highly desirable. In a recent publication (Eur J Wildl Res 62:407–413, 2016), the authors apply a subsampling method, used in several studies to quantify population trends from citizen science data for butterflies, birds, and plants, to assess the status of West European hedgehogs (Erinaceus europaeus) in England. Whilst the findings may be in agreement with expert opinion, we argue that this type of approach does not adequately account for spatial bias common in mammal data and that without further evaluation it is unclear whether the result is reliable or simply coincidental. To explore this concern, we apply the method across a range of terrestrial mammal species and compare the resulting trends to other published studies. Our findings show that the method fails to reproduce the accepted qualitative trends for the majority of species. Furthermore, comparison of trends based on data obtained from different sources produced conflicting predictions suggesting that the method is indeed vulnerable to survey bias. We therefore conclude that at present, without additional modification to address survey bias, this is not a reliable method for predicting population trends for mammals. However, more generally, this raises questions about the validity of subsampling methods based on citizen science data, and we would urge future studies to exercise caution by performing analysis across a suite of species including those with known trends for validation.     

Molecular polymorphism and phenotypic variation in Aspergillus carbonarius

Thirteen collection strains and field isolates of Aspergillus carbonarius were examined by using various genotypic and phenotypic approaches. Restriction fragment length polymorphism analysis of the ribosomal RNA gene cluster and the mitochondrial DNA of the strains revealed only slight variations, except for one field isolate (IN7), which exhibited completely different ribosomal RNA gene cluster and mitochondrial DNA patterns. The mitochondrial DNAs of these strains were found to be much larger (45 to 57 kb) than those found earlier in the A. niger aggregate. Strain-specific characters could be detected by the random amplified polymorphic DNA technique. Isoenzyme analysis and examination of carbon source utilisation patterns of the strains also revealed some intraspecific variability, though much smaller than that observed by using DNA-based techniques. The dendrograms constructed based on genotypic and phenotypic data suggest that strain IN7 might represent a new subspecies of A. carbonarius.Abbreviations kb kilobase pair - mtDNA mitochondrial DNA - RAPD random amplified polymorphic DNA - rDNA ribosomal RNA gene cluster - RFLP restriction fragment length polymorphisms