minimum AIC value indicates the preferred model
Note that likelihood is proportional to sample size. So, AIC comparison between different samplesize are not meaningful.
https://en.wikipedia.org/wiki/Akaike_information_criterion#How_to_apply_AIC_in_practice
Showing posts with label modeling. Show all posts
Showing posts with label modeling. Show all posts
Tuesday, July 17, 2018
Wednesday, March 22, 2017
Wednesday, December 14, 2016
Friday, June 17, 2016
Measurement and Modeling of Drugs in Tissues (Organized in Cooperation with International Society of Anti-Infective Pharmacology (ISAP))
http://www.abstractsonline.com/pp8/#!/4060/session/147
Program Planner Home Print Page
SESSION 044-C - Measurement and Modeling of Drugs in Tissues (Organized in Cooperation with International Society of Anti-Infective Pharmacology (ISAP))
| June 17, 2016, 12:45 - 4:15 PM | Westin, Grand Ballroom A |
Anti-infective pharmacokinetics and pharmacodynamic (PK/PD) modeling traditionally relies on serum or plasma concentration-time data to define exposure-response relationships. Data derived from the blood compartment may or may not accurately reflect target tissue data. This course aims to review current and novel approaches to better quantify target tissue PK/PD. An improved understanding of tissue PK/PD measurement and modeling can help to optimize dosage regimen selection for infections in accessible and restricted compartments.
7 Presentations
| 12:45 - 4:15 PM | - Workshop Convener
David R. Andes; Wisconsin Univ., Madison, WI
| |
| 12:45 - 4:15 PM | - Workshop Convener
Jodi Lestner; Univ. of Liverpool, Liverpool, United Kingdom
| |
| 12:45 - 4:15 PM | - Workshop Faculty
David R. Andes; Wisconsin Univ., Madison, WI
| |
| 12:45 - 4:15 PM | - Workshop Faculty
Jodi Lestner; Univ. of Liverpool, Liverpool, United Kingdom
| |
| 12:45 - 4:15 PM | - Workshop Faculty
Markus Zeitlinger; Med. Univ. Vienna, Vienna, Austria
| |
| 12:45 - 4:15 PM | - Workshop Faculty
George L. Drusano; Univ. of Florida, Lake Nona, FL
| |
| 12:45 - 4:15 PM | - Workshop Speaker
Keith A. Rodvold; Univ. of Illinois at Chicago, Chicago, IL
|
Powered by cOASIS, The Online Abstract Submission and Invitation System SM
© 1996 - 2016 CTI Meeting Technology - All rights reserved.
Feedback
© 1996 - 2016 CTI Meeting Technology - All rights reserved.
Feedback
Tuesday, August 4, 2015
DREAM challenge, open science
wisdom of crowds for robust gene network
nature paper from broad
DREAM project
http://www.ncbi.nlm.nih.gov/pubmed/22796662
http://dreamchallenges.org/upcoming-challenges/
Synapse
https://www.synapse.org/
Thursday, January 30, 2014
Hill coefficient and coperativity, Hill equation
Hill equation
fraction of occupied sites = 1 /((K/L)^n+1) for positive n
When n -> -n,
the Hill equation become:
fraction of occupied sites = 1/ ((L/K)^n + 1)
which can be used to model inhibition.
http://en.wikipedia.org/wiki/Hill_equation_%28biochemistry%29
fraction of occupied sites = 1 /((K/L)^n+1) for positive n
When n -> -n,
the Hill equation become:
fraction of occupied sites = 1/ ((L/K)^n + 1)
which can be used to model inhibition.
http://en.wikipedia.org/wiki/Hill_equation_%28biochemistry%29
Tuesday, December 17, 2013
*** Binomial mortality model, survival function for network aging modeling !!!!
INVERSE Function of binomial CDF
Friday, July 26, 2013
Baudisch 2011 Methods in Ecology and Evolution, The pace and shape of ageing
Baudisch 2011, The pace and shape of ageing.
Baudisch11 standardized survive curves by normalizing age by its own expectation: x / E[x] . This is a very good idea and can be used to compare survive curves in different species with different time scale. For my work, it makes comparing yeast replicative and chronological lifespan possible.
Baudisch11 argues that 'shape' is a unit-less measure, and 'pace' is basically 'rate' with unit 1/time.
Shape measures discussed are:
Omega/L (Omega is age at 1% viability, and L is the average age).
mu(Omega) / mu(0) or \bar{mu}
mu(L)/mu(0) or \bar{mu}
Pace measures argued by Baudisch11 seem to include the two canonical Gompertz parameters.
In its Figure 3, L, Omega, and Maturity seem to be considered as 'pace' measures, too.
Baudisch11 uses 'mortality' for 'mortality rate', which can be seen in her description of the Gompertz model.
Baudisch11 discussed a measure proposed by Ricklef1998 and argued that it is problematic.
It can be seen that x/E[x] is unitless. So, Baudisch11 approach is a nondimensionalization treatment.
Baudisch11 discussed some previous work on dimensionless analysis of aging: Pearl and Miner 1935, Eakin 1994.
Numerically, it is straightforward to calculate the 'shape' measures. However, it is not straightforward to find the analytic form of the shape measures based on the Gompertz or Weibull models.
The median lifespan, i.e., the 50% quantile, has a analytic solution. So, normalization by the median lifespan can be used for both theoretical and empirical analysis.
Baudisch11 standardized survive curves by normalizing age by its own expectation: x / E[x] . This is a very good idea and can be used to compare survive curves in different species with different time scale. For my work, it makes comparing yeast replicative and chronological lifespan possible.
Baudisch11 argues that 'shape' is a unit-less measure, and 'pace' is basically 'rate' with unit 1/time.
Shape measures discussed are:
Omega/L (Omega is age at 1% viability, and L is the average age).
mu(Omega) / mu(0) or \bar{mu}
mu(L)/mu(0) or \bar{mu}
Pace measures argued by Baudisch11 seem to include the two canonical Gompertz parameters.
In its Figure 3, L, Omega, and Maturity seem to be considered as 'pace' measures, too.
Baudisch11 uses 'mortality' for 'mortality rate', which can be seen in her description of the Gompertz model.
Baudisch11 discussed a measure proposed by Ricklef1998 and argued that it is problematic.
It can be seen that x/E[x] is unitless. So, Baudisch11 approach is a nondimensionalization treatment.
Baudisch11 discussed some previous work on dimensionless analysis of aging: Pearl and Miner 1935, Eakin 1994.
Numerically, it is straightforward to calculate the 'shape' measures. However, it is not straightforward to find the analytic form of the shape measures based on the Gompertz or Weibull models.
The median lifespan, i.e., the 50% quantile, has a analytic solution. So, normalization by the median lifespan can be used for both theoretical and empirical analysis.
Thursday, May 9, 2013
Periodical cicadas and birds, student math modeling project
Student math modeling project:
This idea is based on Carl Zimmer's NY Time's writing:
http://www.nytimes.com/2013/05/14/science/marvels-and-a-few-mysteries-in-cicadas-17-years.html?ref=science&_r=0
This oscillation can be done by modifying the predator-prety model with synchronization mechanisms.
This idea is based on Carl Zimmer's NY Time's writing:
http://www.nytimes.com/2013/05/14/science/marvels-and-a-few-mysteries-in-cicadas-17-years.html?ref=science&_r=0
This oscillation can be done by modifying the predator-prety model with synchronization mechanisms.
Monday, April 8, 2013
VCell exercise note, the beginner version
Start VCell and login.
To create a new model, choose File->New-Biomodel,
In the structure diagram menu, click "species" icon, and use the 'arrow' to add a reaction.
Click reaction, and more window will show.
Right click a parameter that I want to change, 'Kf', and a window will show.
Right click 'Application' to add a new one.
Under "Parameters and Functions", we can specify them again.
Under Simulations, click to add a new simulation.
Click the edit icon (a notepad with pencil) to edit the simulation,
Then click "Run", the green arrow.
Use Frap
Choose the two
To create a new model, choose File->New-Biomodel,
In the structure diagram menu, click "species" icon, and use the 'arrow' to add a reaction.
Click reaction, and more window will show.
Right click a parameter that I want to change, 'Kf', and a window will show.
Right click 'Application' to add a new one.
Under Simulations, click to add a new simulation.
Then click "Run", the green arrow.
Use Frap
Choose the two
Subscribe to:
Posts (Atom)