检索结果-内蒙古大学图书馆

Causal Discovery Toolbox: Uncovering causal relationships in Python

JOURNAL OF MACHINE LEARNING RESEARCH 2020年第1期21卷 1-5页

作者： Kalainathan, Diviyan Goudet, Olivier Dutta, Ritik Univ Paris Sud Fen Tech TAU LRIINRIA F-75013 Paris France Univ Angers LERIA 2 Blvd Lavoisier F-49045 Angers France IIT Gandhinagar Gandhinagar 382355 Gujarat India

This paper presents a new open source Python framework for causal discovery from observational data and domain background knowledge, aimed at causal graph and causal mechanism modeling. The CDT package implements an end-to-end approach, recovering the direct dependencies (the skeleton of the causal graph) and the causal relationships between variables. It includes algorithms from the 'BNLEARN' (Scutari, 2018) and 'PCALG' (Kalisch et al., 2018) packages, together with algorithms for pairwise causal discovery such as ANM (Hover et al., 2009). CDT is available under the MIT License at https://***/FenTechSolutions/CausalDiscoveryToolbox.

关键词： Causal Discovery Graph recovery open source constraint-based methods score-based methods pairwise causality Markov blanket

来源：评论

学校读者我要写书评

暂无评论

Optimization of bioprocess productivity based on metabolic-genetic network models with bilevel dynamic programming

引用

BIOTECHNOLOGY AND BIOENGINEERING 2018年第7期115卷 1829-1841页

作者： Jabarivelisdeh, Banafsheh Waldherr, Steffen Int Max Planck Res Sch IMPRS Adv Methods Proc & S Magdeburg Germany Ctr Dynam Syst Inst Automat Engn Magdeburg Saxony Anhalt Germany Katholieke Univ Leuven Dept Chem Engn Celestijnenlaan 200F Bus 2424 B-3001 Leuven Belgium

One of the main goals of metabolic engineering is to obtain high levels of a microbial product through genetic modifications. To improve the productivity of such a process, the dynamic implementation of metabolic engineering strategies has been proven to be more beneficial compared to static genetic manipulations in which the gene expression is not controlled over time, by resolving the trade-off between growth and production. In this work, a bilevel optimization framework based on constraint-based models is applied to identify an optimal strategy for dynamic genetic and process level manipulations to increase productivity. The dynamic enzyme-cost flux balance analysis (deFBA) as underlying metabolic network model captures the network dynamics and enables the analysis of temporal regulation in the metabolic-genetic network. We apply our computational framework to maximize ethanol productivity in a batch process with Escherichia coli. The results highlight the importance of integrating the genetic level and enzyme production and degradation processes for obtaining optimal dynamic gene and process manipulations.

关键词： bilevel optimization constraint-based methods dynamic enzyme-cost flux balance analysis metabolic engineering resource allocation models

来源：评论

学校读者我要写书评

暂无评论

Dynamic optimization of metabolic networks coupled with gene expression

引用

JOURNAL OF THEORETICAL BIOLOGY 2015年 365卷 469-485页

作者： Waldherr, Steffen Oyarzun, Diego A. Bockmayr, Alexander Univ Magdeburg Inst Automat Engn D-39106 Magdeburg Germany Univ London Imperial Coll Sci Technol & Med Dept Math London SW7 2AZ England Free Univ Berlin DFG Res Ctr Matheon D-14195 Berlin Germany

The regulation of metabolic activity by tuning enzyme expression levels is crucial to sustain cellular growth in changing environments. Metabolic networks are often studied at steady state using constraint-based models and optimization techniques. However, metabolic adaptations driven by changes in gene expression cannot be analyzed by steady state models, as these do not account for temporal changes in biomass composition. Here we present a dynamic optimization framework that integrates the metabolic network with the dynamics of biomass production and composition. An approximation by a timescale separation leads to a coupled model of quasi-steady state constraints on the metabolic reactions, and differential equations for the substrate concentrations and biomass composition. We propose a dynamic optimization approach to determine reaction fluxes for this model, explicitly taking into account enzyme production costs and enzymatic capacity. In contrast to the established dynamic flux balance analysis, our approach allows predicting dynamic changes in both the metabolic fluxes and the biomass composition during metabolic adaptations. Discretization of the optimization problems leads to a linear program that can be efficiently solved. We applied our algorithm in two case studies: a minimal nutrient uptake network, and an abstraction of core metabolic processes in bacteria. In the minimal model, we show that the optimized uptake rates reproduce the empirical Monod growth for bacterial cultures. For the network of core metabolic processes, the dynamic optimization algorithm predicted commonly observed metabolic adaptations, such as a diauxic switch with a preference ranking for different nutrients, re-utilization of waste products after depletion of the original substrate, and metabolic adaptation to an impending nutrient depletion. These examples illustrate how dynamic adaptations of enzyme expression can be predicted solely from an optimization principle. (C) 2014 Else

关键词： Flux optimization constraint-based methods Metabolic genetic networks Bacterial growth

来源：评论

学校读者我要写书评

暂无评论

Improving Bioprocess Productivity Using constraint-based Models in a Dynamic Optimization Scheme

Improving Bioprocess Productivity Using Constraint-Based Mod...

引用

6th IFAC Conference on Foundations of Systems Biology in Engineering (FOSBE)

作者： Jabarivelisdeh, Banafsheh Waldherr, Steffen Int Max Planck Res Sch IMPRS Adv Methods Proc & S Magdeburg Germany Otto Von Guericke Univ Ctr Dynam Syst Inst Automat Engn Magdeburg Germany Katholieke Univ Leuven Dept Chem Engn Leuven Belgium Harvard Univ Sch Engn & Appl Sci Cambridge MA 02138 USA

Modulating the expression of target genes is an effective metabolic engineering approach to increase bioprocess productivity. In this work, a bilevel optimization framework is applied for dynamic manipulation of gene expression based on constraint-based models. A dynamic model in the inner problem captures the network dynamics and enables temporal regulation of the metabolic network. Through linearization of nonlinear constraints, this framework is based on linear optimization which saves on computation time especially in a bilevel structure. A case study involving batch fermentation of Escherichia coli for ethanol production is considered to find an optimal manipulation strategy of metabolic networks for maximal productivity. (C) 2016, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

关键词： Productivity constraint-based methods Metabolic engineering Bilevel optimization Dynamic flux balance analysis

来源：评论

学校读者我要写书评

暂无评论

An Efficient and Scalable Algorithm for Local Bayesian Network Structure Discovery

An Efficient and Scalable Algorithm for Local Bayesian Netwo...

引用

European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD)

作者： de Morais, Sergio Rodrigues Aussem, Alex Univ Lyon F-69000 Lyon France Univ Lyon 1 LIESP Lab F-69622 Villeurbanne France

ISBN: (纸本)9783642159381

We present an efficient and scalable constraint-based algorithm, called Hybrid Parents and Children (HPC), to learn the parents and children of a target variable in a Bayesian network. Finding those variables is an important first step in many applications including Bayesian network structure learning, dimensionality reduction and feature selection. The algorithm combines ideas from incremental and divide-and-conquer methods in a principled and effective way, while still being sound in the sample limit. Extensive empirical experiments are provided on public synthetic and real-world data sets of various sample sizes. The most noteworthy feature of HPC is its ability to handle large neighborhoods contrary to current CB algorithm proposals. The number of calls to the statistical test, en hence the run-time, is empirically on the order O(n(1.09)), where n is the number of variables, on the five benchmarks that we considered, and O(n(1.21)) on a real drug design characterized by 138,351 features.

关键词： Bayesian network structure learning constraint-based methods feature selection

来源：评论

学校读者我要写书评

暂无评论

Process Independence Testing in Proximal Graphical Event Models 1

Process Independence Testing in Proximal Graphical Event Mod...

引用

Conference on Causal Learning and Reasoning

作者： Bhattacharjya, Debarun Shanmugam, Karthikeyan Gao, Tian Subramanian, Dharmashankar Thomas J Watson Res Ctr IBM Res Yorktown Hts NY 10598 USA

Datasets involving irregular occurrences of different types of events over the timeline are increasingly commonly available. Proximal graphical event models (PGEMs) are a recent graphical representation for modeling relationships between different event types in such datasets. Existing algorithms for learning PGEMs from event datasets perform poorly on the task of structure discovery, which is particularly important for causal inference since the underlying graph determines the effect of interventions. In this paper, we explore causal semantics in PGEMs and study process independencies implied by the graphical structure of the model. We introduce (conditional) process independence tests for causal PGEMs, deploying them using variations of constraint-based structure discovery algorithms for Bayesian networks. Through experiments with synthetic and real datasets, we show that the proposed approaches are better at balancing precision and recall, demonstrating improved F1 scores over state-of-the-art baselines.

关键词： Event datasets Structure discovery Event models Graphical event models Process independence Local independence constraint-based methods

来源：评论

学校读者我要写书评

暂无评论

Assigning resources to constrained activities

引用

3rd International Conference on Practice and Theory of Automated Timetabling

作者： Meisels, A Ovadia, E Ben Gurion Univ Negev Dept Comp Sci IL-84105 Beer Sheva Israel

ISBN: (纸本)3540424210

Resource allocation problems (RAPs)are naturally represented as constraint networks (CNs), with constraints of inequality among activities that compete for the same resources at the same time [5]. A large variety of timetabling problems can be formulated as CNs with inequality constraints (representing time conflicts among classes of the same teacher, for example). A new algorithm for solving networks of RAPs is described and its detailed behavior is presented on a small example. The proposed algorithm delays assignments of resources to selected activities and processes the network during the assignment procedure, to select delays and values. The proposed algorithm is an enhancement to standard intelligent backtracking algorithms [13], is complete and performs better than two former approaches to solving CNs of resource allocation [15], [3], [4]. Results of comparing the proposed resource assignment algorithm to other ordering heuristics, on randomly generated networks, are reported.

关键词： constraint-based methods timetabling networks

来源：评论

学校读者我要写书评

暂无评论

Learning Bayesian Networks with Cops and Robbers 10

Learning Bayesian Networks with Cops and Robbers

引用

10th International Conference on Probabilistic Graphical Models (PGM)

作者： Talvitie, Topi Parviainen, Pekka Univ Helsinki Dept Comp Sci Helsinki Finland Univ Bergen Dept Informat Bergen Norway

constraint-based methods for learning structures of Bayesian networks are based on testing conditional independencies between variables and constructing a structure that expresses the same conditional independencies as indicated by the tests. We present a constraint-based algorithm that learns the structure of a Bayesian network by simulating a cops-and-a-robber game. The algorithm is designed for learning structures of low treewidth distributions and in such case it conducts conditional independence tests only with small conditioning sets.

关键词： Bayesian networks Structure learning constraint-based methods Treewidth

来源：评论

学校读者我要写书评

暂无评论

Haptic interaction with virtual avatars

Haptic interaction with virtual avatars

引用

6th EuroHaptics International Conference

作者： Keith, Francois Evrard, Paul Chardonnet, Jean-Remy Miossec, Sylvain Kheddar, Abderrahmane AIST CNRS Joint Japanese French Robot Lab Tsukuba Ibaraki Japan

ISBN: (纸本)9783540690566

In this paper we present an interactive dynamic simulator for virtual avatars. It allows creation and manipulation of objects in a collaborative way by virtual avatars or between virtual avatars and users. The users interact with the simulation environment using a haptic probe which provides force feedback. This dynamic simulator uses fast dynamics computation and constraint-based methods with friction. It is part of a general framework that is being devised for studies of collaborative scenarios with haptic feedback.

关键词： dynamic simulator constraint-based methods user-object-avatar haptic interaction

来源：评论

学校读者我要写书评

暂无评论

ParallelPC: An R Package for Efficient Causal Exploration in Genomic Data 22nd

ParallelPC: An R Package for Efficient Causal Exploration in...

引用

22nd Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD)

作者： Thuc Duy Le Xu, Taosheng Liu, Lin Shu, Hu Hoang, Tao Li, Jiuyong Univ South Australia Mawson Lakes SA 5095 Australia Chinese Acad Sci Inst Intelligent Machines Hefei Inst Phys Sci Hefei 230031 Peoples R China

ISBN: (纸本)9783030045036;9783030045029

Discovering causal relationships from genomic data is the ultimate goal in gene regulation research. constraint based causal exploration algorithms, such as PC, FCI, RFCI, PC-simple, IDA and Joint-IDA have achieved significant progress and have many applications. However, their applications in bioinformatics are still limited due to their high computational complexity. In this paper, we present an R package, ParallelPC, that includes the parallelised versions of these causal exploration algorithms and 12 different conditional independence tests for each. The parallelised algorithms help speed up the procedure of experimenting large biological datasets and reduce the memory used when running the algorithms. Our experiment results on a real gene expression dataset show that using the parallelised algorithms it is now practical to explore causal relationships in high dimensional datasets with thousands of variables in a personal multicore computer. We present some typical applications in bioinformatics using different algorithms in ParallelPC. ParallelPC is available in CRAN repository at https://***/web/packages/ParallelPC/***.

关键词： Causality discovery Bayesian networks Parallel computing constraint-based methods

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：