Multisensor Data Fusion for Localization of Pollution Sources in Wastewater Networks

Illegal discharges of pollutants into sewage networks are a growing problem in large European cities. Such events often require restarting wastewater treatment plants, which cost up to a hundred thousand Euros. A system for localization and quantification of pollutants in utility networks could discourage such behavior and indicate a culprit if it happens. We propose an enhanced algorithm for multisensor data fusion for the detection, localization, and quantification of pollutants in wastewater networks. The algorithm processes data from multiple heterogeneous sensors in real-time, producing current estimates of network state and alarms if one or many sensors detect pollutants. Our algorithm models the network as a directed acyclic graph, uses adaptive peak detection, estimates the amount of specific compounds, and tracks the pollutant using a Kalman filter. We performed numerical experiments for several real and artificial sewage networks, and measured the quality of discharge event reconstruction. We report the correctness and performance of our system. We also propose a method to assess the importance of specific sensor locations. The experiments show that the algorithm’s success rate is equal to sensor coverage of the network. Moreover, the median distance between nodes pointed out by the fusion algorithm and nodes where the discharge was introduced equals zero when more than half of the network nodes contain sensors. The system can process around 5000 measurements per second, using 1 MiB of memory per 4600 measurements plus a constant of 97 MiB, and it can process 20 tracks per second, using 1.3 MiB of memory per 100 tracks.

Atom Probe Tomography for Isotopic Analysis: Development of the 34S/32S System in Sulfides

Using a combination of simulated data and pyrite isotopic reference materials, we have refined a methodology to obtain quantitative δ34S measurements from atom probe tomography (APT) datasets. This study builds on previous attempts to characterize relative 34S/32S ratios in gold-containing pyrite using APT. We have also improved our understanding of the artifacts inherent in laser-pulsed APT of insulators. Specifically, we find the probability of multi-hit detection events increases during the APT experiment, which can have a detrimental effect on the accuracy of the analysis. We demonstrate the use of standardized corrected time-of-flight single-hit data for our isotopic analysis. Additionally, we identify issues with the standard methods of extracting background-corrected counts from APT mass spectra. These lead to inaccurate and inconsistent isotopic analyses due to human variability in peak ranging and issues with background correction algorithms. In this study, we use the corrected time-of-flight single-hit data, an adaptive peak fitting algorithm, and an improved deconvolution algorithm to extract 34S/32S ratios from the S2+ peaks. By analyzing against a standard material, acquired under similar conditions, we have extracted δ34S values to within ±5‰ (1‰ = 1 part per thousand) of the published values of our standards.

Towards adaptive method for peak migration time correction: discretization period in electropherograms

Capillary electrophoresis often causes unrepeatable peak migration times in the electropherogram due to changes of electroosmosis, yet in some cases this separation technique does not have a replacement alternative. Some attempts to overcome this issue have been performed introducing internal standards into the sample and compensating peak shifting in time. However, existing vector calculation-based methods are computationally intensive for portable instrumentation and usually limited to post-processing applications with 1 or 2 markers. In this work, an original approach of compensating peak migration time shift via signal discretization period correction is proposed. Using the proposed method, the number of reference points or markers that are used for compensation is extended. This method is effective in compensating migration time of peaks in real samples, where high sample injection volumes are used. Using 4 reference peaks in compensation, the method was capable of reducing the relative standard deviation of migration time of the peaks in the electropherograms more than 15 times. Corrected signal discretization periods indicated very high correlations with recorded separation currents, what can be perspective developing an adaptive peak migration time compensation method in capillary electrophoresis.

Mutation bias interacts with composition bias to influence adaptive evolution

ABSTRACTMutation is a biased stochastic process, with some types of mutations occurring more frequently than others. Previous work has used synthetic genotype-phenotype landscapes to study how such mutation bias affects adaptive evolution. Here, we consider 746 empirical genotype-phenotype landscapes, each of which describes the binding affinity of target DNA sequences to a transcription factor, to study the influence of mutation bias on adaptive evolution of increased binding affinity. By using empirical genotype-phenotype landscapes, we need to make only few assumptions about landscape topography and about the DNA sequences that each landscape contains. The latter is particularly important because the set of sequences that a landscape contains determines the types of mutations that can occur along a mutational path to an adaptive peak. That is, landscapes can exhibit a composition bias — a statistical enrichment of a particular type of mutation relative to a null expectation, throughout an entire landscape or along particular mutational paths — that is independent of any bias in the mutation process. Our results reveal the way in which composition bias interacts with biases in the mutation process under different population genetic conditions, and how such interaction impacts fundamental properties of adaptive evolution, such as its predictability, as well as the evolution of genetic diversity and mutational robustness.AUTHOR SUMMARYMutation is often depicted as a random process due its unpredictable nature. However, such randomness does not imply uniformly distributed outcomes, because some DNA sequence changes happen more frequently than others. Such mutation bias can be an orienting factor in adaptive evolution, influencing the mutational trajectories populations follow toward higher-fitness genotypes. Because these trajectories are typically just a small subset of all possible mutational trajectories, they can exhibit composition bias – an enrichment of a particular kind of DNA sequence change, such as transition or transversion mutations. Here, we use empirical data from eukaryotic transcriptional regulation to study how mutation bias and composition bias interact to influence adaptive evolution.