Integration of software tools for integrative modeling of biomolecular systems

Integrative modeling computes a model based on varied types of input information, be it from experiments or prior models. Often, a type of input information will be best handled by a specific modeling software package. In such a case, we desire to integrate our integrative modeling software package, Integrative Modeling Platform (IMP), with software specialized to the computational demands of the modeling problem at hand. After several attempts, however, we have concluded that even in collaboration with the software’s developers, integration is either impractical or impossible. The reasons for the intractability of integration include software incompatibilities, differing modeling logic, the costs of collaboration, and academic incentives. In the integrative modeling software ecosystem, several large modeling packages exist with often redundant tools. We reason, therefore, that the other development groups have similarly concluded that the benefit of integration does not justify the cost. As a result, modelers are often restricted to the set of tools within a single software package. The inability to integrate tools from distinct software negatively impacts the quality of the models and the efficiency of the modeling. As the complexity of modeling problems grows, we seek to galvanize developers and modelers to consider the long-term benefit that software interoperability yields. In this article, we formulate a demonstrative set of software standards for implementing a model search using tools from independent software packages and discuss our efforts to integrate IMP and the crystallography suite Phenix within the Bayesian modeling framework.     

Statistical techniques for modeling of <Emphasis Type="Italic">Corylus</Emphasis>, <Emphasis Type="Italic">Alnus</Emphasis>, and <Emphasis Type="Italic">Betula</Emphasis> pollen concentration in the air

Prediction of allergic pollen concentration is one of the most important goals of aerobiology. Past studies have used a broad range of modeling techniques; however, the results cannot be directly compared owing to the use of different datasets, validation methods, and evaluation metrics. The main aim of this study was to compare nine statistical modeling techniques using the same dataset. An additional goal was to assess the importance of predictors for the best model. Aerobiological data for Corylus, Alnus, and Betula pollen counts were obtained from nine cities in Poland and covered between five and 16 years of measurements. Meteorological data from the AGRI4CAST project were used as a predictor variables. The results of 243 final models (3 taxa \(\times\)  9 cities \(\times\) 9 techniques) were validated using a repeated k-fold cross-validation and compared using relative and absolute performance statistics. Afterward, the variable importance of predictors in the best models was calculated and compared. Simple models performed poorly. On the other hand, regression trees and rule-based models proved to be the most accurate for all of the taxa. Cumulative growing degree days proved to be the single most important predictor variable in the random forest models of Corylus, Alnus, and Betula. Finally, the study suggested potential improvements in aerobiological modeling, such as the application of robust cross-validation techniques and the use of gridded variables.     

Integrating expert knowledge and ecological niche models to estimate Mexican primates’ distribution

Ecological niche modeling is used to estimate species distributions based on occurrence records and environmental variables, but it seldom includes explicit biotic or historical factors that are important in determining the distribution of species. Expert knowledge can provide additional valuable information regarding ecological or historical attributes of species, but the influence of integrating this information in the modeling process has been poorly explored. Here, we integrated expert knowledge in different stages of the niche modeling process to improve the representation of the actual geographic distributions of Mexican primates (Ateles geoffroyi, Alouatta pigra, and A. palliata mexicana). We designed an elicitation process to acquire information from experts and such information was integrated by an iterative process that consisted of reviews of input data by experts, production of ecological niche models (ENMs), and evaluation of model outputs to provide feedback. We built ENMs using the maximum entropy algorithm along with a dataset of occurrence records gathered from a public source and records provided by the experts. Models without expert knowledge were also built for comparison, and both models, with and without expert knowledge, were evaluated using four validation metrics that provide a measure of accuracy for presence-absence predictions (specificity, sensitivity, kappa, true skill statistic). Integrating expert knowledge to build ENMs produced better results for potential distributions than models without expert knowledge, but a much greater improvement in the transition from potential to realized geographic distributions by reducing overprediction, resulting in better representations of the actual geographic distributions of species. Furthermore, with the combination of niche models and expert knowledge we were able to identify an area of sympatry between A. palliata mexicana and A. pigra. We argue that the inclusion of expert knowledge at different stages in the construction of niche models in an explicit and systematic fashion is a recommended practice as it produces overall positive results for representing realized species distributions.     

思茅松天然林树冠结构模型

以云南省普洱市思茅区思茅松天然林为研究对象,采用枝解析调查了34株思茅松样木的树冠数据,分析了一级枝枝长、枝径、着枝角度、弦长和树冠半径5个树冠形状变量的变化规律,分别构建其预估模型;分析了树冠结构变化,分别构建了一级枝轮枝高度预估模型、一级枝枝条数量预估模型和一级枝枝条数量累积预估模型,并采用独立样本进行模型统计精度检验。结果表明:8个预估模型的预测效果良好,精度达到91%以上,尤其是一级枝着枝角度模型和一级枝轮枝高度模型预测精度达到97%以上。研究结果合理准确描述思茅松树冠结构的变化,为思茅松天然林的经营管理提供科学依据。     

Maxent modeling for predicting the potential distribution of Sanghuang,an important group of medicinal fungi in China

The ability to identify the spatial distribution of economically important fungal species is crucial for understanding the environmental factors that affect them and for conservation management. A potentially valuable approach for this is maximum entropy (Maxent) spatial distribution modeling, which was applied here to map the potential distribution of three “Sanghuang” mushrooms in China, which include Phellinus baumii, Phellinus igniarius and Phellinus vaninii. Nineteen WorldClim bioclimatic variables, with corresponding altitude data, and 89 spatially well-dispersed species occurrence records were used in the modeling. The relative importance of the environmental variables was evaluated by Jackknife tests in the modeling analysis. The maximum entropy models obtained have high Area Under Receiver Operating Characteristic Curve (AUC) values: 0.956, 0.967 and 0.960, for P. baumii, P. igniarius and P. vaninii, respectively. The bioclimatic variable that most strongly affected distributions of P. baumii and P. vaninii was precipitation in the warmest quarter, while the mean temperature in the warmest quarter affected the distribution of P. igniarius most strongly. Overall, these models could provide valuable help in searching for the target species in areas where it is hitherto unknown, and be the reference of conservation measures for these medicinal fungal species.     

A Generalized Radiation Model for Human Mobility: Spatial Scale,Searching Direction and Trip Constraint

We generalized the recently introduced “radiation model”, as an analog to the generalization of the classic “gravity model”, to consolidate its nature of universality for modeling diverse mobility systems. By imposing the appropriate scaling exponent λ, normalization factor κ and system constraints including searching direction and trip OD constraint, the generalized radiation model accurately captures real human movements in various scenarios and spatial scales, including two different countries and four different cities. Our analytical results also indicated that the generalized radiation model outperformed alternative mobility models in various empirical analyses.     

On Finding and Using Identifiable Parameter Combinations in Nonlinear Dynamic Systems Biology Models and COMBOS: A Novel Web Implementation

Parameter identifiability problems can plague biomodelers when they reach the quantification stage of development, even for relatively simple models. Structural identifiability (SI) is the primary question, usually understood as knowing which of P unknown biomodel parameters p ₁,…, p_i,…, p_P are-and which are not-quantifiable in principle from particular input-output (I-O) biodata. It is not widely appreciated that the same database also can provide quantitative information about the structurally unidentifiable (not quantifiable) subset, in the form of explicit algebraic relationships among unidentifiable p_i. Importantly, this is a first step toward finding what else is needed to quantify particular unidentifiable parameters of interest from new I–O experiments. We further develop, implement and exemplify novel algorithms that address and solve the SI problem for a practical class of ordinary differential equation (ODE) systems biology models, as a user-friendly and universally-accessible web application (app)–COMBOS. Users provide the structural ODE and output measurement models in one of two standard forms to a remote server via their web browser. COMBOS provides a list of uniquely and non-uniquely SI model parameters, and–importantly-the combinations of parameters not individually SI. If non-uniquely SI, it also provides the maximum number of different solutions, with important practical implications. The behind-the-scenes symbolic differential algebra algorithms are based on computing Gröbner bases of model attributes established after some algebraic transformations, using the computer-algebra system Maxima. COMBOS was developed for facile instructional and research use as well as modeling. We use it in the classroom to illustrate SI analysis; and have simplified complex models of tumor suppressor p53 and hormone regulation, based on explicit computation of parameter combinations. It’s illustrated and validated here for models of moderate complexity, with and without initial conditions. Built-in examples include unidentifiable 2 to 4-compartment and HIV dynamics models.     

A Presence-Only Model of Suitable Roosting Habitat for the Endangered Indiana Bat in the Southern Appalachians

We know little about how forest bats, which are cryptic and mobile, use roosts on a landscape scale. For widely distributed species like the endangered Indiana bat Myotis sodalis, identifying landscape-scale roost habitat associations will be important for managing the species in different regions where it occurs. For example, in the southern Appalachian Mountains, USA, M. sodalis roosts are scattered across a heavily forested landscape, which makes protecting individual roosts impractical during large-scale management activities. We created a predictive spatial model of summer roosting habitat to identify important predictors using the presence-only modeling program MaxEnt and an information theoretic approach for model comparison. Two of 26 candidate models together accounted for >0.93 of AICc weights. Elevation and forest type were top predictors of presence; aspect north/south and distance-to-ridge were also important. The final average best model indicated that 5% of the study area was suitable habitat and 0.5% was optimal. This model matched our field observations that, in the southern Appalachian Mountains, optimal roosting habitat for M. sodalis is near the ridge top in south-facing mixed pine-hardwood forests at elevations from 260–575 m. Our findings, coupled with data from other studies, suggest M. sodalis is flexible in roost habitat selection across different ecoregions with varying topography and land use patterns. We caution that, while mature pine-hardwood forests are important now, specific areas of suitable and optimal habitat will change over time. Combining the information theoretic approach with presence-only models makes it possible to develop landscape-scale habitat suitability maps for forest bats.     

Neglected and emerging fungal infections: review of hyalohyphomycosis by Paecilomyces lilacinus focusing in disease burden, in vitro antifungal susceptibility and management

Paecilomyces lilacinus is an emerging pathogenic fungus that can cause different clinical manifestations ranging from cutaneous and sub-cutaneous infections to severe oculomycosis. This review discusses infections caused by P. lilacinus, as well as their symptoms and correlates of immune responses, morphological characteristics of the fungus, therapies, in vitro susceptibility tests, laboratory diagnosis and the experimental models available.     

Growth Simulation and Discrimination of Botrytis cinerea,Rhizopus stolonifer and Colletotrichum acutatum Using Hyperspectral Reflectance Imaging

This research aimed to develop a rapid and nondestructive method to model the growth and discrimination of spoilage fungi, like Botrytis cinerea, Rhizopus stolonifer and Colletotrichum acutatum, based on hyperspectral imaging system (HIS). A hyperspectral imaging system was used to measure the spectral response of fungi inoculated on potato dextrose agar plates and stored at 28°C and 85% RH. The fungi were analyzed every 12 h over two days during growth, and optimal simulation models were built based on HIS parameters. The results showed that the coefficients of determination (R²) of simulation models for testing datasets were 0.7223 to 0.9914, and the sum square error (SSE) and root mean square error (RMSE) were in a range of 2.03–53.40×10⁻⁴ and 0.011–0.756, respectively. The correlation coefficients between the HIS parameters and colony forming units of fungi were high from 0.887 to 0.957. In addition, fungi species was discriminated by partial least squares discrimination analysis (PLSDA), with the classification accuracy of 97.5% for the test dataset at 36 h. The application of this method in real food has been addressed through the analysis of Botrytis cinerea, Rhizopus stolonifer and Colletotrichum acutatum inoculated in peaches, demonstrating that the HIS technique was effective for simulation of fungal infection in real food. This paper supplied a new technique and useful information for further study into modeling the growth of fungi and detecting fruit spoilage caused by fungi based on HIS.     

Metabolite-centric approaches for the discovery of antibacterials using genome-scale metabolic networks

Development of genome-scale metabolic models and various constraints-based flux analyses have enabled more sophisticated examination of metabolism. Recently reported metabolite essentiality studies are also based on the constraints-based modeling, but approaches metabolism from a metabolite-centric perspective, providing synthetic lethal combination of reactions and clues for the rational discovery of antibacterials. In this study, metabolite essentiality analysis was applied to the genome-scale metabolic models of four microorganisms: Escherichia coli, Helicobacter pylori, Mycobacterium tuberculosis and Staphylococcus aureus. Furthermore, chokepoints, metabolites surrounded by enzymes that uniquely consume and/or produce them, were also calculated based on the network properties of the above organisms. A systematic drug targeting strategy was developed by combining information from these two methods. Final drug target metabolites are presented and examined with knowledge from the literature.     

Kinetics of Bifidobacterium longum ATCC 15707 growth

This study presents an analysis of a Bifidobacterium longum ATCC 15707 optimization study. Kinetic growth models were fitted to cultivations from a central composite circumscribed (CCC) experiment design for three variables (temperature as well as glucose and yeast extract concentration). The parameters of these kinetic models, μ_max, K_S, Y_XS and m_S, were used as responses of the experiment design. This novel concept of combining optimization and modeling presented slightly different optimal conditions for B. longum growth from the original optimization study. However, the optimum of this study could be based on more scientific arguments. The parameters of the kinetic model represent the physiological effects that the cultivation parameters impose on the organism. A difference was observed in the optimum of the initial glucose concentration, which was originally thought to benefit process efficiency. This re-analysis showed that it is better for all aspects of cell physiology (substrate efficiency and growth rate) to use a lower initial glucose concentration.     

Effect of inventory method on niche models: Random versus systematic error

Data from large-scale biological inventories are essential for understanding and managing Earth's ecosystems. The Forest Inventory and Analysis Program (FIA) of the U.S. Forest Service is the largest biological inventory in North America; however, the FIA inventory recently changed from an amalgam of different approaches to a nationally-standardized approach in 2000. Full use of both data sets is clearly warranted to target many pressing research questions including those related to climate change and forest resources. However, full use requires lumping FIA data from different regionally-based designs (pre-2000) and/or lumping the data across the temporal changeover. Combining data from different inventory types must be approached with caution as inventory types represent different probabilities of detecting trees per sample unit, which can ultimately confound temporal and spatial patterns found in the data. Consequently, the main goal of this study is to evaluate the effect of inventory on a common analysis in ecology, modeling of climatic niches (or species-climate relations). We use non-parametric multiplicative regression (NPMR) to build and compare niche models for 41 tree species from the old and new FIA design in the Pacific coastal United States. We discover two likely effects of inventory on niche models and their predictions. First, there is an increase from 4 to 6% in random error for modeled predictions from the different inventories when compared to modeled predictions from two samples of the same inventory. Second, systematic error (or directional disagreement among modeled predictions) is detectable for 4 out of 41 species among the different inventories: Calocedrus decurrens, Pseudotsuga menziesii, and Pinus ponderosa, and Abies concolor. Hence, at least 90% of niche models and predictions of probability of occurrence demonstrate no obvious effect from the change in inventory design. Further, niche models built from sub-samples of the same data set can yield systematic error that rivals systematic error in predictions for models built from two separate data sets. This work corroborates the pervasive and pressing need to quantify different types of error in niche modeling to address issues associated with data quality and large-scale data integration.     

Comparison of Models for IP3 Receptor Kinetics Using Stochastic Simulations

Inositol 1,4,5-trisphosphate receptor (IP₃R) is a ubiquitous intracellular calcium (Ca²⁺) channel which has a major role in controlling Ca²⁺ levels in neurons. A variety of computational models have been developed to describe the kinetic function of IP₃R under different conditions. In the field of computational neuroscience, it is of great interest to apply the existing models of IP₃R when modeling local Ca²⁺ transients in dendrites or overall Ca²⁺ dynamics in large neuronal models. The goal of this study was to evaluate existing IP₃R models, based on electrophysiological data. This was done in order to be able to suggest suitable models for neuronal modeling. Altogether four models (Othmer and Tang, 1993; Dawson et al., 2003; Fraiman and Dawson, 2004; Doi et al., 2005) were selected for a more detailed comparison. The selection was based on the computational efficiency of the models and the type of experimental data that was used in developing the model. The kinetics of all four models were simulated by stochastic means, using the simulation software STEPS, which implements the Gillespie stochastic simulation algorithm. The results show major differences in the statistical properties of model functionality. Of the four compared models, the one by Fraiman and Dawson (2004) proved most satisfactory in producing the specific features of experimental findings reported in literature. To our knowledge, the present study is the first detailed evaluation of IP₃R models using stochastic simulation methods, thus providing an important setting for constructing a new, realistic model of IP₃R channel kinetics for compartmental modeling of neuronal functions. We conclude that the kinetics of IP₃R with different concentrations of Ca²⁺ and IP₃ should be more carefully addressed when new models for IP₃R are developed.     

Sharing detection heterogeneity information among species in community models of occupancy and abundance can strengthen inference

1. The estimation of abundance and distribution and factors governing patterns in these parameters is central to the field of ecology. The continued development of hierarchical models that best utilize available information to inform these processes is a key goal of quantitative ecologists. However, much remains to be learned about simultaneously modeling true abundance, presence, and trajectories of ecological communities.
2. Simultaneous modeling of the population dynamics of multiple species provides an interesting mechanism to examine patterns in community processes and, as we emphasize herein, to improve species‐specific estimates by leveraging detection information among species. Here, we demonstrate a simple but effective approach to share information about observation parameters among species in hierarchical community abundance and occupancy models, where we use shared random effects among species to account for spatiotemporal heterogeneity in detection probability.
3. We demonstrate the efficacy of our modeling approach using simulated abundance data, where we recover well our simulated parameters using N‐mixture models. Our approach substantially increases precision in estimates of abundance compared with models that do not share detection information among species. We then expand this model and apply it to repeated detection/non‐detection data collected on six species of tits (Paridae) breeding at 119 1 km² sampling sites across a P. montanus hybrid zone in northern Switzerland (2004–2020). We find strong impacts of forest cover and elevation on population persistence and colonization in all species. We also demonstrate evidence for interspecific competition on population persistence and colonization probabilities, where the presence of marsh tits reduces population persistence and colonization probability of sympatric willow tits, potentially decreasing gene flow among willow tit subspecies.
4. While conceptually simple, our results have important implications for the future modeling of population abundance, colonization, persistence, and trajectories in community frameworks. We suggest potential extensions of our modeling in this paper and discuss how leveraging data from multiple species can improve model performance and sharpen ecological inference.

     

Assessment and improvement of the appropriateness of an LCI data set on a system level – application to textile manufacturing

Purpose

This paper aims at assessing the appropriateness at the system level of different Life Cycle Inventory (LCI) data sets (including default models) selected by the Life Cycle Assessment (LCA) practitioner. This means that the uncertainty measurements are applied on some specific main parameters of the LCI data set instead of measured input values. This approach aims at providing a pragmatic method to approximate and reduce the uncertainties resulting from a lack of information on a specific step.

Methods

The method proposed in this paper to assess the percentage errors on appropriateness includes three main steps. First, different systems including different versions of the same process with technological or geographical changes are assessed. Second, a hierarchical cluster analysis (HCA) or a principal component analysis (PCA) is performed to identify the main variables influencing the results. Third, a multivariate analysis of variances (MANOVA) assesses the significance of the main variables on the results. An appropriate default model can be developed by setting the variables introducing high variations in results.

Results and discussion

When comparing the same spinning process located in different countries, the HCA enabled us to identify the electricity mix as the main variable influencing the results. The “world average default models” has proven inappropriate to represent country specific locations. When comparing spinning technologies, the PCAs identified the electricity and the cotton input required as the main variables influencing the results and explained the variations in results due to technological changes. The HCA performed on different yarn manufacturing procedures identified the location and the yarn thickness as the two main variables influencing the results. The MANOVA assessed the percentage marginal variance (PMV) explained by the variable location between 85 and 92 % for four impact categories. The MANOVA performed on different fabric manufacturing systems assessed the PMV explained by the variables harvest, spinning, and weaving locations above 68 % for all impact categories. The HCA and MANOVA analyses helped design an appropriate “technological average default model.”

Conclusions

From the identification of the main influencing variables (HCA and PCA) to the quantitative appropriateness assessment (MANOVA) and the development of appropriate default models, the method has proven effective in assisting the LCA practitioner in the modeling of textile manufacturing systems, and for other worldwide multi-step manufacturing systems.     

On the dangers of making scientific models ontologically independent: taking Richard Levins’ warnings seriously

Levins and Lewontin have contributed significantly to our philosophical understanding of the structures, processes, and purposes of biological mathematical theorizing and modeling. Here I explore their separate and joint pleas to avoid making abstract and ideal scientific models ontologically independent by confusing or conflating our scientific models and the world. I differentiate two views of theorizing and modeling, orthodox and dialectical, in order to examine Levins and Lewontin’s, among others, advocacy of the latter view. I compare the positions of these two views with respect to four points regarding ontological assumptions: (1) the origin of ontological assumptions, (2) the relation of such assumptions to the formal models of the same theory, (3) their use in integrating and negotiating different formal models of distinct theories, and (4) their employment in explanatory activity. Dialectical is here used in both its Hegelian–Marxist sense of opposition and tension between alternative positions and in its Platonic sense of dialogue between advocates of distinct theories. I investigate three case studies, from Levins and Lewontin as well as from a recent paper of mine, that show the relevance and power of the dialectical understanding of theorizing and modeling.     

The neural binding problem(s)

The famous Neural Binding Problem (NBP) comprises at least four distinct problems with different computational and neural requirements. This review discusses the current state of work on General Coordination, Visual Feature-Binding, Variable Binding, and the Subjective Unity of Perception. There is significant continuing progress, partially masked by confusing the different versions of the NBP.     

Variance component analysis of a multi-site study for the reproducibility of multiple reaction monitoring measurements of peptides in human plasma

Background

In the Addona et al. paper (Nature Biotechnology 2009), a large-scale multi-site study was performed to quantify Multiple Reaction Monitoring (MRM) measurements of proteins spiked in human plasma. The unlabeled signature peptides derived from the seven target proteins were measured at nine different concentration levels, and their isotopic counterparts were served as the internal standards.

Methodology/Principal Findings

In this paper, the sources of variation are analyzed by decomposing the variance into parts attributable to specific experimental factors: technical replicates, sites, peptides, transitions within each peptide, and higher-order interaction terms based on carefully built mixed effects models. The factors of peptides and transitions are shown to be major contributors to the variance of the measurements considering heavy (isotopic) peptides alone. For the light (¹²C) peptides alone, in addition to these factors, the factor of study*peptide also contributes significantly to the variance of the measurements. Heterogeneous peptide component models as well as influence analysis identify the outlier peptides in the study, which are then excluded from the analysis. Using a log-log scale transformation and subtracting the heavy/isotopic peptide [internal standard] measurement from the peptide measurements (i.e., taking the logarithm of the peak area ratio in the original scale establishes that), the MRM measurements are overall consistent across laboratories following the same standard operating procedures, and the variance components related to sites, transitions and higher-order interaction terms involving sites have greatly reduced impact. Thus the heavy peptides have been effective in reducing apparent inter-site variability. In addition, the estimates of intercepts and slopes of the calibration curves are calculated for the sub-studies.

Conclusions/Significance

The MRM measurements are overall consistent across laboratories following the same standard operating procedures, and heavy peptides can be used as an effective internal standard for reducing apparent inter-site variability. Mixed effects modeling is a valuable tool in mass spectrometry-based proteomics research.