�ݺ�ߣshows by User: csgillespie

�ݺ�ߣshows by User: csgillespie / http://www.slideshare.net/images/logo.gif �ݺ�ߣshows by User: csgillespie / Tue, 01 Apr 2014 15:50:19 GMT �ݺ�ߣShare feed for �ݺ�ߣshows by User: csgillespie Bayesian Experimental Design for Stochastic Kinetic Models /slideshow/gillespie-33005092/33005092 gillespie-140401155019-phpapp01
In recent years, the use of the Bayesian paradigm for estimating the optimal experimental design has increased. However, standard techniques are computationally intensive for even relatively small stochastic kinetic models. One solution to this problem is to couple cloud computing with a model emulator. By running simulations simultaneously in the cloud, the large design space can be explored. A Gaussian process is then fitted to this output, enabling the optimal design parameters to be estimated. ]]>
In recent years, the use of the Bayesian paradigm for estimating the optimal experimental design has increased. However, standard techniques are computationally intensive for even relatively small stochastic kinetic models. One solution to this problem is to couple cloud computing with a model emulator. By running simulations simultaneously in the cloud, the large design space can be explored. A Gaussian process is then fitted to this output, enabling the optimal design parameters to be estimated. ]]> Tue, 01 Apr 2014 15:50:19 GMT /slideshow/gillespie-33005092/33005092 csgillespie@slideshare.net(csgillespie) Bayesian Experimental Design for Stochastic Kinetic Models csgillespie In recent years, the use of the Bayesian paradigm for estimating the optimal experimental design has increased. However, standard techniques are computationally intensive for even relatively small stochastic kinetic models. One solution to this problem is to couple cloud computing with a model emulator. By running simulations simultaneously in the cloud, the large design space can be explored. A Gaussian process is then fitted to this output, enabling the optimal design parameters to be estimated. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/gillespie-140401155019-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> In recent years, the use of the Bayesian paradigm for estimating the optimal experimental design has increased. However, standard techniques are computationally intensive for even relatively small stochastic kinetic models. One solution to this problem is to couple cloud computing with a model emulator. By running simulations simultaneously in the cloud, the large design space can be explored. A Gaussian process is then fitted to this output, enabling the optimal design parameters to be estimated.

Bayesian Experimental Design for Stochastic Kinetic Models from Colin Gillespie

]]> 1140 8 https://cdn.slidesharecdn.com/ss_thumbnails/gillespie-140401155019-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 The tau-leap method for simulating stochastic kinetic models /slideshow/the-27257701/27257701 tau-leap-131016120808-phpapp02
]]>
]]> Wed, 16 Oct 2013 12:08:08 GMT /slideshow/the-27257701/27257701 csgillespie@slideshare.net(csgillespie) The tau-leap method for simulating stochastic kinetic models csgillespie <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/tau-leap-131016120808-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" />

The tau-leap method for simulating stochastic kinetic models from Colin Gillespie

]]> 3557 5 https://cdn.slidesharecdn.com/ss_thumbnails/tau-leap-131016120808-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Poster for Information, probability and inference in systems biology (IPISB 2013) /slideshow/poster-24425940/24425940 poster-130719100731-phpapp02
Interest lies in inference for the rate parameters in a complex stochastic biological model describing the aggregation of proteins within human cells. Protein aggregation is a factor in many age-related diseases such as Alzheimer's disease. Ideally time-course measurements on all chemical species in the model would be available. However, current experimental techniques only allow noisy observations on the proportions of cell death at a few time points. Although the model has a large state space and is analytically intractable, realisations from the model can be obtained using a stochastic simulator. The time evolution of a cell can be repeatedly simulated giving an estimate of the proportion of cell death. Unfortunately, simulation from the model is too slow to be used in an MCMC inference scheme. A Gaussian process emulator, which is very fast, can be used to approximate the simulator. An MCMC scheme can be constructed targeting the posterior distribution of interest, however evaluating the marginal likelihood is challenging. A pseudo-marginal approach replaces the marginal likelihood with an easy to construct unbiased estimate while still targeting the true posterior. The methods will be illustrated using a toy birth-death model, allowing comparison with the exact model.]]>
Interest lies in inference for the rate parameters in a complex stochastic biological model describing the aggregation of proteins within human cells. Protein aggregation is a factor in many age-related diseases such as Alzheimer's disease. Ideally time-course measurements on all chemical species in the model would be available. However, current experimental techniques only allow noisy observations on the proportions of cell death at a few time points. Although the model has a large state space and is analytically intractable, realisations from the model can be obtained using a stochastic simulator. The time evolution of a cell can be repeatedly simulated giving an estimate of the proportion of cell death. Unfortunately, simulation from the model is too slow to be used in an MCMC inference scheme. A Gaussian process emulator, which is very fast, can be used to approximate the simulator. An MCMC scheme can be constructed targeting the posterior distribution of interest, however evaluating the marginal likelihood is challenging. A pseudo-marginal approach replaces the marginal likelihood with an easy to construct unbiased estimate while still targeting the true posterior. The methods will be illustrated using a toy birth-death model, allowing comparison with the exact model.]]> Fri, 19 Jul 2013 10:07:31 GMT /slideshow/poster-24425940/24425940 csgillespie@slideshare.net(csgillespie) Poster for Information, probability and inference in systems biology (IPISB 2013) csgillespie Interest lies in inference for the rate parameters in a complex stochastic biological model describing the aggregation of proteins within human cells. Protein aggregation is a factor in many age-related diseases such as Alzheimer's disease. Ideally time-course measurements on all chemical species in the model would be available. However, current experimental techniques only allow noisy observations on the proportions of cell death at a few time points. Although the model has a large state space and is analytically intractable, realisations from the model can be obtained using a stochastic simulator. The time evolution of a cell can be repeatedly simulated giving an estimate of the proportion of cell death. Unfortunately, simulation from the model is too slow to be used in an MCMC inference scheme. A Gaussian process emulator, which is very fast, can be used to approximate the simulator. An MCMC scheme can be constructed targeting the posterior distribution of interest, however evaluating the marginal likelihood is challenging. A pseudo-marginal approach replaces the marginal likelihood with an easy to construct unbiased estimate while still targeting the true posterior. The methods will be illustrated using a toy birth-death model, allowing comparison with the exact model. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/poster-130719100731-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> Interest lies in inference for the rate parameters in a complex stochastic biological model describing the aggregation of proteins within human cells. Protein aggregation is a factor in many age-related diseases such as Alzheimer's disease. Ideally time-course measurements on all chemical species in the model would be available. However, current experimental techniques only allow noisy observations on the proportions of cell death at a few time points. Although the model has a large state space and is analytically intractable, realisations from the model can be obtained using a stochastic simulator. The time evolution of a cell can be repeatedly simulated giving an estimate of the proportion of cell death. Unfortunately, simulation from the model is too slow to be used in an MCMC inference scheme. A Gaussian process emulator, which is very fast, can be used to approximate the simulator. An MCMC scheme can be constructed targeting the posterior distribution of interest, however evaluating the marginal likelihood is challenging. A pseudo-marginal approach replaces the marginal likelihood with an easy to construct unbiased estimate while still targeting the true posterior. The methods will be illustrated using a toy birth-death model, allowing comparison with the exact model.

Poster for Information, probability and inference in systems biology (IPISB 2013) from Colin Gillespie

]]> 881 4 https://cdn.slidesharecdn.com/ss_thumbnails/poster-130719100731-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds document Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Reference classes: a case study with the poweRlaw package /slideshow/user2013/24259763 user2013-130715130620-phpapp01
Power-law distributions have been used extensively to characterise many disparate scenarios, inter alia, the sizes of moon craters and annual incomes. Recently power-laws have even been used to characterize terrorist attacks and interstate wars. However, for every correct characterisation that a particular process obeys a power-law, there are many systems that have been incorrectly labelled as being scale-free. Part of the reason for incorrectly categorising systems with power-law properties is the lack of easy to use software. The poweRlaw package aims to tackles this problem by allowing multiple heavy tail distributions, to be fitted within a standard framework. Within this package, different distributions are represented using reference classes. This enables a consistent interface to be constructed for plotting and parameter inference. This talk will describe the advantages (and disadvantages) of using reference classes. In particular, how reference classes can be leveraged to allow fast, efficient computation via parameter caching. The talk will also touch upon potential difficulties such as combining reference classes with parallel computation.]]>
Power-law distributions have been used extensively to characterise many disparate scenarios, inter alia, the sizes of moon craters and annual incomes. Recently power-laws have even been used to characterize terrorist attacks and interstate wars. However, for every correct characterisation that a particular process obeys a power-law, there are many systems that have been incorrectly labelled as being scale-free. Part of the reason for incorrectly categorising systems with power-law properties is the lack of easy to use software. The poweRlaw package aims to tackles this problem by allowing multiple heavy tail distributions, to be fitted within a standard framework. Within this package, different distributions are represented using reference classes. This enables a consistent interface to be constructed for plotting and parameter inference. This talk will describe the advantages (and disadvantages) of using reference classes. In particular, how reference classes can be leveraged to allow fast, efficient computation via parameter caching. The talk will also touch upon potential difficulties such as combining reference classes with parallel computation.]]> Mon, 15 Jul 2013 13:06:20 GMT /slideshow/user2013/24259763 csgillespie@slideshare.net(csgillespie) Reference classes: a case study with the poweRlaw package csgillespie Power-law distributions have been used extensively to characterise many disparate scenarios, inter alia, the sizes of moon craters and annual incomes. Recently power-laws have even been used to characterize terrorist attacks and interstate wars. However, for every correct characterisation that a particular process obeys a power-law, there are many systems that have been incorrectly labelled as being scale-free. Part of the reason for incorrectly categorising systems with power-law properties is the lack of easy to use software. The poweRlaw package aims to tackles this problem by allowing multiple heavy tail distributions, to be fitted within a standard framework. Within this package, different distributions are represented using reference classes. This enables a consistent interface to be constructed for plotting and parameter inference. This talk will describe the advantages (and disadvantages) of using reference classes. In particular, how reference classes can be leveraged to allow fast, efficient computation via parameter caching. The talk will also touch upon potential difficulties such as combining reference classes with parallel computation. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/user2013-130715130620-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> Power-law distributions have been used extensively to characterise many disparate scenarios, inter alia, the sizes of moon craters and annual incomes. Recently power-laws have even been used to characterize terrorist attacks and interstate wars. However, for every correct characterisation that a particular process obeys a power-law, there are many systems that have been incorrectly labelled as being scale-free. Part of the reason for incorrectly categorising systems with power-law properties is the lack of easy to use software. The poweRlaw package aims to tackles this problem by allowing multiple heavy tail distributions, to be fitted within a standard framework. Within this package, different distributions are represented using reference classes. This enables a consistent interface to be constructed for plotting and parameter inference. This talk will describe the advantages (and disadvantages) of using reference classes. In particular, how reference classes can be leveraged to allow fast, efficient computation via parameter caching. The talk will also touch upon potential difficulties such as combining reference classes with parallel computation.

Reference classes: a case study with the poweRlaw package from Colin Gillespie

]]> 2234 3 https://cdn.slidesharecdn.com/ss_thumbnails/user2013-130715130620-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Introduction to power laws /slideshow/introduction-to-powerlaws/15186908 lecture-121115031759-phpapp01
]]>
]]> Thu, 15 Nov 2012 03:17:57 GMT /slideshow/introduction-to-powerlaws/15186908 csgillespie@slideshare.net(csgillespie) Introduction to power laws csgillespie <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/lecture-121115031759-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" />

Introduction to power laws from Colin Gillespie

]]> 3432 8 https://cdn.slidesharecdn.com/ss_thumbnails/lecture-121115031759-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Moment Closure Based Parameter Inference of Stochastic Kinetic Models /slideshow/gillespie/13939505 gillespie-120810171920-phpapp01
Talk on moment closure parameter which I gave at the SIAM conference on life sciences 2012, http://www.siam.org/meetings/ls12/]]>
Talk on moment closure parameter which I gave at the SIAM conference on life sciences 2012, http://www.siam.org/meetings/ls12/]]> Fri, 10 Aug 2012 17:19:17 GMT /slideshow/gillespie/13939505 csgillespie@slideshare.net(csgillespie) Moment Closure Based Parameter Inference of Stochastic Kinetic Models csgillespie Talk on moment closure parameter which I gave at the SIAM conference on life sciences 2012, http://www.siam.org/meetings/ls12/ <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/gillespie-120810171920-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> Talk on moment closure parameter which I gave at the SIAM conference on life sciences 2012, http://www.siam.org/meetings/ls12/

Moment Closure Based Parameter Inference of Stochastic Kinetic Models from Colin Gillespie

]]> 1182 4 https://cdn.slidesharecdn.com/ss_thumbnails/gillespie-120810171920-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 An introduction to moment closure techniques /slideshow/an-introduction-to-moment-closure-techniques/13395762 cisban08-120620122658-phpapp02
An internal seminar introducing the moment closure technique for stochastic kinetic models]]>
An internal seminar introducing the moment closure technique for stochastic kinetic models]]> Wed, 20 Jun 2012 12:26:55 GMT /slideshow/an-introduction-to-moment-closure-techniques/13395762 csgillespie@slideshare.net(csgillespie) An introduction to moment closure techniques csgillespie An internal seminar introducing the moment closure technique for stochastic kinetic models <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/cisban08-120620122658-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> An internal seminar introducing the moment closure technique for stochastic kinetic models

An introduction to moment closure techniques from Colin Gillespie

]]> 2439 2 https://cdn.slidesharecdn.com/ss_thumbnails/cisban08-120620122658-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Speeding up the Gillespie algorithm /slideshow/speeding-up-the-gillespie-algorithm/13247321 talk-120608045102-phpapp01
A review of the techniques used to make the Gillespie algorithm computationally efficient.]]>
A review of the techniques used to make the Gillespie algorithm computationally efficient.]]> Fri, 08 Jun 2012 04:51:00 GMT /slideshow/speeding-up-the-gillespie-algorithm/13247321 csgillespie@slideshare.net(csgillespie) Speeding up the Gillespie algorithm csgillespie A review of the techniques used to make the Gillespie algorithm computationally efficient. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/talk-120608045102-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> A review of the techniques used to make the Gillespie algorithm computationally efficient.

Speeding up the Gillespie algorithm from Colin Gillespie

]]> 2705 3 https://cdn.slidesharecdn.com/ss_thumbnails/talk-120608045102-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Moment closure inference for stochastic kinetic models /slideshow/ohio-13247290/13247290 ohio-120608044623-phpapp01
My talk from the MBI Workshop on Recent Advances in Statistical Inference for Mathematical Biology 2012 http://www.mbi.osu.edu/2011/rasschedule.html]]>
My talk from the MBI Workshop on Recent Advances in Statistical Inference for Mathematical Biology 2012 http://www.mbi.osu.edu/2011/rasschedule.html]]> Fri, 08 Jun 2012 04:46:22 GMT /slideshow/ohio-13247290/13247290 csgillespie@slideshare.net(csgillespie) Moment closure inference for stochastic kinetic models csgillespie My talk from the MBI Workshop on Recent Advances in Statistical Inference for Mathematical Biology 2012 http://www.mbi.osu.edu/2011/rasschedule.html <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/ohio-120608044623-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> My talk from the MBI Workshop on Recent Advances in Statistical Inference for Mathematical Biology 2012 http://www.mbi.osu.edu/2011/rasschedule.html

Moment closure inference for stochastic kinetic models from Colin Gillespie

]]> 967 3 https://cdn.slidesharecdn.com/ss_thumbnails/ohio-120608044623-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 WCSB 2012 /slideshow/wcsb-2012/13247255 talk-120608044134-phpapp02
My slides from the 9th Workshop on Computational Systems Biology]]>
My slides from the 9th Workshop on Computational Systems Biology]]> Fri, 08 Jun 2012 04:41:32 GMT /slideshow/wcsb-2012/13247255 csgillespie@slideshare.net(csgillespie) WCSB 2012 csgillespie My slides from the 9th Workshop on Computational Systems Biology <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/talk-120608044134-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> My slides from the 9th Workshop on Computational Systems Biology

WCSB 2012 from Colin Gillespie

]]> 498 3 https://cdn.slidesharecdn.com/ss_thumbnails/talk-120608044134-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Bayesian inference for stochastic population models with application to aphids /slideshow/bayesian-inference-for-stochastic-population-models-with-application-to-aphids-3136933/3136933 aphids09-100211122409-phpapp01
]]>
]]> Thu, 11 Feb 2010 12:16:08 GMT /slideshow/bayesian-inference-for-stochastic-population-models-with-application-to-aphids-3136933/3136933 csgillespie@slideshare.net(csgillespie) Bayesian inference for stochastic population models with application to aphids csgillespie <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/aphids09-100211122409-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" />

Bayesian inference for stochastic population models with application to aphids from Colin Gillespie

]]> 825 3 https://cdn.slidesharecdn.com/ss_thumbnails/aphids09-100211122409-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 https://public.slidesharecdn.com/v2/images/profile-picture.png Nothing interesting www.mas.ncl.ac.uk/~ncsg3/ https://cdn.slidesharecdn.com/ss_thumbnails/gillespie-140401155019-phpapp01-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/gillespie-33005092/33005092 Bayesian Experimental ... https://cdn.slidesharecdn.com/ss_thumbnails/tau-leap-131016120808-phpapp02-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/the-27257701/27257701 The tau-leap method fo... https://cdn.slidesharecdn.com/ss_thumbnails/poster-130719100731-phpapp02-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/poster-24425940/24425940 Poster for Information...