Tiling array data analysis: a multiscale approach using wavelets

Background  

Tiling array data is hard to interpret due to noise. The wavelet transformation is a widely used technique in signal processing for elucidating the true signal from noisy data. Consequently, we attempted to denoise representative tiling array datasets for ChIP-chip experiments using wavelets. In doing this, we used specific wavelet basis functions, Coiflets, since their triangular shape closely resembles the expected profiles of true ChIP-chip peaks.     

Local ecological divergence of two closely related stag beetles based on genetic,morphological, and environmental analyses

The process of phenotypic adaptation to the environments is widely recognized. However, comprehensive studies integrating phylogenetic, phenotypic, and ecological approaches to assess this process are scarce. Our study aims to assess whether local adaptation may explain intraspecific differentiation by quantifying multidimensional differences among populations in closely related lucanid species, Platycerus delicatulus and Platycerus kawadai, which are endemic saproxylic beetles in Japan. First, we determined intraspecific analysis units based on nuclear and mitochondrial gene analyses of Platycerus delicatulus and Platycerus kawadai under sympatric and allopatric conditions. Then, we compared differences in morphology and environmental niche between populations (analysis units) within species. We examined the relationship between morphology and environmental niche via geographic distance. P. kawadai was subdivided into the “No introgression” and “Introgression” populations based on mitochondrial COI gene – nuclear ITS region discordance. P. delicatulus was subdivided into “Allopatric” and “Sympatric” populations. Body length differed significantly among the populations of each species. For P. delicatulus, character displacement was suggested. For P. kawadai, the morphological difference was likely caused by geographic distance or genetic divergence rather than environmental differences. The finding showed that the observed mitochondrial–nuclear discordance is likely due to historical mitochondrial introgression following a range of expansion. Our results show that morphological variation among populations of P. delicatulus and P. kawadai reflects an ecological adaptation process based on interspecific interactions, geographic distance, or genetic divergence. Our results will deepen understanding of ecological specialization processes across the distribution and adaptation of species in natural systems.     

Barriers acting on the genetic structure of a highly aerial seabird: The magnificent frigatebird

The sometimes-distant association between dispersal movements and mating strategies may govern seabird ecology. Despite their large dispersal capability, pelagic seabirds are subject to the influence of physical and non-physical barriers that may limit the magnitude of realized effective dispersal and gene flow. The Magnificent Frigatebird (Fregata magnificens) is a widespread, tropical, pelagic, sequentially monogamous seabird capable to breach physical barriers, such as land masses. Nevertheless, Mexican frigatebirds have shown movement preferences by sex (female migration and high male site fidelity), which, along with the complex female mate choice behavior, may promote breeding isolation. In order to ascertain if levels of population structure are influenced by physical or non-physical barriers, we analyzed mitochondrial DNA variation of adult Magnificent Frigatebirds breeding in the main nesting colonies in Mexico, three in the eastern Pacific Ocean and one in the Caribbean Sea. Control region sequences revealed significant genetic structure between Pacific and Caribbean frigatebirds (AMOVA, Φ_SC = 0.105, P < 0.0001), but none among Pacific breeding colonies (non-significant pairwise Φ_ST). Whereas the C-haplotype was shared by Pacific and Caribbean frigatebirds, the T-haplotype was absent in the Caribbean. Genetic differentiation between ocean basins and available evidence of Mexican frigatebird movements suggest that in spite of the capability of frigatebirds to breach physical barriers, movement preferences by sex and mate selection acting in a large inter-oceanic scale are the prime behaviors promoting population structure, which is highly consequential for their conservation.     

Offspring body condition and immunocompetence are negatively affected by high breeding densities in a colonial seabird: a multiscale approach.

Why avian colonies vary in size and how food competition among nearby colonies affects offspring quality are still not completely understood. We simultaneously examined the effects of four scales of breeding density on two measures of offspring viability (body condition and T-cell-mediated immunity) in the colonial Magellanic penguin. Body condition of fledglings was inversely correlated with breeding density within 100 m(2) of nests, and decreased with increasing numbers of breeding pairs competing within the parental foraging ranges (100 km), probably as a result of density-dependent food depletion. The T-cell-mediated immune response was positively correlated with body condition, reflecting, to some extent, the previous breeding-density effects, and was negatively correlated with colony size, which may be related to social stress. However, given the effect of protein intake on cell immunity, this result could also indicate a thus far neglected cost of coloniality, namely the consumption of low-protein food to compensate for the depletion of optimal prey. These results were not influenced by other traits, nor by the current exposure of birds to parasites and diseases, as measured by serological variables. Since body condition and the T-cell-mediated immune response of fledgling birds are indicators of their survival and recruitment prospects, the costs we have identified can explain variability in colony size in relation to food competition with surrounding colonies, as well as the skewed distribution toward small colonies in this species.     

Estimating seed and pollen movement in a monoecious plant: a hierarchical Bayesian approach integrating genetic and ecological data

The scale of seed and pollen movement in plants has a critical influence on population dynamics and interspecific interactions, as well as on their capacity to respond to environmental change through migration or local adaptation. However, dispersal can be challenging to quantify. Here, we present a Bayesian model that integrates genetic and ecological data to simultaneously estimate effective seed and pollen dispersal parameters and the parentage of sampled seedlings. This model is the first developed for monoecious plants that accounts for genotyping error and treats dispersal from within and beyond a plot in a fully consistent manner. The flexible Bayesian framework allows the incorporation of a variety of ecological variables, including individual variation in seed production, as well as multiple sources of uncertainty. We illustrate the method using data from a mixed population of red oak (Quercus rubra, Q. velutina, Q. falcata) in the NC piedmont. For simulated test data sets, the model successfully recovered the simulated dispersal parameters and pedigrees. Pollen dispersal in the example population was extensive, with an average father-mother distance of 178 m. Estimated seed dispersal distances at the piedmont site were substantially longer than previous estimates based on seed-trap data (average 128 m vs. 9.3 m), suggesting that, under some circumstances, oaks may be less dispersal-limited than is commonly thought, with a greater potential for range shifts in response to climate change.     

Inference of hazel grouse population structure using multilocus data: a landscape genetic approach

In conservation and management of species it is important to make inferences about gene flow, dispersal and population structure. In this study, we used 613 georeferenced tissue samples from hazel grouse (Bonasa bonasia) where each individual was genotyped at 12 microsatellite loci to make inference on population genetic structure, gene flow and dispersal in northern Sweden. Observed levels of genetic diversity suggest that Swedish hazel grouse do not suffer loss of genetic diversity compared with other grouse species. We found significant F(IS) (deviation from Hardy-Weinberg expectations) over the entire sample using jack-knifed estimators over loci, which is most likely explained by a Wahlund effect. With the use of spatial autocorrelation methods, we detected significant isolation by distance among individuals. Neighbourhood size was estimated in the order of 62-158 individuals corresponding to a dispersal distance of 950-1500 m. Using a spatial statistical model for landscape genetics to infer the number of populations and the spatial location of genetic discontinuities between these populations we found indications that Swedish hazel grouse are divided into a northern and a southern population. We could not find a sharp border between these two populations and none of the observed borders appeared to coincide with any potential geographical barriers.These results imply that gene flow appears somewhat unrestricted in the boreal taiga forests of northern Sweden and that the two populations of hazel grouse in Sweden may be explained by the post-glacial reinvasion history of the Scandinavian Peninsula.     

Within and between species competition in a seabird community: statistical exploration and modeling of time-series data

In a changing environment, the maintenance of communities is subject to many constraints (phenology, resources, climate, etc.). One such constraint is the relationship between conspecifics and competitors. In mixed colonies, seabirds may have to cope with interspecific and intraspecific competition for both space and food resources. We applied competitive interaction models to data on three seabird breeding populations: black-legged kittiwake (Rissa tridactyla), common guillemot (Uria aalge) and Brünnich's guillemot (Uria lomvia) collected over 27-years at Kharlov Island in the Barents Sea. We found a competitive effect only for the kittiwake breeding population size on the common guillemot breeding population size when kittiwakes were abundant. The timing of kittiwake breeding negatively affected the number of breeding Brünnich's guillemots. The timing of breeding was negatively correlated to biomass of the main pelagic fish in the Barents Sea, the capelin (Mallotus villosus), which suggests an indirect action. The community matrix shows that the community was not stable. The kittiwake population did not decrease as seen in north Norwegian populations. Likewise, the common guillemot population, after a crash in 1985, was recovering at Kharlov while Norwegian populations were decreasing. Only the Brünnich's guillemot showed a decrease at Kharlov until 1999. We suggest that the stability of the kittiwake and common guillemot populations at Kharlov is due to better feeding conditions than in colonies of the Norwegian coast, linked to a possible eastward shift of the capelin population with the temperature increase of the Barents Sea.     

Selection by parasitoid females among closely related hosts based on volatiles: Identifying relevant chemical cues

Parasitoid fitness is influenced by the ability to overcome host defense strategies and by the ability of parasitoid females to select high‐quality host individuals. When females are unable to differentiate among hosts, their fitness will decrease with an increasing abundance of resistant hosts. To understand the effect of mixed host populations on female fitness, it is therefore necessary to investigate the ability of female parasitoids to select among hosts. Here, we used behavioral assays, headspace volatile collection, and electrophysiology to study the ability of Asecodes parviclava to use olfactory cues to select between a susceptible host (Galerucella calmariensis) and a resistant host (Galerucella pusilla) from a distance. Our studies show that parasitoid females have the capacity to distinguish the two hosts and that the selection behavior is acquired through experiences during earlier life stages. Further, we identified two volatiles (α‐terpinolene and [E]‐β‐ocimene) which amounts differ between the two plant–herbivore systems and that caused behavioral and electrophysiological responses. The consequence of this selection behavior is that females have the capacity to avoid laying eggs in G. pusilla, where the egg mortality is higher due to much stronger immune responses toward A. parviclava than in larvae of G. calmariensis.     

Recent speciation in three closely related sympatric specialists: inferences using multi-locus sequence, post-mating isolation and endosymbiont data

Shifting between unrelated host plants is relatively rare for phytophagous insects, and distinct host specificity may play crucial roles in reproductive isolation. However, the isolation status and the relationship between parental divergence and post-mating isolation among closely related sympatric specialists are still poorly understood. Here, multi-locus sequence were used to estimate the relationship among three host plant-specific closely related flea beetles, Altica cirsicola, A. fragariae and A. viridicyanea (abbreviated as AC, AF and AV respectively). The tree topologies were inconsistent using different gene or different combinations of gene fragments. The relationship of AF+(AC+AV) was supported, however, by both gene tree and species tree based on concatenated data. Post-mating reproductive data on the results of crossing these three species are best interpreted in the light of a well established phylogeny. Nuclear-induced but not Wolbachia-induced unidirectional cytoplasmic incompatibility, which was detected in AC-AF and AF-AV but not in AC-AV, may also suggest more close genetic affinity between AC and AV. Prevalence of Wolbachia in these three beetles, and the endosymbiont in most individuals of AV and AC sharing a same wsp haplotype may give another evidence of AF+(AC+AV). Our study also suggested that these three flea beetles diverged in a relative short time (0.94 My), which may be the result of shifting between unrelated host plants and distinct host specificity. Incomplete post-mating isolation while almost complete lineage sorting indicated that effective pre-mating isolation among these three species should have evolved.     

群落构建研究的新进展:进化和生态相结合的群落谱系结构研究

群落如何构建足群落生态学中的重要问题.群落谱系结构研究将物种间的亲缘进化关系运用到群落生态学研究中,利用物种的系统发育状况推测历史因素对现有群落的影响,为推断影响群落组成的生态学机制提供了有效方法.群落谱系结构的研究方法是首先建立可代表群落物种库的超级系统进化树,然后计算群落内物种间的谱系距离,最后通过统计方法检测其与随机模型下的谱系距离是否有显著差异来获得谱系结构(如谱系聚集、谱系发散),从而揭示群落构建中的关键生态过程(如生境过滤、竞争作用).群落谱系结构与空间尺度、分类群尺度、时间尺度等不同研究尺度有关.在小的空间尺度下,随着分类群尺度降低、树木年龄级增大,群落谱系结构从聚集逐渐转为发散;而随群落空间尺度的增大,谱系趋向于聚集.谱系结构受到环境因素影响,因此分析集合群落下的谱系可以揭示区域生态过程的影响.另外,群落谱系结构研究还有助于探讨中性理论、密度制约假说等生态学理论,并预测干扰作用下的群落演化趋势.在利用谱系结构深入探讨群落构建成因时,需要基于生态特征和环境变量共同分析,同时考虑小尺度局域过程(群落的微环境或群落内种间相互作用等)和大尺度区域过程(地史过程和物种形成等),并可结合生态控制实验,以确认群落构建的关键因素.在研究方法和手段上,今后需要注重通过选择合适的基因片段建立系统树,然后通过生态特征来加以校正,以更准确地反映物种间的亲缘距离.另外,获得谱系树后还需要寻找更加合理的统计模型和指数,增加统计分析和解决问题的能力.     

Rhizobium leguminosarum exopolysaccharide mutants: biochemical and genetic analyses and symbiotic behavior on three hosts

Ten independently generated mutants of Rhizobium leguminosarum biovar phaseoli CFN42 isolated after Tn5 mutagenesis formed nonmucoid colonies on all agar media tested and lacked detectable production of the normal acidic exopolysaccharide in liquid culture. The mutants were classified into three groups. Three mutants harbored Tn5 insertions on a 3.6-kilobase-pair EcoRI fragment and were complemented to have normal exopolysaccharide production by cosmids that shared an EcoRI fragment of this size from the CFN42 genome. The Tn5 inserts of five other mutants appeared to be located on a second, slightly smaller EcoRI fragment. Attempts to complement mutants of this second group with cloned DNA were unsuccessful. The mutations of the other two mutants were located in apparently adjacent EcoRI fragments carried on two cosmids that complemented those two mutants. The latter two mutants also lacked O-antigen-containing lipopolysaccharides and induced underdeveloped nodules that lacked nitrogenase activity on bean plants. The other eight mutants had normal lipopolysaccharides and wild-type symbiotic proficiencies on bean plants. Mutants in each of these groups were mated with R. leguminosarum strains that nodulated peas (R. leguminosarum biovar viciae) or clovers (R. leguminosarum biovar trifolii). Transfer of the Tn5 mutations resulted in exopolysaccharide-deficient R. leguminosarum biovar viciae or R. leguminosarum biovar trifolii transconjugants that were symbiotically deficient in all cases. These results support earlier suggestions that successful symbiosis with peas or clovers requires that rhizobia be capable of acidic exopolysaccharide production, whereas symbiosis with beans does not have this requirement.     

Scaling of biological community structure: a systems approach to community complexity

Local community dynamics are determined by the interaction of environmental variation and the biotic properties of communities. This interaction occurs on many spatial and temporal scales, hence the expectation is that community dynamics will be complex. Previous theoretical approaches to communities have assumed linear, near equilibrium dynamics. An alternative approach suggests that community dynamics are the result of the balance between energy use by the community and its tendency to move towards thermodynamic equilibrium, in this case extinction of all species in the community. Because this balance will be imprecise, community dynamics should be oscillatory. Furthermore, because energy use by a community can be broken down into a hierarchical set of processes occurring on different time scales, community dynamics should reflect multiple periodicities. The above theoretical treatment suggests that since community dynamics are scaled, a hierarchical observational approach should help resolve important aspects of community structure. This approach of scaling community observations provides a technique for evaluation of community responses to environmental change, including human induced perturbations. A thermodynamic approach to community dynamics can also provide the basis for new theoretical and empricial discoveries about biological communities.     

Population genetic structure and long-distance dispersal among seabird populations: implications for colony persistence

Dramatic local population decline brought about by anthropogenic-driven change is an increasingly common threat to biodiversity. Seabird life history traits make them particularly vulnerable to such change; therefore, understanding population connectivity and dispersal dynamics is vital for successful management. Our study used a 357-base pair mitochondrial control region locus sequenced for 103 individuals and 18 nuclear microsatellite loci genotyped for 245 individuals to investigate population structure in the Atlantic and Pacific populations of the pelagic seabird, Leach's storm-petrel Oceanodroma leucorhoa leucorhoa. This species is under intense predation pressure at one regionally important colony on St Kilda, Scotland, where a disparity between population decline and predation rates hints at immigration from other large colonies. AMOVA, F(ST), Φ(ST) and Bayesian cluster analyses revealed no genetic structure among Atlantic colonies (Global Φ(ST) = -0.02 P > 0.05, Global F(ST) = 0.003, P > 0.05, STRUCTURE K = 1), consistent with either contemporary gene flow or strong historical association within the ocean basin. The Pacific and Atlantic populations are genetically distinct (Global Φ(ST) = 0.32 P < 0.0001, Global F(ST) = 0.04, P < 0.0001, STRUCTURE K = 2), but evidence for interocean exchange was found with individual exclusion/assignment and population coalescent analyses. These findings highlight the importance of conserving multiple colonies at a number of different sites and suggest that management of this seabird may be best viewed at an oceanic scale. Moreover, our study provides an illustration of how long-distance movement may ameliorate the potentially deleterious impacts of localized environmental change, although direct measures of dispersal are still required to better understand this process.     

Low genetic variation and no detectable population structure in aspergillus fumigatus compared to closely related Neosartorya species

Aspergillus fumigatus is an anamorphic euascomycete mold with a ubiquitous presence worldwide. Despite intensive work to understand its success as a pathogen infecting immunosuppressed patients, the population dynamics and recent evolutionary history of A. fumigatus remain understudied. We examined patterns of genetic variation at three intergenic loci for 70 natural isolates from Europe, North America, South America, Asia, Africa, and Australia. The same loci were used to analyze within-population genetic variation for 33 isolates obtained from five geographic locations. Neither data set detected evidence of population differentiation or found any association between the genetic and geographic distances among these isolates. No evidence for genetic differentiation within the two A. fumigatus mating types was detected. The genetic diversity of A. fumigatus, contrasted with that of its close teleomorphic relatives, Neosartorya fischeri and Neosartorya spinosa, is remarkably low.     

Variance stabilization and normalization for one-color microarray data using a data-driven multiscale approach

MOTIVATION: Many standard statistical techniques are effective on data that are normally distributed with constant variance. Microarray data typically violate these assumptions since they come from non-Gaussian distributions with a non-trivial mean-variance relationship. Several methods have been proposed that transform microarray data to stabilize variance and draw its distribution towards the Gaussian. Some methods, such as log or generalized log, rely on an underlying model for the data. Others, such as the spread-versus-level plot, do not. We propose an alternative data-driven multiscale approach, called the Data-Driven Haar-Fisz for microarrays (DDHFm) with replicates. DDHFm has the advantage of being 'distribution-free' in the sense that no parametric model for the underlying microarray data is required to be specified or estimated; hence, DDHFm can be applied very generally, not just to microarray data. RESULTS: DDHFm achieves very good variance stabilization of microarray data with replicates and produces transformed intensities that are approximately normally distributed. Simulation studies show that it performs better than other existing methods. Application of DDHFm to real one-color cDNA data validates these results. AVAILABILITY: The R package of the Data-Driven Haar-Fisz transform (DDHFm) for microarrays is available in Bioconductor and CRAN.