Nonparametric variable importance assessment using machine learning techniques

In a regression setting, it is often of interest to quantify the importance of various features in predicting the response. Commonly, the variable importance measure used is determined by the regression technique employed. For this reason, practitioners often only resort to one of a few regression techniques for which a variable importance measure is naturally defined. Unfortunately, these regression techniques are often suboptimal for predicting the response. Additionally, because the variable importance measures native to different regression techniques generally have a different interpretation, comparisons across techniques can be difficult. In this work, we study a variable importance measure that can be used with any regression technique, and whose interpretation is agnostic to the technique used. This measure is a property of the true data‐generating mechanism. Specifically, we discuss a generalization of the analysis of variance variable importance measure and discuss how it facilitates the use of machine learning techniques to flexibly estimate the variable importance of a single feature or group of features. The importance of each feature or group of features in the data can then be described individually, using this measure. We describe how to construct an efficient estimator of this measure as well as a valid confidence interval. Through simulations, we show that our proposal has good practical operating characteristics, and we illustrate its use with data from a study of risk factors for cardiovascular disease in South Africa.     

Identification of SNP interactions using logic regression

Interactions of single nucleotide polymorphisms (SNPs) are assumed to be responsible for complex diseases such as sporadic breast cancer. Important goals of studies concerned with such genetic data are thus to identify combinations of SNPs that lead to a higher risk of developing a disease and to measure the importance of these interactions. There are many approaches based on classification methods such as CART and random forests that allow measuring the importance of single variables. But none of these methods enable the importance of combinations of variables to be quantified directly. In this paper, we show how logic regression can be employed to identify SNP interactions explanatory for the disease status in a case-control study and propose 2 measures for quantifying the importance of these interactions for classification. These approaches are then applied on the one hand to simulated data sets and on the other hand to the SNP data of the GENICA study, a study dedicated to the identification of genetic and gene-environment interactions associated with sporadic breast cancer.     

Use of Mechanistic Models to Integrate and Analyze Multiple Proteomic Datasets

Proteins in cell signaling networks tend to interact promiscuously through low-affinity interactions. Consequently, evaluating the physiological importance of mapped interactions can be difficult. Attempts to do so have tended to focus on single, measurable physicochemical factors, such as affinity or abundance. For example, interaction importance has been assessed on the basis of the relative affinities of binding partners for a protein of interest, such as a receptor. However, multiple factors can be expected to simultaneously influence the recruitment of proteins to a receptor (and the potential of these proteins to contribute to receptor signaling), including affinity, abundance, and competition, which is a network property. Here, we demonstrate that measurements of protein copy numbers and binding affinities can be integrated within the framework of a mechanistic, computational model that accounts for mass action and competition. We use cell line-specific models to rank the relative importance of protein-protein interactions in the epidermal growth factor receptor (EGFR) signaling network for 11 different cell lines. Each model accounts for experimentally characterized interactions of six autophosphorylation sites in EGFR with proteins containing a Src homology 2 and/or phosphotyrosine-binding domain. We measure importance as the predicted maximal extent of recruitment of a protein to EGFR following ligand-stimulated activation of EGFR signaling. We find that interactions ranked highly by this metric include experimentally detected interactions. Proteins with high importance rank in multiple cell lines include proteins with recognized, well-characterized roles in EGFR signaling, such as GRB2 and SHC1, as well as a protein with a less well-defined role, YES1. Our results reveal potential cell line-specific differences in recruitment.     

Synthesis and relevance of cardiac eicosanoids with particular emphasis on ischemia and reperfusion

Eicosanoids represent a family of compounds derived primarily from arachidonic acid. It is now known that arachidonic acid can undergo metabolism via at least three distinct pathways, although the most readily understood are those resulting in prostaglandin or leukotriene formation via cyclooxygenase and 5-lipoxygenase, respectively. These products can be synthesized by the heart or can be released from accumulating neutrophils under various pathological conditions. Eicosanoids possess a wide array of pharmacological actions that could be of importance either in the initiation or modulation of various cardiac diseases. Here, we review the potential importance of eicosanoids to ischemic heart disease. Data are cited that examine the potential importance of these compounds to experimentally induced cardiac injury as well as clinically observed ischemic heart disease. Particular emphasis is placed on recent studies that document the relevance of endogenously synthesized arachidonic acid metabolites as well as the consequence of modulating eicosanoid synthesis through pharmacological or dietary means on cardiac injury under experimental or clinical situations.     

鄂尔多斯高原风蚀沙化梁地克隆植物的分布及其与物种多样性的关系

 采用样线法和点样法对鄂尔多斯风蚀沙化梁地上的植物进行了调查,研究了梁地上克隆植物的分布、克隆植物在群落中的重要性及其与群落物种多样性之间的关系。结果表明：1）在梁顶的典型草原植物群落和梁底的滩地盐生植物群落中克隆植物的物种丰富度较高,而在梁坡的沙生植物群落中较少,非克隆植物物种在梁顶出现频率较高,在梁坡和梁底较少;密集型克隆植物物种丰富度的变化与非克隆植物相同,游击型克隆植物在梁顶植物群落中没有出现,在梁坡群落中也很少,而在梁底的滩地中较多。2）梁底群落中克隆植物的重要值高于梁顶和梁坡群落中克隆植物的重要值,梁顶与梁坡群落中非克隆植物的重要值高于梁底群落中非克隆植物的重要值;在梁坡与梁底群落中,克隆植物的重要值都高于非克隆植物的重要值,而在梁顶群落中克隆植物与非克隆植物的重要值之间无差异;梁顶与梁坡群落中密集型克隆植物的重要值高于梁底群落中密集型克隆植物的重要值,而游击型克隆植物的重要值在梁底群落中较高;在梁顶与梁坡群落中,密集型克隆植物的重要值分别高于游击型克隆植物的重要值,而在梁底群落中,密集型克隆植物的重要值低于游击型克隆植物的重要值。3）梁顶的典型草原植物群落中,群落物种多样性随非克隆植物的重要值、克隆植物的重要值、密集型克隆植物的重要值的变化趋势相同,都为抛物线型。梁坡的沙生植物群落中,群落物种多样性与非克隆植物的重要值呈正相关,而分别与克隆植物和密集型克隆植物的重要值呈负相关。梁底的滩地盐生植物群落中,群落物种多样性与非克隆植物的重要值呈正相关,而与克隆植物和游击型克隆植物的重要值呈负相关。     

Translating clinical proteomics: the importance of study design

Mass spectrometry-based clinical proteomics approaches were introduced into the biomedical field more than two decades ago. Despite recent developments both in the field of mass spectrometry and bioinformatics, the gap between proteomics results and their translation into clinical practice still needs to be closed, as implementation of proteomics results in the clinic appears to be scarce. An extra focus on the importance of the experimental design is therefore of crucial importance.     

Importance of ticks and tick-borne diseases of domestic ruminants in Southern Europe

In Southern Europe, as elsewhere, the importance of ticks and tick-borne diseases (TBD) depends on the species of tick vectors present and also varies to a great extent according to the genetic susceptibility of the livestock involved, as well as other factors such as management and vegetation. There are large gaps in our knowledge of the tick species in the various countries, and of the prevalence and economic importance of the diseases they transmit. There is a great need for tick surveys as well as surveys of tick-borne infections. It is essential to distinguish between infection and disease. A list is given of TBD and their vectors known or to be expected to occur in domestic ruminants in Southern Europe. It is hoped that the exchange of knowledge during this workshop will indicate the importance of these parasites in the various countries involved.     

Comparison of parents' expectations and importance ratings for specific aspects of childbirth.

We examined parents'' expectations of many aspects of the birth experience and compared them with the importance they attached to these aspects. Expectation was defined as a respondent''s rating that a given practice would be reality, and importance was defined as a respondent''s rating of the personal importance of a practice were all options possible. Subjects in the last 6 to 8 weeks of pregnancy were asked by their obstetricians to complete a questionnaire; 231 mothers and 227 fathers responded, for a response rate of 95%. In each of six subcategories parents'' importance ratings significantly exceeded their expectation ratings. Certain items were rated as relatively less important postnatally than prenatally and by multigravid women than by primigravid women. Parents'' perceptions of available options reflect consistent discrepancy with what they wish were possible. However, increased efforts to inform parents of existing options and to provide the rationale for specific practices may reduce the discrepancy between importance and expectations. This would in turn heighten the likelihood of a psychologically positive birth experience for parents.     

Political polarization of conservation issues in the era of COVID-19: An examination of partisan perspectives and priorities in the United States

As a zoonotic disease with unprecedented global impacts, COVID-19 may influence how people prioritize issues related to wildlife conservation. Using a nationally representative sample of US residents, we investigated: (1) how COVID-19 affected the relative importance of conservation issues among adults with different political ideologies, and (2) how the pandemic affected political polarization of conservation issues during the 2020 general election in the United States. Conservation issues such as endangered species and controlling zoonotic disease ranked low in importance among the 14 policy issues considered, even lower than environmental issues such as climate change and environmental protection; however, the importance of all conservation issues increased as a result of COVID-19. Political polarization surrounding the perceived importance of conservation issues also increased during the COVID-19 pandemic, with Democrats reporting larger increases in importance than Republicans. Polarization was driven by the most conservative Republicans and the most liberal Democrats. But this polarization was less extreme than it was for other issues such as climate change and healthcare. Findings highlight the need for communicating links between zoonotic disease and human interactions with wildlife and the environment. Acting quickly may be critical in areas where conservation issues are primed to succumb to political polarization.     

Conceptual problems of Ecological Continuity and its bioindicators

Very old, undisturbed forest stands may be important for biodiversity through their content of microhabitats or for the long periods available for colonisation, or for both. The term Ecological Continuity (EC) has been used to ascribe value to old forest stands. The relative importance of microhabitat and time for colonisation are usually not kept apart when EC is used as a conservation criterium. EC is broadly applied but poorly defined. Use of EC may lead to underestimation of the importance of forest dynamics and dispersal, and to overestimation of the importance of local land use history. If bioindicators of long-term habitat persistence are to be used, species with low dispersal capacity should be chosen. However, many lichens and other fungi, bryophytes and insects cited as indicators of EC, seem to have a patch-tracking lifestyle. They are 'colonists' according to life history strategy classification, and rather seem to indicate specific microhabitats. Terrestrial molluscs, some vascular forest plants, and those bryophytes and lichenized fungi classified as 'perennial stayers' in life history strategy classification, might be used to indicate long-term habitat persistence in forests, but more research is needed to evaluate such indicators.     

The importance of silicon for marine production*

A review of silicon, with emphasis on its forms, uptake, dissolution and role in marine primary production, is given. The importance of silicon in marine food webs is discussed, as well as the concentrations of silicon in various areas and the importance of changing N:Si:P ratios. The methodology for measuring silicate transformations has recently been improved by the introduction of the highly enriched ³²Si isotope. Results from uptake experiments using ³²Si in nutrient rich coastal water and in open ocean are presented. The uptake kinetic experiments showed that the silicate uptake usually is unsaturated. We propose that closer attention in the future should be paid to the importance of balanced nutrient composition as well as nutrient supply dynamics for the development of eutrophication versus efficient trophic transfer and fish production in nutrient enriched systems. Close attention should also be paid to the mechanisms that reduce the inputs of silicate to coastal waters.     

BRCA1 exon 11 alternative splicing, multiple functions and the association with cancer

BRCA1 (breast cancer early-onset 1) alternative splicing levels are regulated in a cell-cycle- and cell-type-specific manner, with splice variants being present in different proportions in tumour cell lines as well as in normal mammary epithelial cells. The importance of this difference in the pathogenesis of breast cancer has yet to be determined. Developing an understanding of the impact of BRCA1 isoform ratio changes on cell phenotype will be of value in breast cancer and may offer therapeutic options. In the present paper, we describe the splicing isoforms of BRCA1 exon 11, their possible role in cancer biology and the importance of maintaining a balanced ratio.     

西南地区生态重要性格局研究

生态保护重要性是表征区域生态系统结构和功能重要性程度的综合指标。在明确西南地区自然环境特征和关键生态问题的基础上,综合分析生态敏感性和生态服务功能重要性,构建西南生态重要性评估指标体系,并运用GIS技术识别具有重要保护意义的地区,为优化生态保护策略、划定生态红线和主体功能区划提供技术支持。研究表明:西南地区生态保护极重要区为75.7万km2,主要分布于雅鲁藏布江流域、横断山区、川西高原等地区。该区占西南地区总面积的32.5%,提供了59.4%的土壤保持总量和54.5%的水源涵养总量,保护了69.7%的重要物种。研究建议,划定极重要区为生态红线进行强制性严格保护,以保障和维护生态安全,促进西南地区可持续发展。     

Functional diversity (FD), species richness and community composition

Functional diversity is an important component of biodiversity, yet in comparison to taxonomic diversity, methods of quantifying functional diversity are less well developed. Here, we propose a means for quantifying functional diversity that may be particularly useful for determining how functional diversity is related to ecosystem functioning. This measure of functional diversity “FD” is defined as the total branch length of a functional dendrogram. Various characteristics of FD make it preferable to other measures of functional diversity, such as the number of functional groups in a community. Simulating species' trait values illustrates how the relative importance of richness and composition for FD depends on the effective dimensionality of the trait space in which species separate. Fewer dimensions increase the importance of community composition and functional redundancy. More dimensions increase the importance of species richness and decreases functional redundancy. Clumping of species in trait space increases the relative importance of community composition. Five natural communities show remarkably similar relationships between FD and species richness.     

Developing a breeding strategy to exploit quantitative variation in symbiotic nitrogen fixation

Summary This paper examines evidence which quantifies the relative importance of legume and Rhizobium genotypes as determinants of phenotypic variation in symbiotic nitrogen fixation. It demonstrates potentially large and unpredictable effects of the Rhizobium genotype. The likely importance of such effects on crop yield is considered. The information is then used to assess ways in which legume breeding programmes may be altered to encompass the effects of genetic variation in Rhizobium.     

重要值-面积曲线在热带亚热带森林中的应用

重要值—面积曲线不仅可以确定群落最小面积,并可绘出种群的相对稳定的重要值,表达出群落的最主要特征,因而,具有一定的理论意义和实用价值。 在种类复杂而偶见种又很多的森林群落中,如以传统的种—面积曲线确定的最小面积会很大,但实际上许多偶见种对群落性质的影响并不重要。如应用重要值—面积曲线获得的最小面积虽然相对较小,但它能满足群落学研究的需要。在寡种群落中,以种—面积曲线确定的最小面积会过小,不足以表达群落的主要特征,而重要值—面积曲线则可得到较大的,足以表达群落特征的最小面积。本文提出重要值数学公式,以及重复随机集合取样的数学公式为：以助于确定最小面积。     

都江堰市水源涵养功能空间格局

水源涵养是生态系统的重要服务功能之一.识别水源涵养重要区对于制定水源保护规划,落实陆域-水域综合保护,防治水体污染,保障区域用水安全具有重要意义.以2008年四川都江堰市为研究范围,采用生态系统服务功能综合权衡工具-Integrated Valuation of Ecosystem Services and Tradeoffs(InVEST)中的水源涵养模型对县域生态系统的水源涵养功能进行了空间制图,然后采用综合指数法将水源涵养功能与植被减洪能力和水源地保护区进行叠加计算水源涵养重要性指数.结果表明:尽管汶川地震导致都江堰市域部分地区水源涵养功能降低,但都江堰市水源涵养功能整体仍然较高,部分地区多年平均水源涵养量超过200mm/a.都江堰市的水源涵养极重要区面积为421km2,占全市总面积的34.9％,远大于目前县域的水源保护区面积,未来应进一步加强对这些极重要区的保护,以保证本市以及成都平原的供水安全.水域涵养功能及其重要性都存在较明显的空间分异,大致呈现由西北向东南减少的趋势,水源涵养重要性与水源功能的分布密切相关,但又不完全一致.因此,在县域尺度上采用服务功能空间化评估结果进行重要性评价能够比较精细地反映生态系统水源涵养重要性的空间差异.但该方法还需要将人类用水需求进行空间表达,与本文的方法进行结合,以更准确反映水源涵养重要性的空间分异.     

Estimating and Projecting Trends in HIV/AIDS Generalized Epidemics Using Incremental Mixture Importance Sampling

Summary The Joint United Nations Programme on HIV/AIDS (UNAIDS) has decided to use Bayesian melding as the basis for its probabilistic projections of HIV prevalence in countries with generalized epidemics. This combines a mechanistic epidemiological model, prevalence data, and expert opinion. Initially, the posterior distribution was approximated by sampling‐importance‐resampling, which is simple to implement, easy to interpret, transparent to users, and gave acceptable results for most countries. For some countries, however, this is not computationally efficient because the posterior distribution tends to be concentrated around nonlinear ridges and can also be multimodal. We propose instead incremental mixture importance sampling (IMIS), which iteratively builds up a better importance sampling function. This retains the simplicity and transparency of sampling importance resampling, but is much more efficient computationally. It also leads to a simple estimator of the integrated likelihood that is the basis for Bayesian model comparison and model averaging. In simulation experiments and on real data, it outperformed both sampling importance resampling and three publicly available generic Markov chain Monte Carlo algorithms for this kind of problem.     

Alpheus Spring Packard and cave fauna in the evolution debate

Conclusion Packard attempted to incorporate cave fauna into a general theory of evolution that would be consistent with the principle of recapitulation, and would have as the primary mechanism the inheritance of the effects of the environment. Beyond this, he also attempted to demonstrate that the evolution of cave fauna was consistent with progressive evolution. The use he made of comparative anatomy and embryology places him within the tradition of classical morphology that was dominant through much of the last half of the nineteenth century, but of waning importance by the time of Packard's death in 1905. The importance Packard gave to cave fauna as evidence for Lamarckian evolution stimulated interest in the phenomenon; this interest, and references to cave fauna in the scientific literature, declined after his death. Since then, the importance of cave fauna in evolutionary theory has declined from their status as the star evidence in Packard's theory to their present status as a difficult anomaly within the modern synthetic theory.