Freezing of gait in Parkinson's disease: Classification using computational intelligence.

Parkinson's disease (PD) is a neurodegenerative disease represented by the progressive loss of dopamine producing neurons, with motor and non-motor symptoms that may be hard to distinguish from other disorders. Affecting millions of people across the world, its symptoms include bradykinesia, tremors, depression, rigidity, postural instability, cognitive decline, and falls. Furthermore, changes in gait can be used as a primary diagnosis factor. A dataset is described that records data on healthy individuals and on PD patients, including those who experience freezing of gait, in both the ON and OFF-medication states. The dataset is comprised of data for four separate tasks: voluntary stop, timed up and go, simple motor task, and dual motor and cognitive task. Seven different classifiers are applied to two problems relating to this data. The first problem is to distinguish PD patients from healthy individuals, both overall and per task. The second problem is to determine the effectiveness of medication. A thorough analysis on the classifiers and their results is performed. Overall, multilayer perceptron and decision tree provide the most consistent results.

All lockdowns are not equal: Reducing epidemic impact through evolutionary computation.

The impact of different lockdown strategies upon the total number of infections in an epidemic are evaluated for two models of infection: one in which the disease confers permanent immunity, and one in which it does not. The strategies are based upon the proportion of the population infected at a time in order to trigger lockdown, combined with the proportion of interactions removed during lockdown. The population, its interactions, and the relative strengths of those interactions are stored in a weighted contact network, from which edges are removed during lockdown. These edges are selected using an evolutionary algorithm (EA) designed to minimize total infections. Using the EA to select edges significantly reduces total infections in comparison to random selection. In fact, the EA results for the least strict conditions were similar or better to the random results for the most strict conditions, showing that a judicious choice of restrictions during lockdown has the greatest effect on reducing infections. Further, when using the most strict rules a smaller proportion of interactions can be removed to obtain similar or better results in comparison to removing a higher proportion of interactions for less strict rules.

Effective decoders for DNA codes.

A number of applications use DNA as a storage mechanism. Because processes in these applications may cause errors in the data, the information must be encoded as one of a chosen set of words that are well separated from one another - a DNA error-correcting code. Typically, the types of errors that may occur include insertions, deletions and substitutions of symbols, making the edit metric the most suitable choice to measure the distance between strings. Decoding, the process of recovering the original word when errors occur, is complicated by biological restrictions combined with a high cost to calculate edit distance. Side effect machines (SEMs), an extension of finite state machines, can provide efficient decoding algorithms for such codes. Several codes of varying lengths are used to study the effectiveness of evolutionary programming (EP) as a general approach for finding SEMs for edit metric decoding. Two classification methods (direct and fuzzy classification) are compared, and different EP settings are examined to observe how decoding accuracy is affected. Regardless of code length, the best results are found using fuzzy classification. The best accuracy is seen for codes of length 10, for which a maximum accuracy of up to 99.4% is achieved for distance 1 and distance 2 and 3 achieve up to 97.1% and 85.9%, respectively. Additionally, the SEMs are examined for potential bloat by comparing the number of reachable states against the total number of states. Bloat is seen more in larger machines than in smaller machines. Furthermore, the results are analysed to find potential trends and relationships among the parameters, with the most consistent trend being that, when allowed, the longer codes generally show a propensity for larger machines.

Lossy Compression of Quality Values in Sequencing Data.

The dropping cost of sequencing human DNA has allowed for fast development of several projects around the world generating huge amounts of DNA sequencing data. This deluge of data has run up against limited storage space, a problem that researchers are trying to solve through compression techniques. In this study we address the compression of SAM files, the standard output files for DNA alignment. We specifically study lossy compression techniques used for quality values reported in the SAM file and analyze the impact of such lossy techniques on the CRAM format. We present a series of experiments using a data set corresponding to individual NA12878 with three different fold coverages. We introduce a new lossy model, dynamic binning, and compare its performance to other lossy techniques, namely Illumina binning, LEON and QVZ. We analyze the compression ratio when using CRAM and also study the impact of the lossy techniques on SNP calling. Our results show that lossy techniques allow a better CRAM compression ratio. Furthermore, we show that SNP calling performance is not negatively affected and may even be boosted.

A centrality based multi-objective approach to disease gene association.

Disease Gene Association finds genes that are involved in the presentation of a given genetic disease. We present a hybrid approach which implements a multi-objective genetic algorithm, where input consists of centrality measures based on various relational biological evidence types merged into a complex network. Multiple objective settings and parameters are studied including the development of a new exchange methodology, safe dealer-based crossover. Successful results with respect to breast cancer and Parkinson's disease compared to previous techniques and popular known databases are shown. In addition, the newly developed methodology is also successfully applied to Alzheimer's disease, further demonstrating its flexibility. Across all three case studies the strongest results were produced by the shortest path-based measures stress and betweenness, either in a single objective parameter setting or when used in conjunction in a multi-objective environment. The new crossover technique achieved the best results when applied to Alzheimer's disease.

Models of Parkinson's Disease Patient Gait.

Parkinson's Disease is a disorder with diagnostic symptoms that include a change to a walking gait. The disease is problematic to diagnose. An objective method of monitoring the gait of a patient is required to ensure the effectiveness of diagnosis and treatments. We examine the suitability of Extreme Gradient Boosting (XGBoost) and Artificial Neural Network (ANN) Models compared to Symbolic Regression (SR) using genetic programming that was demonstrated to be successful in previous works on gait. The XGBoost and ANN models are found to out-perform SR, but the SR model is more human explainable.

Restarting and recentering genetic algorithm variations for DNA fragment assembly: The necessity of a multi-strategy approach.

DNA Fragment assembly - an NP-Hard problem - is one of the major steps in of DNA sequencing. Multiple strategies have been used for this problem, including greedy graph-based algorithms, deBruijn graphs, and the overlap-layout-consensus approach. This study focuses on the overlap-layout-consensus approach. Heuristics and computational intelligence methods are combined to exploit their respective benefits. These algorithm combinations were able to produce high quality results surpassing the best results obtained by a number of competitive algorithms specially designed and tuned for this problem on thirteen of sixteen popular benchmarks. This work also reinforces the necessity of using multiple search strategies as it is clearly observed that algorithm performance is dependent on problem instance; without a deeper look into many searches, top solutions could be missed entirely.

On the synthesis of DNA error correcting codes.

DNA error correcting codes over the edit metric consist of embeddable markers for sequencing projects that are tolerant of sequencing errors. When a genetic library has multiple sources for its sequences, use of embedded markers permit tracking of sequence origin. This study compares different methods for synthesizing DNA error correcting codes. A new code-finding technique called the salmon algorithm is introduced and used to improve the size of best known codes in five difficult cases of the problem, including the most studied case: length six, distance three codes. An updated table of the best known code sizes with 36 improved values, resulting from three different algorithms, is presented. Mathematical background results for the problem from multiple sources are summarized. A discussion of practical details that arise in application, including biological design and decoding, is also given in this study.

Genetic algorithm with alternating selection pressure for protein side-chain packing and pK(a) prediction.

The prediction of protein side-chain conformation is central for understanding protein functions. Side-chain packing is a sub-problem of protein folding and its computational complexity has been shown to be NP-hard. We investigated the capabilities of a hybrid (genetic algorithm/simulated annealing) technique for side-chain packing and for the generation of an ensemble of low energy side-chain conformations. Our method first relies on obtaining a near-optimal low energy protein conformation by optimizing its amino-acid side-chains. Upon convergence, the genetic algorithm is allowed to undergo forward and "backward" evolution by alternating selection pressures between minimal and higher energy setpoints. We show that this technique is very efficient for obtaining distributions of solutions centered at any desired energy from the minimum. We outline the general concepts of our evolutionary sampling methodology using three different alternating selective pressure schemes. Quality of the method was assessed by using it for protein pK(a) prediction.