Robustness of ecological niche modeling algorithms for mammals in Guyana

The genetic algorithm for rule-set prediction (GARP) has beensuccessfully used in modeling species' distributions with environmental data forwell-studied birds in the United States and elsewhere. GARP's efficiency hasbeen demonstrated to be robust even with incomplete occurrence and geographicdata. Thorough biological sampling in conjunction with comprehensive geographicinformation, however, is not the norm for many tropical areas where mostbiodiversity occurs. Mammals from Guyana were used as a test of the robustnessof these approaches in a worst-case scenario of uneven sampling combined withcoarse geographic data. The occurrence of species in poorly surveyed regions,such as the Pakaraima Highlands of west-central Guyana, was consistentlyunder-predicted, whereas presence in well-surveyed areas such as thesouthwestern Rupununi was usually correctly predicted. Comparisons of numbersof species and specimens collected also indicate that lowland forests in thesoutheast and coastal forests in the northwest are under-sampled. For robustdistributional predictions in Guyana, more thorough inventories are needed inthese diverse environments.     

The Treeterbi and Parallel Treeterbi algorithms: efficient, optimal decoding for ordinary, generalized and pair HMMs

MOTIVATION: Hidden Markov models (HMMs) and generalized HMMs been successfully applied to many problems, but the standard Viterbi algorithm for computing the most probable interpretation of an input sequence (known as decoding) requires memory proportional to the length of the sequence, which can be prohibitive. Existing approaches to reducing memory usage either sacrifice optimality or trade increased running time for reduced memory. RESULTS: We developed two novel decoding algorithms, Treeterbi and Parallel Treeterbi, and implemented them in the TWINSCAN/N-SCAN gene-prediction system. The worst case asymptotic space and time are the same as for standard Viterbi, but in practice, Treeterbi optimally decodes arbitrarily long sequences with generalized HMMs in bounded memory without increasing running time. Parallel Treeterbi uses the same ideas to split optimal decoding across processors, dividing latency to completion by approximately the number of available processors with constant average overhead per processor. Using these algorithms, we were able to optimally decode all human chromosomes with N-SCAN, which increased its accuracy relative to heuristic solutions. We also implemented Treeterbi for Pairagon, our pair HMM based cDNA-to-genome aligner. AVAILABILITY: The TWINSCAN/N-SCAN/PAIRAGON open source software package is available from http://genes.cse.wustl.edu.     

Robustness of cellular functions

Robustness, the ability to maintain performance in the face of perturbations and uncertainty, is a long-recognized key property of living systems. Owing to intimate links to cellular complexity, however, its molecular and cellular basis has only recently begun to be understood. Theoretical approaches to complex engineered systems can provide guidelines for investigating cellular robustness because biology and engineering employ a common set of basic mechanisms in different combinations. Robustness may be a key to understanding cellular complexity, elucidating design principles, and fostering closer interactions between experimentation and theory.     

Robustness of Selection Procedures

In the article Bechhofers Indifference-zone formulation for selecting the t populations with the t highest means is considered in a set of non-normal distributions. Selection rules based on the sample mean, the 10% and the 20% trimmed means, two estimators proposed by Tiku (1981) for valuating the smallest and highest accepted sample values higher, the sample median and a linear combination of quantile estimators, two adaptive procedures and a ranksum procedure are investigated in a large scale simulation experiment in respect of their robustness against deviations from an assumed distribution. Robustness is understood as a small percentage of the difference β_A-β between the actual probability of incorrect selection β_A and the nominal β-value. We obtained a relatively good robustness for the classical sample mean selection rule and useful derivations for the employment of other selection rules in an area of practical importance.     

Robustness of metabolic map reconstruction

With the ever increasing amount of genomic data available, the interest for generating biochemical pathways has grown tremendously. So far, mainly complete genomes have been used to reconstruct the biochemical pathways and their associated interactions. However, a large number of low coverage genomes, as well as other sources of partial genomic data, are currently available for many organisms. In order to be able to use incomplete data for metabolic reconstruction, the inherent properties of this procedure need to be investigated. In this short note, we describe the robustness and predictive power of metabolic reconstructions using partial information from Schizosaccharomyces pombe. We also discuss the implications of the results on reference genome projects as well as other large-scale sequencing data.     

Robustness of optimal mixed strategies

 Mixed strategies, or variable phenotypes, can evolve in fluctuating environments when at the time that a strategy is chosen the consequences of that decision are relatively uncertain. In a previous paper, we have shown several examples of explicit forms of optimal mixed strategies when an environmental distribution and payoff function are given. In many of these examples, the mixed strategy has a continuous distribution. In a recent study, however, Sasaki and Ellner proved that, if the distribution of the environmental parameter is modified in certain ways, the exact ESS distribution becomes discrete rather than continuous. This forces us to take a closer look at the robustness of optimal mixed strategies. In the current paper we prove that such strategies are indeed robust against small perturbations of the environmental distribution and/or the payoff function, in the sense that the optimal strategy distribution for the perturbed system, converges weakly to the optimal strategy distribution for the unperturbed system as the magnitude of the perturbation goes to zero. Furthermore, we show that the fitness difference between the two strategies converges to zero. Thus, although optimal strategies in ‘ideal’ and perturbed systems can be qualitatively different, the difference between the distributions (in a measure theoretic sense) is small. Received: 27 October 1996 / Revised version: 5 March 1997     

Dynamics of intracellular information decoding

A variety of cellular functions are robust even to substantial intrinsic and extrinsic noise in intracellular reactions and the environment that could be strong enough to impair or limit them. In particular, of substantial importance is cellular decision-making in which a cell chooses a fate or behavior on the basis of information conveyed in noisy external signals. For robust decoding, the crucial step is filtering out the noise inevitably added during information transmission. As a minimal and optimal implementation of such an information decoding process, the autocatalytic phosphorylation and autocatalytic dephosphorylation (aPadP) cycle was recently proposed. Here, we analyze the dynamical properties of the aPadP cycle in detail. We describe the dynamical roles of the stationary and short-term responses in determining the efficiency of information decoding and clarify the optimality of the threshold value of the stationary response and its information-theoretical meaning. Furthermore, we investigate the robustness of the aPadP cycle against the receptor inactivation time and intrinsic noise. Finally, we discuss the relationship among information decoding with information-dependent actions, bet-hedging and network modularity.     

Frequency decoding of calcium oscillations

Background

Calcium (Ca^2 +) oscillations are ubiquitous signals present in all cells that provide efficient means to transmit intracellular biological information. Either spontaneously or upon receptor ligand binding, the otherwise stable cytosolic Ca^2 + concentration starts to oscillate. The resulting specific oscillatory pattern is interpreted by intracellular downstream effectors that subsequently activate different cellular processes. This signal transduction can occur through frequency modulation (FM) or amplitude modulation (AM), much similar to a radio signal. The decoding of the oscillatory signal is typically performed by enzymes with multiple Ca^2 + binding residues that diversely can regulate its total phosphorylation, thereby activating cellular program. To date, NFAT, NF-κB, CaMKII, MAPK and calpain have been reported to have frequency decoding properties.

Scope of review

The basic principles and recent discoveries reporting frequency decoding of FM Ca^2 + oscillations are reviewed here.

Major conclusions

A limited number of cellular frequency decoding molecules of Ca^2 + oscillations have yet been reported. Interestingly, their responsiveness to Ca^2 + oscillatory frequencies shows little overlap, suggesting their specific roles in cells.

General significance

Frequency modulation of Ca^2 + oscillations provides an efficient means to differentiate biological responses in the cell, both in health and in disease. Thus, it is crucial to identify and characterize all cellular frequency decoding molecules to understand how cells control important cell programs.     

Robustness and fragility in immunosenescence

We construct a model to study tradeoffs associated with aging in the adaptive immune system, focusing on cumulative effects of replacing naive cells with memory cells. Binding affinities are characterized by a stochastic shape space model. System loss arising from an individual infection is associated with disease severity, as measured by the total antigen population over the course of an infection. We monitor evolution of cell populations on the shape space over a string of infections, and find that the distribution of losses becomes increasingly heavy-tailed with time. Initially this lowers the average loss: the memory cell population becomes tuned to the history of past exposures, reducing the loss of the system when subjected to a second, similar infection. This is accompanied by a corresponding increase in vulnerability to novel infections, which ultimately causes the expected loss to increase due to overspecialization, leading to increasing fragility with age (i.e., immunosenescence). In our model, immunosenescence is not the result of a performance degradation of some specific lymphocyte, but rather a natural consequence of the built-in mechanisms for system adaptation. This “robust, yet fragile” behavior is a key signature of Highly Optimized Tolerance.     

Robustness Analysis of Stochastic Biochemical Systems

We propose a new framework for rigorous robustness analysis of stochastic biochemical systems that is based on probabilistic model checking techniques. We adapt the general definition of robustness introduced by Kitano to the class of stochastic systems modelled as continuous time Markov Chains in order to extensively analyse and compare robustness of biological models with uncertain parameters. The framework utilises novel computational methods that enable to effectively evaluate the robustness of models with respect to quantitative temporal properties and parameters such as reaction rate constants and initial conditions. We have applied the framework to gene regulation as an example of a central biological mechanism where intrinsic and extrinsic stochasticity plays crucial role due to low numbers of DNA and RNA molecules. Using our methods we have obtained a comprehensive and precise analysis of stochastic dynamics under parameter uncertainty. Furthermore, we apply our framework to compare several variants of two-component signalling networks from the perspective of robustness with respect to intrinsic noise caused by low populations of signalling components. We have successfully extended previous studies performed on deterministic models (ODE) and showed that stochasticity may significantly affect obtained predictions. Our case studies demonstrate that the framework can provide deeper insight into the role of key parameters in maintaining the system functionality and thus it significantly contributes to formal methods in computational systems biology.     

The Quantitative Genetics of Phenotypic Robustness

Phenotypic robustness, or canalization, has been extensively investigated both experimentally and theoretically. However, it remains unknown to what extent robustness varies between individuals, and whether factors buffering environmental variation also buffer genetic variation. Here we introduce a quantitative genetic approach to these issues, and apply this approach to data from three species. In mice, we find suggestive evidence that for hundreds of gene expression traits, robustness is polymorphic and can be genetically mapped to discrete genomic loci. Moreover, we find that the polymorphisms buffering genetic variation are distinct from those buffering environmental variation. In fact, these two classes have quite distinct mechanistic bases: environmental buffers of gene expression are predominantly sex-specific and trans-acting, whereas genetic buffers are not sex-specific and often cis-acting. Data from studies of morphological and life-history traits in plants and yeast support the distinction between polymorphisms buffering genetic and environmental variation, and further suggest that loci buffering different types of environmental variation do overlap with one another. These preliminary results suggest that naturally occurring polymorphisms affecting phenotypic robustness could be abundant, and that these polymorphisms may generally buffer either genetic or environmental variation, but not both.     

Robustness of Massively Parallel Sequencing Platforms

The improvements in high throughput sequencing technologies (HTS) made clinical sequencing projects such as ClinSeq and Genomics England feasible. Although there are significant improvements in accuracy and reproducibility of HTS based analyses, the usability of these types of data for diagnostic and prognostic applications necessitates a near perfect data generation. To assess the usability of a widely used HTS platform for accurate and reproducible clinical applications in terms of robustness, we generated whole genome shotgun (WGS) sequence data from the genomes of two human individuals in two different genome sequencing centers. After analyzing the data to characterize SNPs and indels using the same tools (BWA, SAMtools, and GATK), we observed significant number of discrepancies in the call sets. As expected, the most of the disagreements between the call sets were found within genomic regions containing common repeats and segmental duplications, albeit only a small fraction of the discordant variants were within the exons and other functionally relevant regions such as promoters. We conclude that although HTS platforms are sufficiently powerful for providing data for first-pass clinical tests, the variant predictions still need to be confirmed using orthogonal methods before using in clinical applications.     

Robustness of Sequential Tests for Means

Three sequential tests of the null hypothesis, μ = μ₀, versus an alternative of the form (μ ? μ₀)²= =σ²d² are compared for different forms of violation of the underlying normal assumption. 10000 samples were simulated for d=0.6; 1.0 and 1.6 (σ² = 1), four (α, β)-combinations and seven alternative distributions. The results show that for small d-values one test is robust for α and another for β. For large d all tests can be used.     

Robustness analysis of biochemical network models

Biological systems that have been experimentally verified to be robust to significant changes in their environments require mathematical models that are themselves robust. In this context, a necessary condition for model robustness is that the model dynamics should not be sensitive to small variations in the model's parameters. Robustness analysis problems of this type have been extensively studied in the field of robust control theory and have been found to be very difficult to solve in general. The authors describe how some tools from robust control theory and nonlinear optimisation can be used to analyse the robustness of a recently proposed model of the molecular network underlying adenosine 3',5'-cyclic monophosphate (cAMP) oscillations observed in fields of chemotactic Dictyostelium cells. The network model, which consists of a system of seven coupled nonlinear differential equations, accurately reproduces the spontaneous oscillations in cAMP observed during the early development of D. discoideum. The analysis by the authors reveals, however, that very small variations in the model parameters can effectively destroy the required oscillatory dynamics. A biological interpretation of the analysis results is that correct functioning of a particular positive feedback loop in the proposed model is crucial to maintaining the required oscillatory dynamics.