Statistical and Combinatorial Aspects of Comparative Genomics*

Abstract. This document presents a survey of the statistical and combinatorial aspects of four areas of comparative genomics: gene order based measures of evolutionary distances between species, construction of phylogenetic trees, detection of horizontal transfer of genes, and detection of ancient whole genome duplications.     

Modeling service-time distributions for queueing network simulation

Shapes of service-time distributions in queueing network models have a great impact on the distribution of system response-times. It is essential for the analysis of response-time distribution that the modeled service-time distributions have the correct shape. Tradionally modeling of service-time distributions is based on a parametric approach by assuming a specific distribution and estimating its parameters. We introduce an alternative approach based on the principles of exploratory data analysis and nonparametric data modeling. The proposed method applies nonlinear data transformation and resistant curve fitting. The method can be used in cases, where the available data is a complete sample, a histogram, or the mean and a set of 5-10 quantiles. The reported results indicate that the proposed method is able to approximate the distribution of measured service times so that accurate estimates for quantiles of the response-time distribution are obtained     

Health Effects Caused by Primary Fine Particulate Matter (PM2.5) Emitted from Buses in the Helsinki Metropolitan Area, Finland

Fine particle (PM(2.5)) emissions from traffic have been associated with premature mortality. The current work compares PM(2.5)-induced mortality in alternative public bus transportation strategies as being considered by the Helsinki Metropolitan Area Council, Finland. The current bus fleet and transportation volume is compared to four alternative hypothetical bus fleet strategies for the year 2020: (1) the current bus fleet for 2020 traffic volume, (2) modern diesel buses without particle traps, (3) diesel buses with particle traps, and (4) buses using natural gas engines. The average population PM(2.5) exposure level attributable to the bus emissions was determined for the 1996-1997 situation using PM(2.5) exposure measurements including elemental composition from the EXPOLIS-Helsinki study and similar element-based source apportionment of ambient PM(2.5) concentrations observed in the ULTRA study. Average population exposure to particles originating from the bus traffic in the year 2020 is assumed to be proportional to the bus emissions in each strategy. Associated mortality was calculated using dose-response relationships from two large cohort studies on PM(2.5) mortality from the United States. Estimated number of deaths per year (90% confidence intervals in parenthesis) associated with primary PM(2.5) emissions from buses in Helsinki Metropolitan Area in 2020 were 18 (0-55), 9 (0-27), 4 (0-14), and 3 (0-8) for the strategies 1-4, respectively. The relative differences in the associated mortalities for the alternative strategies are substantial, but the number of deaths in the lowest alternative, the gas buses, is only marginally lower than what would be achieved by diesel engines equipped with particle trap technology. The dose-response relationship and the emission factors were identified as the main sources of uncertainty in the model.     

Talking with like-minded people—Equality and efficacy in enclave deliberation

Enclaves of like-minded people are often seen as problematic from a democratic point of view, as they have been found to lead to both group polarization and an amplification of cognitive errors. Nevertheless, enclaves can also act as protected spaces have the opportunity to discuss politics with their peers. As a result, people who are less well-endowed to face political disagreement can find it easier to engage in politics. In order to study the ‘empowering’ potential of enclave deliberation, we use data from a population-based experiment (n = 207). The participants were randomly allocated to two treatments. Some participants deliberated in groups consisting of people with similar baseline views on immigration (like-minded treatment), whereas others deliberated in groups where both restrictive and permissive participants were present (mixed treatment). We hypothesize that (1) discussion in like-minded groups is more equal than in mixed opinion groups and that (2) participants with lower resources feel politically more efficacious after deliberation in like-minded than in mixed groups. Our results suggest that people with higher resources tend to be more active regardless of treatment. Nevertheless, we also find that among those with lower resources deliberation in like-minded groups generates a higher sense of equality than discussion in mixed opinion groups.     

A comparative study of social network models: Network evolution models and nodal attribute models

This paper reviews, classifies and compares recent models for social networks that have mainly been published within the physics-oriented complex networks literature. The models fall into two categories: those in which the addition of new links is dependent on the (typically local) network structure (network evolution models, NEMs), and those in which links are generated based only on nodal attributes (nodal attribute models, NAMs). An exponential random graph model (ERGM) with structural dependencies is included for comparison. We fit models from each of these categories to two empirical acquaintance networks with respect to basic network properties. We compare higher order structures in the resulting networks with those in the data, with the aim of determining which models produce the most realistic network structure with respect to degree distributions, assortativity, clustering spectra, geodesic path distributions, and community structure (subgroups with dense internal connections). We find that the nodal attribute models successfully produce assortative networks and very clear community structure. However, they generate unrealistic clustering spectra and peaked degree distributions that do not match empirical data on large social networks. On the other hand, many of the network evolution models produce degree distributions and clustering spectra that agree more closely with data. They also generate assortative networks and community structure, although often not to the same extent as in the data. The ERGM model, which turned out to be near-degenerate in the parameter region best fitting our data, produces the weakest community structure.     

Likelihood-free inference via classification

Increasingly complex generative models are being used across disciplines as they allow for realistic characterization of data, but a common difficulty with them is the prohibitively large computational cost to evaluate the likelihood function and thus to perform likelihood-based statistical inference. A likelihood-free inference framework has emerged where the parameters are identified by finding values that yield simulated data resembling the observed data. While widely applicable, a major difficulty in this framework is how to measure the discrepancy between the simulated and observed data. Transforming the original problem into a problem of classifying the data into simulated versus observed, we find that classification accuracy can be used to assess the discrepancy. The complete arsenal of classification methods becomes thereby available for inference of intractable generative models. We validate our approach using theory and simulations for both point estimation and Bayesian inference, and demonstrate its use on real data by inferring an individual-based epidemiological model for bacterial infections in child care centers.     

Local dimension reduction of summary statistics for likelihood-free inference

Statistics and Computing - Approximate Bayesian computation (ABC) and other likelihood-free inference methods have gained popularity in the last decade, as they allow rigorous statistical inference...     

A lay-statistician explanation of minority discrimination

We outline a new explanation of discrimination against numerical minorities. In contrast to prior work that focuses on how the content of categories affects discrimination, our argument describes how the size of categories leads to discrimination. Specifically, we argue that, when comparing multiple categories, actors tend to view larger categories as more closely approximating an underlying population than smaller ones. As a result, a decision maker will tend to expect that members of a numerical majority are more likely to be what he/she is searching for, whether it is the best or worst candidate. We report the results of two studies designed to test these arguments. To demonstrate the generality of the proposed mechanism, Study 1 tested the argument in a non-social domain. Participants disproportionately favored the majority (vs. minority) category when searching for a single winning lottery ticket, and favored the minority category when the goal was to avoid a single losing ticket. Our second study supported an additional implication of the argument in a social domain: decision makers tended to rank highly qualified majority job candidates as better than equally qualified minority candidates, and relatively unqualified majority candidates as worse than equally unqualified minority candidates.     

Solving Survo puzzles using matrix combinatorial products

We investigate combinatorial matrix problems that are related to restricted integer partitions. They arise from Survo puzzles, where the basic task is to fill an m×n table by integers 1, 2,?…?, mn, so that each number appears only once, when the column sums and the row sums are fixed. We present a new computational method for solving Survo puzzles with binary matrices that are recoded and combined using the Hadamard, Kronecker, and Khatri–Rao products. The idea of our method is based on using the matrix interpreter and other data analytic tools of Survo R, which represents the newest generation of the Survo computing environment, recently implemented as a multiplatform, open source R package. We illustrate our method with detailed examples.