Do pseudo-absence selection strategies influence species distribution models and their predictions? An information-theoretic approach based on simulated data

Background Multiple logistic regression is precluded from many practical applications in ecology that aim to predict the geographic distributions of species because it requires absence data, which are rarely available or are unreliable. In order to use multiple logistic regression, many studies have simulated "pseudo-absences" through a number of strategies, but it is unknown how the choice of strategy influences models and their geographic predictions of species. In this paper we evaluate the effect of several prevailing pseudo-absence strategies on the predictions of the geographic distribution of a virtual species whose "true" distribution and relationship to three environmental predictors was predefined. We evaluated the effect of using a) real absences b) pseudo-absences selected randomly from the background and c) two-step approaches: pseudo-absences selected from low suitability areas predicted by either Ecological Niche Factor Analysis: (ENFA) or BIOCLIM. We compared how the choice of pseudo-absence strategy affected model fit, predictive power, and information-theoretic model selection results. Results Models built with true absences had the best predictive power, best discriminatory power, and the "true" model (the one that contained the correct predictors) was supported by the data according to AIC, as expected. Models based on random pseudo-absences had among the lowest fit, but yielded the second highest AUC value (0.97), and the "true" model was also supported by the data. Models based on two-step approaches had intermediate fit, the lowest predictive power, and the "true" model was not supported by the data. Conclusion If ecologists wish to build parsimonious GLM models that will allow them to make robust predictions, a reasonable approach is to use a large number of randomly selected pseudo-absences, and perform model selection based on an information theoretic approach. However, the resulting models can be expected to have limited fit.

The MicroArray Quality Control (MAQC)-II study of common practices for the development and validation of microarray-based predictive models.

Gene expression data from microarrays are being applied to predict preclinical and clinical endpoints, but the reliability of these predictions has not been established. In the MAQC-II project, 36 independent teams analyzed six microarray data sets to generate predictive models for classifying a sample with respect to one of 13 endpoints indicative of lung or liver toxicity in rodents, or of breast cancer, multiple myeloma or neuroblastoma in humans. In total, >30,000 models were built using many combinations of analytical methods. The teams generated predictive models without knowing the biological meaning of some of the endpoints and, to mimic clinical reality, tested the models on data that had not been used for training. We found that model performance depended largely on the endpoint and team proficiency and that different approaches generated models of similar performance. The conclusions and recommendations from MAQC-II should be useful for regulatory agencies, study committees and independent investigators that evaluate methods for global gene expression analysis.

Weighted Shape-based Averaging with Neighborhood Prior Model for Multiple Atlas Fusion-based Medical Image Segmentation

In medical imaging, merging automated segmentations obtained from multiple atlases has become a standard practice for improving the accuracy. In this letter, we propose two new fusion methods: "Global Weighted Shape-Based Averaging" (GWSBA) and "Local Weighted Shape-Based Averaging" (LWSBA). These methods extend the well known Shape-Based Averaging (SBA) by additionally incorporating the similarity information between the reference (i.e., atlas) images and the target image to be segmented. We also propose a new spatially-varying similarity-weighted neighborhood prior model, and an edge-preserving smoothness term that can be used with many of the existing fusion methods. We first present our new Markov Random Field (MRF) based fusion framework that models the above mentioned information. The proposed methods are evaluated in the context of segmentation of lymph nodes in the head and neck 3D CT images, and they resulted in more accurate segmentations compared to the existing SBA.

One naive T cell, multiple fates in CD8+ T cell differentiation.

The mechanism by which the immune system produces effector and memory T cells is largely unclear. To allow a large-scale assessment of the development of single naive T cells into different subsets, we have developed a technology that introduces unique genetic tags (barcodes) into naive T cells. By comparing the barcodes present in antigen-specific effector and memory T cell populations in systemic and local infection models, at different anatomical sites, and for TCR-pMHC interactions of different avidities, we demonstrate that under all conditions tested, individual naive T cells yield both effector and memory CD8+ T cell progeny. This indicates that effector and memory fate decisions are not determined by the nature of the priming antigen-presenting cell or the time of T cell priming. Instead, for both low and high avidity T cells, individual naive T cells have multiple fates and can differentiate into effector and memory T cell subsets.

Simultaneous co-expression of memory- and effector-related genes by individual human CD8 T cells depends on antigen specificity and differentiation

Purpose/Objective: Phenotypic and functional T cell properties are usually analyzed at the level of defined cell populations. However, large differences between individual T cells may have important functional consequences. To answer this issue, we performed highly sensitive single-cell gene expression profiling, which allows the direct ex vivo characterization of individual virus- and tumor-specific T cells from healthy donors and melanoma patients. Materials and methods: HLA-A*0201-positive patients with stage III/ IV metastatic melanoma were included in a phase I clinical trial (LUD- 00-018). Patients received monthly low-dose of the Melan-AMART- 1 26_35 unmodified natural (EAAGIGILTV) or the analog A27L (ELAGIGILTV) peptides, mixed CPG and IFA. Individual effector memory CD28+ (EM28+) and EM28- tetramer-specific CD8pos T cells were sorted by flow cytometer. Following direct cell lysis and reverse transcription, the resulting cDNA was precipitated and globally amplified. Semi-quantitative PCR was used for gene expression and TCR BV repertoire analyses. Results: We have previously shown that vaccination with the natural Melan-A peptide induced T cells with superior effector functions as compared to the analog peptide optimized for enhanced HLA binding. Here we found that natural peptide vaccination induced EM28+ T cells with frequent co-expression of both memory/homing-associated genes (CD27, IL7R, EOMES, CXCR3 and CCR5) and effector-related genes (IFNG, KLRD1, PRF1 and GZMB), comparable to protective EBV- and CMV-specific T cells. In contrast, memory/homing- and effectorassociated genes were less frequently co-expressed after vaccination with the analog peptide. Conclusions: These findings reveal a previously unknown level of gene expression diversity among vaccine- and virus-specific T cells with the simultaneous co-expression of multiple memory/homing- and effector- related genes by the same cell. Such broad functional gene expression signatures within antigen-specific T cells may be critical for mounting efficient responses to pathogens or tumors. In summary, direct ex vivo high-resolution molecular characterization of individual T cells provides key insights into the processes shaping the functional properties of tumor- and virus-specific T cells.

Dispersal strategies, few dominating or many coexisting: the effect of environmental spatial structure and multiple sources of mortality.

Interspecific competition, life history traits, environmental heterogeneity and spatial structure as well as disturbance are known to impact the successful dispersal strategies in metacommunities. However, studies on the direction of impact of those factors on dispersal have yielded contradictory results and often considered only few competing dispersal strategies at the same time. We used a unifying modeling approach to contrast the combined effects of species traits (adult survival, specialization), environmental heterogeneity and structure (spatial autocorrelation, habitat availability) and disturbance on the selected, maintained and coexisting dispersal strategies in heterogeneous metacommunities. Using a negative exponential dispersal kernel, we allowed for variation of both species dispersal distance and dispersal rate. We showed that strong disturbance promotes species with high dispersal abilities, while low local adult survival and habitat availability select against them. Spatial autocorrelation favors species with higher dispersal ability when adult survival and disturbance rate are low, and selects against them in the opposite situation. Interestingly, several dispersal strategies coexist when disturbance and adult survival act in opposition, as for example when strong disturbance regime favors species with high dispersal abilities while low adult survival selects species with low dispersal. Our results unify apparently contradictory previous results and demonstrate that spatial structure, disturbance and adult survival determine the success and diversity of coexisting dispersal strategies in competing metacommunities.

Simultaneous determination of antidementia drugs in human plasma: Procedure transfer from HPLC-MS to UPLC-MS/MS.

A previously developed high performance liquid chromatography mass spectrometry (HPLC-MS) procedure for the simultaneous determination of antidementia drugs, including donepezil, galantamine, memantine, rivastigmine and its metabolite NAP 226-90, was transferred to an ultra performance liquid chromatography system coupled to a tandem mass spectrometer (UPLC-MS/MS). The drugs and their internal standards ([(2)H(7)]-donepezil, [(13)C,(2)H(3)]-galantamine, [(13)C(2),(2)H(6)]-memantine, [(2)H(6)]-rivastigmine) were extracted from 250μL human plasma by protein precipitation with acetonitrile. Chromatographic separation was achieved on a reverse phase column (BEH C18 2.1mm×50mm; 1.7μm) with a gradient elution of an ammonium acetate buffer at pH 9.3 and acetonitrile at a flow rate of 0.4mL/min and an overall run time of 4.5min. The analytes were detected on a tandem quadrupole mass spectrometer operated in positive electrospray ionization mode, and quantification was performed using multiple reaction monitoring. The method was validated according to the recommendations of international guidelines over a calibration range of 1-300ng/mL for donepezil, galantamine and memantine, and 0.2-50ng/mL for rivastimgine and NAP 226-90. The trueness (86-108%), repeatability (0.8-8.3%), intermediate precision (2.3-10.9%) and selectivity of the method were found to be satisfactory. Matrix effects variability was inferior to 15% for the analytes and inferior to 5% after correction by internal standards. A method comparison was performed with patients' samples showing similar results between the HPLC-MS and UPLC-MS/MS procedures. Thus, this validated UPLC-MS/MS method allows to reduce the required amount of plasma, to use a simplified sample preparation, and to obtain a higher sensitivity and specificity with a much shortened run-time.

Further insights on the French WISC-IV factor structure through Bayesian structural equation modeling

The interpretation of the Wechsler Intelligence Scale for Children-Fourth Edition (WISC-IV) is based on a 4-factor model, which is only partially compatible with the mainstream Cattell-Horn-Carroll (CHC) model of intelligence measurement. The structure of cognitive batteries is frequently analyzed via exploratory factor analysis and/or confirmatory factor analysis. With classical confirmatory factor analysis, almost all crossloadings between latent variables and measures are fixed to zero in order to allow the model to be identified. However, inappropriate zero cross-loadings can contribute to poor model fit, distorted factors, and biased factor correlations; most important, they do not necessarily faithfully reflect theory. To deal with these methodological and theoretical limitations, we used a new statistical approach, Bayesian structural equation modeling (BSEM), among a sample of 249 French-speaking Swiss children (8-12 years). With BSEM, zero-fixed cross-loadings between latent variables and measures are replaced by approximate zeros, based on informative, small-variance priors. Results indicated that a direct hierarchical CHC-based model with 5 factors plus a general intelligence factor better represented the structure of the WISC-IV than did the 4-factor structure and the higher order models. Because a direct hierarchical CHC model was more adequate, it was concluded that the general factor should be considered as a breadth rather than a superordinate factor. Because it was possible for us to estimate the influence of each of the latent variables on the 15 subtest scores, BSEM allowed improvement of the understanding of the structure of intelligence tests and the clinical interpretation of the subtest scores.

Predicting species distribution: offering more than simple habitat models

n the last two decades, interest in species distribution models (SDMs) of plants and animals has grown dramatically. Recent advances in SDMs allow us to potentially forecast anthropogenic effects on patterns of biodiversity at different spatial scales. However, some limitations still preclude the use of SDMs in many theoretical and practical applications. Here, we provide an overview of recent advances in this field, discuss the ecological principles and assumptions underpinning SDMs, and highlight critical limitations and decisions inherent in the construction and evaluation of SDMs. Particular emphasis is given to the use of SDMs for the assessment of climate change impacts and conservation management issues. We suggest new avenues for incorporating species migration, population dynamics, biotic interactions and community ecology into SDMs at multiple spatial scales. Addressing all these issues requires a better integration of SDMs with ecological theory.

Uncertainty Quantification with Support Vector Regression Prediction Models

Uncertainty quantification of petroleum reservoir models is one of the present challenges, which is usually approached with a wide range of geostatistical tools linked with statistical optimisation or/and inference algorithms. The paper considers a data driven approach in modelling uncertainty in spatial predictions. Proposed semi-supervised Support Vector Regression (SVR) model has demonstrated its capability to represent realistic features and describe stochastic variability and non-uniqueness of spatial properties. It is able to capture and preserve key spatial dependencies such as connectivity, which is often difficult to achieve with two-point geostatistical models. Semi-supervised SVR is designed to integrate various kinds of conditioning data and learn dependences from them. A stochastic semi-supervised SVR model is integrated into a Bayesian framework to quantify uncertainty with multiple models fitted to dynamic observations. The developed approach is illustrated with a reservoir case study. The resulting probabilistic production forecasts are described by uncertainty envelopes.

MP2RAGE Multiple Sclerosis Magnetic Resonance Imaging at 3 T.

OBJECTIVES: Lesion detection and characterization in multiple sclerosis (MS) are an essential part of its clinical diagnosis and an important research field. In this pilot study, we applied the recently introduced two inversion-contrast magnetization-prepared rapid gradient echo sequence (MP2RAGE) to patients with early-stage MS.¦MATERIALS AND METHODS: The MP2RAGE is a 3-dimensional (3D) magnetization-prepared rapid gradient echo derivative providing homogeneous T1 weighting and simultaneous T1 mapping. The MP2RAGE performance was compared with that of 2 clinical routine sequences (2D fluid-attenuated inversion recovery [FLAIR] and 3D magnetization-prepared rapid gradient echo [MP-RAGE]) and 2 state-of-the art clinical research sequences (the 3D FLAIR-SPACE [sampling perfection with application-optimized contrasts by using different flip-angle evolutions], a fluid-attenuated variable flip-angle fast spin echo technique, and the 3D double-inversion recovery SPACE). A cohort of 10 early-stage female MS patients (age, 31.6 ± 4.7 years; disease duration, 3.8 ± 1.9 years; median expanded disability status scale score, 1.75) and 10 age- and gender-matched controls were enrolled after approval of the local institutional review board was obtained. Multiple sclerosis lesions were identified and assigned to brain locations and tissue types by two experienced physicians in all 5 contrasts. Subsequently, lesions were manually delineated for comparison and statistical analysis of lesion count, volume and quantitative measures.¦RESULTS AND CONCLUSIONS: The results show that the 3D T1-weighted high-resolution MP2RAGE contrast provides a sensitive means for MS lesion assessment. The additional quantitative T1 relaxation time maps obtained with the MP2RAGE provide further potential diagnostic and prognostic information that could help (a) to better discriminate lesion subtypes and (b) to stage and predict the activity and the evolution of MS. Results also indicate that the T2-weighted double-inversion recovery and FLAIR-SPACE contrasts are attractive complements to the MP2RAGE for lesion detection.

Retrospective feasibility study of simultaneous integrated boost in cervical cancer using Tomotherapy: the impact of organ motion and tumor regression.

BACKGROUND: Whole pelvis intensity modulated radiotherapy (IMRT) is increasingly being used to treat cervical cancer aiming to reduce side effects. Encouraged by this, some groups have proposed the use of simultaneous integrated boost (SIB) to target the tumor, either to get a higher tumoricidal effect or to replace brachytherapy. Nevertheless, physiological organ movement and rapid tumor regression throughout treatment might substantially reduce any benefit of this approach. PURPOSE: To evaluate the clinical target volume - simultaneous integrated boost (CTV-SIB) regression and motion during chemo-radiotherapy (CRT) for cervical cancer, and to monitor treatment progress dosimetrically and volumetrically to ensure treatment goals are met. METHODS AND MATERIALS: Ten patients treated with standard doses of CRT and brachytherapy were retrospectively re-planned using a helical Tomotherapy - SIB technique for the hypothetical scenario of this feasibility study. Target and organs at risk (OAR) were contoured on deformable fused planning-computed tomography and megavoltage computed tomography images. The CTV-SIB volume regression was determined. The center of mass (CM) was used to evaluate the degree of motion. The Dice's similarity coefficient (DSC) was used to assess the spatial overlap of CTV-SIBs between scans. A cumulative dose-volume histogram modeled estimated delivered doses. RESULTS: The CTV-SIB relative reduction was between 31 and 70%. The mean maximum CM change was 12.5, 9, and 3 mm in the superior-inferior, antero-posterior, and right-left dimensions, respectively. The CTV-SIB-DSC approached 1 in the first week of treatment, indicating almost perfect overlap. CTV-SIB-DSC regressed linearly during therapy, and by the end of treatment was 0.5, indicating 50% discordance. Two patients received less than 95% of the prescribed dose. Much higher doses to the OAR were observed. A multiple regression analysis showed a significant interaction between CTV-SIB reduction and OAR dose increase. CONCLUSIONS: The CTV-SIB had important regression and motion during CRT, receiving lower therapeutic doses than expected. The OAR had unpredictable shifts and received higher doses. The use of SIB without frequent adaptation of the treatment plan exposes cervical cancer patients to an unpredictable risk of under-dosing the target and/or overdosing adjacent critical structures. In that scenario, brachytherapy continues to be the gold standard approach.

Roles of multiple acyl-CoA oxidases in the routing of carbon flow towards β-oxidation and polyhydroxyalkanoate biosynthesis in Yarrowia lipolytica.

The oleaginous yeast Yarrowia lipolytica possesses six acyl-CoA oxidase (Aox) isoenzymes encoded by genes POX1-POX6. The respective roles of these multiple Aox isoenzymes were studied in recombinant Y. lipolytica strains that express heterologous polyhydroxyalkanoate (PHA) synthase (phaC) of Pseudomonas aeruginosa in varying POX genetic backgrounds, thus allowing assessment of the impact of specific Aox enzymes on the routing of carbon flow to β-oxidation or to PHA biosynthesis. Analysis of PHA production yields during growth on fatty acids with different chain lengths has revealed that the POX genotype significantly affects the PHA levels, but not the monomer composition of PHA. Aox3p function was found to be responsible for 90% and 75% of the total PHA produced from either C9:0 or C13:0 fatty acid, respectively, whereas Aox5p encodes the main Aox involved in the biosynthesis of 70% of PHA from C9:0 fatty acid. Other Aoxs, such as Aox1p, Aox2p, Aox4p and Aox6p, were not found to play a significant role in PHA biosynthesis, independent of the chain length of the fatty acid used. Finally, three known models of β-oxidation are discussed and it is shown that a 'leaky-hose pipe model' of the cycle can be applied to Y. lipolytica.

Simultaneous coexpression of memory-related and effector-related genes by individual human CD8 T cells depends on antigen specificity and differentiation.

Phenotypic and functional cell properties are usually analyzed at the level of defined cell populations but not single cells. Yet, large differences between individual cells may have important functional consequences. It is likely that T-cell-mediated immunity depends on the polyfunctionality of individual T cells, rather than the sum of functions of responding T-cell subpopulations. We performed highly sensitive single-cell gene expression profiling, allowing the direct ex vivo characterization of individual virus-specific and tumor-specific T cells from healthy donors and melanoma patients. We have previously shown that vaccination with the natural tumor peptide Melan-A-induced T cells with superior effector functions as compared with vaccination with the analog peptide optimized for enhanced HLA-A*0201 binding. Here we found that natural peptide vaccination induced tumor-reactive CD8 T cells with frequent coexpression of both memory/homing-associated genes (CD27, IL7R, EOMES, CXCR3, and CCR5) and effector-related genes (IFNG, KLRD1, PRF1, and GZMB), comparable with protective Epstein-Barr virus-specific and cytomegalovirus-specific T cells. In contrast, memory/homing-associated and effector-associated genes were less frequently coexpressed after vaccination with the analog peptide. Remarkably, these findings reveal a previously unknown level of gene expression diversity among vaccine-specific and virus-specific T cells with the simultaneous coexpression of multiple memory/homing-related and effector-related genes by the same cell. Such broad functional gene expression signatures within antigen-specific T cells may be critical for mounting efficient responses to pathogens or tumors. In summary, direct ex vivo high-resolution molecular characterization of individual T cells provides key insights into the processes shaping the functional properties of tumor-specific and virus-specific T cells.

Learning wind fields with multiple kernels

This paper presents multiple kernel learning (MKL) regression as an exploratory spatial data analysis and modelling tool. The MKL approach is introduced as an extension of support vector regression, where MKL uses dedicated kernels to divide a given task into sub-problems and to treat them separately in an effective way. It provides better interpretability to non-linear robust kernel regression at the cost of a more complex numerical optimization. In particular, we investigate the use of MKL as a tool that allows us to avoid using ad-hoc topographic indices as covariables in statistical models in complex terrains. Instead, MKL learns these relationships from the data in a non-parametric fashion. A study on data simulated from real terrain features confirms the ability of MKL to enhance the interpretability of data-driven models and to aid feature selection without degrading predictive performances. Here we examine the stability of the MKL algorithm with respect to the number of training data samples and to the presence of noise. The results of a real case study are also presented, where MKL is able to exploit a large set of terrain features computed at multiple spatial scales, when predicting mean wind speed in an Alpine region.