Search results for: Numerical probability modeling

Authors: Joanna Sooknanan, Donna Comissiong

Abstract:

We propose a phenomenological model for the process of polymer desorption. In so doing, we omit the usual theoretical approach of incorporating a fictitious viscoelastic stress term into the flux equation. As a result, we obtain a model that captures the essence of the phenomenon of trapping skinning, while preserving the integrity of the experimentally verified Fickian law for diffusion. An appropriate asymptotic analysis is carried out, and a parameter is introduced to represent the speed of the desorption front. Numerical simulations are performed to illustrate the desorption dynamics of the model. Recommendations are made for future modifications of the model, and provisions are made for the inclusion of experimentally determined frontal speeds.

Keywords: Phenomenological Model, Polymer, Desorption, Trapping Skinning

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1293

Authors: Jan Barák, Karel Fraňa, Jörg Stiller

Abstract:

This work presents results of moist air condensation in heat exchanger. It describes theoretical knowledge and definition of moist air. Model with geometry of square canal was created for better understanding and postprocessing of condensation phenomena. Different approaches were examined on this model to find suitable software and model. Obtained knowledge was applied to geometry of real heat exchanger and results from experiment were compared with numerical results. One of the goals is to solve this issue without creating any user defined function in the applied code. It also contains summary of knowledge and outlook for future work.

Keywords: Condensation, exchanger, experiment, validation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 5589

Authors: C. Ardil

Abstract:

A mathematical model and an effective numerical method for calculating the temperature field of the profile part of convection cooled blades have been developed. The theoretical substantiation of the method is proved by corresponding theorems. To this end, convergent quadrature processes were developed and error estimates were obtained in terms of the Zygmund continuity moduli.The boundary conditions for heat exchange are determined from the solution of the corresponding integral equations and empirical relations.The reliability of the developed methods is confirmed by the calculation-experimental studies of the thermohydraulic characteristics of the nozzle apparatus of the first stage of a gas turbine.

Keywords: Integral methods, determination of temperature fields, cooled blades, gas turbines.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 757

Authors: C. Clements, M. Hasenohr, A. Anvar

Abstract:

A glider is in essence an unpowered vehicle and in this project we designed and built an oceanic glider, designed to operate underwater. This Glider was designed to collect ocean data such as temperature, pressure and (in future measures physical dimensions of the operating environment) and output this data to an external source. Development of the Oceanic Glider required research into various actuation systems that control buoyancy, pitch and yaw and the dynamics of these systems. It also involved the design and manufacture of the Glider and the design and implementation of a controller that enabled the Glider to navigate and move in an appropriate manner.

Keywords: Ocean Glider, Robot, Automation, Command, Control, Navigation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1763

Authors: Myung-Gon Yoon, Jung-Ho Moon, Tae Kwon Ha

Abstract:

This paper deals with a stabilization problem for multi-agent systems, when all agents in a multi-agent system receive the same broadcasting control signal and the controller can measure not each agent output but the sum of all agent outputs. It is analytically shown that when the sum of all agent outputs is bounded with a certain broadcasting controller for a given reference, each agent output is separately bounded: stabilization of the sum of agent outputs always results in the stability of every agent output. A numerical example is presented to illustrate our theoretic findings in this paper.

Keywords: Broadcasting Control, Multi-agent System, Transfer Function

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1840

Authors: Ju H. Park, S.M. Lee

Abstract:

In this paper, we propose a robust controller design method for discrete-time systems with sector-bounded nonlinearities and time-varying delay. Based on the Lyapunov theory, delaydependent stabilization criteria are obtained in terms of linear matrix inequalities (LMIs) by constructing the new Lyapunov-Krasovskii functional and using some inequalities. A robust state feedback controller is designed by LMI framework and a reciprocally convex combination technique. The effectiveness of the proposed method is verified throughout a numerical example.

Keywords: Lur'e systems, Time-delay, Stabilization, LMIs.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1687

Authors: Yong-jun Zhao, Heng-li Zhang, Zong-yun Hu

Abstract:

Based on the sources- smoothed rank profile (SRP) and modified minimum description length (MMDL) principle, a method for estimation of the source coherency structure (SCS) and the number of wideband sources is proposed in this paper. Instead of focusing, we first use a spatial smoothing technique to pre-process the array covariance matrix of each frequency for de-correlating the sources and then use smoothed rank profile to determine the SCS and the number of wideband sources. We demonstrate the availability of the method by numerical simulations.

Keywords: Wideband sources, source coherency structure (SCS), smoothed rank profile (SRP).

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1317

Authors: Ali Vaseghi Ardekani, Hamid Reza Forutan, Amir Habibi, Ali Reza Rajabi, Hasan Adloo

Abstract:

Distributed generation units (DGs) are grid-connected or stand-alone electric generation units located within the electric distribution system at or near the end user. It is generally accepted that centralized electric power plants will remain the major source of the electric power supply for the near future. DGs, however, can complement central power by providing incremental capacity to the utility grid or to an end user. This paper presents an efficient power dispatching model for hybrid wind-Solar power generation system, to satisfy some essential requirements, such as dispersed electric power demand, electric power quality and reducing generation cost and so on. In this paper, presented some elements of the main parts in the hybrid system; and then made fundamental dispatching strategies according to different situations; then pointed out four improving measures to improve genetic algorithm, such as: modify the producing way of selection probability, improve the way of crossover, protect excellent chromosomes, and change mutation range and so on. Finally, propose a technique for solving the unit's commitment for dispatching problem based on an improved genetic algorithm.

Keywords: Power Quality, Wind-Solar System, Genetic Algorithm, Hybrid System.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1646

Authors: Amer S. Al-Yahmadi, T.C. Hsia

Abstract:

This paper seeks to develop simple yet practical and efficient control scheme that enables cooperating arms to handle a flexible beam. Specifically the problem studied herein is that of two arms rigidly grasping a flexible beam and such capable of generating forces/moments in such away as to move a flexible beam along a predefined trajectory. The paper develops a sliding mode control law that provides robustness against model imperfection and uncertainty. It also provides an implicit stability proof. Simulation results for two three joint arms moving a flexible beam, are presented to validate the theoretical results.

Keywords: Sliding mode control, cooperative manipulators.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1635

Authors: I. López , J. Garay, R. Carreño, Z. Varga

Abstract:

Monitoring of ecological systems is one of the major issues in ecosystem research. The concepts and methodology of mathematical systems theory provide useful tools to face this problem. In many cases, state monitoring of a complex ecological system consists in observation (measurement) of certain state variables, and the whole state process has to be determined from the observed data. The solution proposed in the paper is the design of an observer system, which makes it possible to approximately recover the state process from its partial observation. The method is illustrated with a trophic chain of resource – producer – primary consumer type and a numerical example is also presented.

Keywords: Monitoring, observer system, trophic chain

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1433

Authors: M. A. Serag-Eldin

Abstract:

The paper presents the results of simple measurements conducted on a model of a wind-driven venturi-type room ventilator. The ventilator design is new and was developed employing mathematical modeling. However, the computational model was not validated experimentally for the particular application considered. The paper presents the performance of the ventilator model under laboratory conditions, for five different wind tunnel speeds. The results are used to both demonstrate the effectiveness of the new design and to validate the computational model employed to develop it.

Keywords: Venturi-flow, ventilation, Wind-energy, Wind flow.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1357

Authors: Jatinderkumar R. Saini, Apurva A. Desai

Abstract:

e-mail has become an important means of electronic communication but the viability of its usage is marred by Unsolicited Bulk e-mail (UBE) messages. UBE consists of many types like pornographic, virus infected and 'cry-for-help' messages as well as fake and fraudulent offers for jobs, winnings and medicines. UBE poses technical and socio-economic challenges to usage of e-mails. To meet this challenge and combat this menace, we need to understand UBE. Towards this end, the current paper presents a content-based textual analysis of nearly 3000 winnings-announcing UBE. Technically, this is an application of Text Parsing and Tokenization for an un-structured textual document and we approach it using Bag Of Words (BOW) and Vector Space Document Model techniques. We have attempted to identify the most frequently occurring lexis in the winnings-announcing UBE documents. The analysis of such top 100 lexis is also presented. We exhibit the relationship between occurrence of a word from the identified lexisset in the given UBE and the probability that the given UBE will be the one announcing fake winnings. To the best of our knowledge and survey of related literature, this is the first formal attempt for identification of most frequently occurring lexis in winningsannouncing UBE by its textual analysis. Finally, this is a sincere attempt to bring about alertness against and mitigate the threat of such luring but fake UBE.

Keywords: Lexis, Unsolicited Bulk e-mail (UBE), Vector SpaceDocument Model, Winnings, Lottery

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1540

Authors: Chung-Hung Tsai

Abstract:

Global environmental changes lead to increased frequency and scale of natural disaster, Taiwan is under the influence of global warming and extreme weather. Therefore, the vulnerability was increased and variability and complexity of disasters is relatively enhanced. The purpose of this study is to consider the source and magnitude of hazard characteristics on the tourism industry. Using modern risk management concepts, integration of related domestic and international basic research, this goes beyond the Taiwan typhoon disaster risk assessment model and evaluation of loss. This loss evaluation index system considers the impact of extreme weather, in particular heavy rain on the tourism industry in Taiwan. Consider the extreme climate of the compound impact of disaster for the tourism industry; we try to make multi-hazard risk assessment model, strategies and suggestions. Related risk analysis results are expected to provide government department, the tourism industry asset owners, insurance companies and banking include tourist disaster risk necessary information to help its tourism industry for effective natural disaster risk management.

Keywords: Tourism industry, extreme weather, multi-hazard, vulnerability analysis, loss exceeding probability, risk management.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3830

Authors: Miaomiao Yang, Shouming Zhong

Abstract:

This paper studies the problem of exponential stability analysis for recurrent neural networks with time-varying delay.By establishing a suitable augmented LyapunovCKrasovskii function and a novel sufficient condition is obtained to guarantee the exponential stability of the considered system.In order to get a less conservative results of the condition,zero equalities and reciprocally convex approach are employed. The several exponential stability criterion proposed in this paper is simpler and effective. A numerical example is provided to demonstrate the feasibility and effectiveness of our results.

Keywords: Exponential stability , Neural networks, Linear matrix inequality, Lyapunov-Krasovskii, Time-varying.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1767

Authors: M. M. Kassem, A. S. Rashed

Abstract:

The time dependent progress of a chemical reaction over a flat horizontal plate is here considered. The problem is solved through the group similarity transformation method which reduces the number of independent by one and leads to a set of nonlinear ordinary differential equation. The problem shows a singularity at the chemical reaction order n=1 and is analytically solved through the perturbation method. The behavior of the process is then numerically investigated for n≠1 and different Schmidt numbers. Graphical results for the velocity and concentration of chemicals based on the analytical and numerical solutions are presented and discussed.

Keywords: Time dependent, chemical convection, grouptransformation method, perturbation method.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1628

Authors: Panagiotis H. Tsarouhas

Abstract:

the reliability analysis of the medical equipments can help to increase the availability and the efficiency of the systems. In this manuscript we present a simple method of decomposition that could be easily applied on the complex medical systems. Using this method we can easily calculate the effect of the subsystems or components on the reliability of the overall system. Furthermore, to investigate the effect of subsystems or components on system performance, we perform a numerical study varying every time the worst reliability of subsystem or component with another which has higher reliability. It can also be useful to engineers and designers of medical equipment, who wishes to optimize the complex systems.

Keywords: Reliability, Availability, Series-parallel System, medical equipment.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2397

Authors: Meysam Lak, Mohammad Fatehi Marji, Alireza Yarahamdi Bafghi, Abolfazl Abdollahipour

Abstract:

A wellbore is a hole that is drilled to aid in the exploration and recovery of natural resources including oil and gas. Occasionally, in order to increase productivity index and porosity of the wellbore and reservoir, the well stimulation methods have been used. Hydraulic fracturing is one of these methods. Moreover, several explosions at the end of the well can stimulate the reservoir and create fractures around it. In this study, crack initiation in rock around the wellbore has been numerically modeled due to explosion. One, two, three, and four pairs of explosion have been set at the end of the wellbore on its wall. After each stage of the explosion, results have been presented and discussed. Results show that this method can initiate and probably propagate several fractures around the wellbore.

Keywords: Crack initiation, explosion, finite difference modelling, well productivity.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 810

Authors: Ehan Sabah Shukri, Wirachman Wisnoe

Abstract:

Numerical study is performed to investigate the temperature distribution in an annular diffuser fitted with helical tape hub. Different pitches (Y = 20 mm, and Y = 30 mm) for the helical tape are studied with different heights (H = 20 mm, 22 mm, and 24 mm) to be compared. The geometry of the annular diffuser and the inlet condition for both hub arrangements are kept constant. The result obtains that using helical tape insert with different pitches and different heights will force the temperature to distribute in a helical direction; however the use of helical tape hub with height (H = 22 mm) for both pitches enhance the temperature distribution in a good manner.

Keywords: Helical tape, divergent fluid flow, temperature distribution, swirl flow, CFD.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1785

Authors: Arash Mir Abdolah Lavasani, Hamidreza Bayat

Abstract:

Heat transfer from two cam shape cylinder in tandem arrangement had been studied numerically. The distance between the centers of cylinders (L) is allowed to vary to change the longitudinal pitch ratio (L/Deq). The equivalent diameter of the cylinder (Deq) is 27.6 mm and longitudinal pitch ratio varies in range 2<L/Deq<6. The Reynolds number based on equivalent circular cylinder are within 50< Reeq <300. Results show that Nusselt number of second cylinder increases about 5 to 33 times when longitudinal pitch ratio increases from 2 to 6.

Keywords: Cam Shaped, tandem Cylinders, Numerical, Heat Transfer.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1556

Authors: Deyu Sun, Guangbin Wang

Abstract:

In this paper, we present preconditioned generalized accelerated overrelaxation (GAOR) methods for solving certain nonsingular linear system. We compare the spectral radii of the iteration matrices of the preconditioned and the original methods. The comparison results show that the preconditioned GAOR methods converge faster than the GAOR method whenever the GAOR method is convergent. Finally, we give two numerical examples to confirm our theoretical results.

Keywords: Preconditioned, GAOR method, linear system, convergence, comparison.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1567

Authors: Dmitry D. Zakharov

Abstract:

The theoretical prediction of the acoustical polarization effects in the heterogeneous composites, made of thick elastic solids with thin nematic films, is presented. The numericalanalytical solution to the problem of the different wave propagation exhibits some new physical effects in the low frequency domain: the appearance of the critical frequency and the existence of the narrow transition zone where the wave rapidly changes its speed. The associated wave attenuation is highly perturbed in this zone. We also show the possible appearance of the critical frequencies where the attenuation changes the sign. The numerical results of parametrical analysis are presented and discussed.

Keywords: Surface wave, guided wave, heterogeneous composite, nematic coating.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1362

Authors: Artem Nuriev, Olga Zaitseva

Abstract:

This paper is devoted to the study of a viscous incompressible flow around a circular cylinder performing harmonic oscillations, especially the steady streaming phenomenon. The research methodology is based on the asymptotic explanation method combined with the computational bifurcation analysis. The research approach develops Schlichting and Wang decomposition method. Present studies allow to identify several regimes of the secondary streaming with different flow structures. The results of the research are in good agreement with experimental and numerical simulation data.

Keywords: Oscillating cylinder, Secondary Streaming, Flow Regimes, Asymptotic and Bifurcation Analysis.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2173

Authors: Ezzatallah Baloui Jamkhaneh, Bahram Sadeghpour-Gildeh, Gholamhossein Yari

Abstract:

This purpose of this paper is to present the acceptance single sampling plan when the fraction of nonconforming items is a fuzzy number and being modeled based on the fuzzy Poisson distribution. We have shown that the operating characteristic (oc) curves of the plan is like a band having a high and low bounds whose width depends on the ambiguity proportion parameter in the lot when that sample size and acceptance numbers is fixed. Finally we completed discuss opinion by a numerical example. And then we compared the oc bands of using of binomial with the oc bands of using of Poisson distribution.

Keywords: Statistical quality control, acceptance single sampling, fuzzy number.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1991

Authors: Mahboobeh Abdoos, Karim Faez, Masoud Sabaei

Abstract:

Position based routing protocols are the kinds of routing protocols, which they use of nodes location information, instead of links information to routing. In position based routing protocols, it supposed that the packet source node has position information of itself and it's neighbors and packet destination node. Greedy is a very important position based routing protocol. In one of it's kinds, named MFR (Most Forward Within Radius), source node or packet forwarder node, sends packet to one of it's neighbors with most forward progress towards destination node (closest neighbor to destination). Using distance deciding metric in Greedy to forward packet to a neighbor node, is not suitable for all conditions. If closest neighbor to destination node, has high speed, in comparison with source node or intermediate packet forwarder node speed or has very low remained battery power, then packet loss probability is increased. Proposed strategy uses combination of metrics distancevelocity similarity-power, to deciding about giving the packet to which neighbor. Simulation results show that the proposed strategy has lower lost packets average than Greedy, so it has more reliability.

Keywords: Mobile Ad Hoc Network, Position Based, Reliability, Routing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1763

Authors: R. Gunabalan, S. Immanuel Prabakaran, J. Reegan, S. Ganesh

Abstract:

This paper describes a software approach for modeling inverter fed induction motor drive using Laboratory Virtual Instrument Engineering Workbench (LabVIEW). The reason behind the selection of LabVIEW software is because of its strong graphical interface, flexibility of its programming language combined with built-in tools designed specifically for test, measurement and control. LabVIEW is generally used in most of the applications for data acquisition, test and control. In this paper, inverter and induction motor are modeled using LabVIEW toolkits. Simulation results are presented and are validated.

Keywords: Induction motor, LabVIEW, State model.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 8576

Authors: R. M. Farouk

Abstract:

In this paper, we present an analytical analysis of the representation of images as the magnitudes of their transform with the discrete wavelets. Such a representation plays as a model for complex cells in the early stage of visual processing and of high technical usefulness for image understanding, because it makes the representation insensitive to small local shifts. We found that if the signals are band limited and of zero mean, then reconstruction from the magnitudes is unique up to the sign for almost all signals. We also present an iterative reconstruction algorithm which yields very good reconstruction up to the sign minor numerical errors in the very low frequencies.

Keywords: Wavelets, Image processing signal processing, Image reconstruction

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1388

Authors: A. P. Singh

Abstract:

The results reported in this paper are the part of an extensive laboratory investigation undertaken to study the effects of fibre parameters on the permeability and strength characteristics of steel fibre reinforced concrete (SFRC). The effect of varying fibre content and curing age on the water permeability, compressive and split tensile strengths of SFRC was investigated using straight steel fibres having an aspect ratio of 65. Samples containing three different weight fractions of 1.0%, 2.0% and 4.0% were cast and tested for permeability and strength after 7, 14, 28 and 60 days of curing. Plain concrete samples were also cast and tested for reference purposes.

Permeability was observed to decrease significantly with the addition of steel fibres and continued to decrease with increasing fibre content and increasing curing age. An exponential relationship was observed between permeability and compressive and split tensile strengths for SFRC as well as PCC. To evaluate the effect of fibre content on the permeability and strength characteristics, the Analysis of Variance (ANOVA) statistical method was used. An a level (probability of error) of 0.05 was used for ANOVA test. Regression analysis was carried out to develop relationship between permeability, compressive strength and curing age.

Keywords: Permeability, grade of concrete, fibre shape, fibre content, curing age, steady state, Darcy’s law, method of penetration.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3083

Authors: M. Shakaib, M. Ehtesham-ul Haq, I. Ahmed, R.M. Yunus

Abstract:

Computational fluid dynamics (CFD) simulations carried out in this paper show that spacer orientation has a major influence on temperature patterns and on the heat transfer rates. The local heat flux values significantly vary from high to very low values at each filament when spacer touches the membrane surface. The heat flux profile is more uniform when spacer filaments are not in contact with the membrane thus making this arrangement more beneficial. The temperature polarization is also found to be less in this case when compared to the empty channel.

Keywords: heat transfer, membrane distillation, spacer, temperature polarization.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1724

Authors: Panupong Chanplin, Kathaleeya Chanda., Wilailuk Mepracha

Abstract:

The aims of this research were twofold: 1) to brand identity design of Creative Agriculture Community Enterprise, Bang Khonthi District, Samut Songkhram Province and 2) to study the level of tourist satisfaction towards brand identity design of Creative Agriculture Community Enterprise, Bang Khonthi District, Samut Songkhram Province. tourist satisfaction was measured using six criteria: clear brand positioning, likeable brand personality, memorable logo, attractive color palette, professional typography and on-brand supporting graphics. The researcher utilized a probability sampling method via simple random sampling. The sample consisted of 30 tourists in the Creative Agriculture Community Enterprise. Statistics utilized for data analysis were percentage, mean, and standard deviation. The results suggest that tourist had high levels of satisfaction towards all six criteria of the brand identity design that was designed to target them. This study proposes that specifically brand identity designed of Creative Agriculture Community Enterprise could also be implemented with other real media already available on the market.

Keywords: Satisfaction, brand identity, logo, creative agriculture community enterprise.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 815

Authors: Maxence Queyrel, Edi Prifti, Alexandre Templier, Jean-Daniel Zucker

Abstract:

Analysis of the human microbiome using metagenomic sequencing data has demonstrated high ability in discriminating various human diseases. Raw metagenomic sequencing data require multiple complex and computationally heavy bioinformatics steps prior to data analysis. Such data contain millions of short sequences read from the fragmented DNA sequences and stored as fastq files. Conventional processing pipelines consist in multiple steps including quality control, filtering, alignment of sequences against genomic catalogs (genes, species, taxonomic levels, functional pathways, etc.). These pipelines are complex to use, time consuming and rely on a large number of parameters that often provide variability and impact the estimation of the microbiome elements. Training Deep Neural Networks directly from raw sequencing data is a promising approach to bypass some of the challenges associated with mainstream bioinformatics pipelines. Most of these methods use the concept of word and sentence embeddings that create a meaningful and numerical representation of DNA sequences, while extracting features and reducing the dimensionality of the data. In this paper we present an end-to-end approach that classifies patients into disease groups directly from raw metagenomic reads: metagenome2vec. This approach is composed of four steps (i) generating a vocabulary of k-mers and learning their numerical embeddings; (ii) learning DNA sequence (read) embeddings; (iii) identifying the genome from which the sequence is most likely to come and (iv) training a multiple instance learning classifier which predicts the phenotype based on the vector representation of the raw data. An attention mechanism is applied in the network so that the model can be interpreted, assigning a weight to the influence of the prediction for each genome. Using two public real-life data-sets as well a simulated one, we demonstrated that this original approach reaches high performance, comparable with the state-of-the-art methods applied directly on processed data though mainstream bioinformatics workflows. These results are encouraging for this proof of concept work. We believe that with further dedication, the DNN models have the potential to surpass mainstream bioinformatics workflows in disease classification tasks.

Keywords: Metagenomics, phenotype prediction, deep learning, embeddings, multiple instance learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 910