Computational Modeling - Science topic

yes its good

Hi Moustapha Alhaj Dibo , please visit the Conference website（English) for more details: https://ais.cn/u/iyMvqe

The thing is, computational models usually show the most stable state, but at any temperature above 0K there will be some activated state that has alternate geometry. In organic synthesis the usual way to build this cycle would be to use some metal catalyst that will form a complex with both double bonds in suitable positions and then oxidize it.

whether the article will be included in the IEEE Xplore library?

Davide Evangelista Material is failing.

Finite element analysis (FEA) and computational fluid dynamics (CFD), play a crucial role in the precise characterization and optimization of thermal courses.

1. Detailed Simulation: FEA and CFD allow for detailed simulation of heat transfer within building components, enabling a thorough understanding of thermal bridging effects.

2. Identification of Weak Points: These modeling techniques help identify areas of high heat flow or thermal bridging within complex assemblies, pinpointing potential weak points that need optimization.

3. Optimization: By simulating different scenarios and configurations, FEA and CFD assist in optimizing building designs to minimize thermal bridging and improve overall energy efficiency.

4. Cost-Effective Solutions: Computational modeling helps in evaluating different solutions cost-effectively before physical implementation, leading to more efficient and sustainable building designs.

I'm forwarding the AI's response to the forum.

ChatGPT 3.5

📷

"You

Здравствуйте, уважаемая GPT 3.5 Я прошу вас сделать приближенный расчет силы притяжения Луны к Земле.

ChatGPT

ChatGPT

Конечно, с удовольствием помогу с этим расчетом.

Сила притяжения между Луной и Землей можно вычислить с использованием закона всемирного притяжения Ньютона. Формула для расчета этой силы выглядит следующим образом:

�=�×�1×�2�2F=G×r2m1​×m2​​

Где:

Масса Луны �1≈7.35×1022 кгm1​≈7.35×1022кг, а масса Земли �2≈5.972×1024 кгm2​≈5.972×1024кг. Расстояние между центрами масс Луны и Земли составляет приблизительно �≈384,400 кмr≈384,400км (среднее расстояние между ними).

Подставляя значения в формулу, мы можем вычислить приближенную силу притяжения:

�=6.674×10−11×7.35×1022×5.972×1024(384,400×103)2F=6.674×10−11×(384,400×103)27.35×1022×5.972×1024​

�≈1.99×1020 НF≈1.99×1020Н

Таким образом, приближенная сила притяжения между Луной и Землей составляет около 1.99×1020 Н1.99×1020Н.

📷

You

Напомните, пожалуйста размерность Н.

ChatGPT

Конечно, единица измерения "Н" обозначает ньютон, который является единицей измерения силы в системе Международной системы единиц (СИ). Сила, измеряемая в ньютонах, определяется как масса, умноженная на ускорение:

1 Н=1 кг×1 мс21Н=1кг×с21м​

Это означает, что один ньютон — это сила, которая приложена к массе в один килограмм и приводит её к ускорению в один метр в квадрате за секунду (м/с²).

📷

You

Большое спасибо. Объедините, пожалуйста первый ответ и второй и переведите его на английский. Я это отправлю ученым и студентам на форуме ученых.

ChatGPT

"

Certainly! Here's the combined response in English:

Hello esteemed GPT 3.5,

I request you to make an approximate calculation of the gravitational force of the Moon on Earth.

Certainly, I'll assist you with this calculation.

The gravitational force between the Moon and Earth can be calculated using Newton's law of universal gravitation. The formula for calculating this force is as follows:

�=�×�1×�2�2F=G×r2m1​×m2​​

Where:

The mass of the Moon �1≈7.35×1022 kgm1​≈7.35×1022kg, and the mass of the Earth �2≈5.972×1024 kgm2​≈5.972×1024kg. The distance between the centers of mass of the Moon and Earth is approximately �≈384,400 kmr≈384,400km (average distance between them).

Substituting the values into the formula, we can calculate the approximate gravitational force:

�=6.674×10−11×7.35×1022×5.972×1024(384,400×103)2F=6.674×10−11×(384,400×103)27.35×1022×5.972×1024​

�≈1.99×1020 NF≈1.99×1020N

Therefore, the approximate gravitational force between the Moon and Earth is about 1.99×1020 N1.99×1020N.

Thank you! If you have any further questions or need clarification, feel free to ask."

Как Вы полагаете, этой силы достаточно, чтоб создать прилив в океане?

GPT:

Да, сила притяжения между Луной и Землей вполне достаточна для создания приливов в океане. Этот феномен называется приливами, и он обусловлен гравитационным влиянием Луны и, в меньшей степени, Солнца на Землю.

Приливы возникают из-за разницы в силе притяжения гравитационных полей Луны и Земли на разных сторонах Земли. Как следствие, вода в океанах перемещается в направлении, которое выравнивает эту разницу, создавая приливные волны.

Таким образом, сила притяжения Луны играет ключевую роль в создании приливов на Земле.

Yes, the gravitational force between the Moon and Earth is indeed sufficient to create tides in the ocean. This phenomenon is called tides, and it is caused by the gravitational influence of the Moon and, to a lesser extent, the Sun on Earth.

Tides occur due to the difference in gravitational force between the Moon and Earth's gravitational fields on different sides of the Earth. As a result, water in the oceans moves in a way that equalizes this difference, creating tidal waves.

Thus, the gravitational force of the Moon plays a key role in generating tides on Earth.

Below is not clear to me:

what kind of a resistance you are looking for.

Interested

Thanks, Mohammed Yaseen Alzahawi , I am of the same opinion, however, I am more interested in the underlying rationale.

Dear , thanks for your elaboration on the query. With my limited understanding and what the literature outlined, it depends on the VSA (Visual-spatial abilities) capability of the individuals. Individuals with high VSA are able to process spatially complex procedures while still retaining sufficient cognitive resources to benefit from using 3D visualisation while learning. With low VSA, individuals utilise more cognitive resources when performing a spatially complex task. This results in an increase in cognitive load as they learn. Therefore, in individuals with low VSA, it may be possible to compensate for the higher cognitive demands by improving instructional methods.

Dear Mohammadali,

Take a look at the following papers:

-

-

-

Bests,

Mojtaba

I think you are asking about how to tell if two or more different inhibitors in the essential oil are binding to the enzyme simultaneously in different sites.

To answer this question, you will have to fractionate the oil to purify the individual inhibitory components. This would probably be done using some type of chromatography. Individual fractions from the chromatography would then be tested for inhibition of the enzyme in an enzyme activity assay.

Once you have the individual inhibitors purified, you will be able to use steady-state kinetics, equilibrium binding, and/or structural biology to determine whether any pair of inhibitors can bind at the same time.

You need to remove lipid before MD simulation. You can not delete or add any atom/residue/molecule during and after MD simulation, as it will destroy your trajectory data.

When dealing with QM calculations on chemical reactions, one basic problem is thermochemistry. The evaluation of thermodynamic parameters allows a theoretical determination of enthalpy, free energy, and entropy variations associated with the reaction and hence reaction constant. If this is your goal, I suggest strongly this paper, available for Gaussian but whose concept can be easily generalised:

I hope this may be a good starting point (as it was for me!).

Superimpose using spdbv or pymol or you can stitch both the fragments using Python programming.

Try PS2 http://ps2.life.nctu.edu.tw/

cph http://www.cbs.dtu.dk/services/CPHmodels/

phyre2 http://www.sbg.bio.ic.ac.uk/phyre2/html/page.cgi?id=index

I-TSSER: https://zhanggroup.org/I-TASSER/

Best!!

Try BiGGR

dynamicFBA

fbar

sybilDynFBA

Have fun and Happy Research!

Opinion that use SIESTA code which is available free and complete tutorials on that are given below in the links:

Dear Erika, there is something better than Rt that has many limitations, in the case of mood the major limitations is the supposed ergodicity at the basis of Rt that is seldom verified in case of epidemics and never verified by mood spreading that happens on networks or, in any case, with very different spreading potential of the different 'infected' units.

This is why you should look at the papers related to the spreading of a signal across a social network, see for example:

Ciao

Alessandro

One place to get a sense (census?) of things related to the Pezzulo paper is to do a forward literature search: https://scholar.google.com/scholar?cites=14415125872999652537&as_sdt=400005&sciodt=0,14&hl=en

244 citations in 10 years. You can get a sense of the movement in the field by classifying these citations in various ways.

Hi Abhay Vincent , I'm not sure if you have been able to resolve this yet, but there are few things that need to be clear:

1) How do you initialize the configuration, since dealing with water would be different from initializing your particles in any specific lattice.

2) What potential are you using? You can go through sample water simulation scripts here in the tutorial by Christopher Brien: https://sites.google.com/a/ncsu.edu/cjobrien/tutorials-and-guides/working-with-water-in-lammps

Once, you have that, npt ensemble will allow you to linearly ramp up the temperatures. Ideally, I have worked with cooling rates of 1e0 to 1e2 K/ps. You might want to reduce your heating rate but do check the potential you're using and the way the configuration is defined for your simulation.

Dealing with H and O atoms separately along with an SPC or TIP3P potential will also help you in identifying other effects, that may or may not be the target of your current analysis.

consult

The pressure and the normal components of the velocity have to be continuous across the boundaries. See for example in Fundamentals of Acoustics by Kinsler, Frey, Coppens, and Sanders Chapter 6.

you may try our webserver :) John Tainer http://rnabinding.com/P3DOCK/P3DOCK.html

It is certainly possible that you simply need to use different software (as suggested by the other answers), or perhaps the scoring functions used by restrained vs. global docking are different for this particular docking algorithm (making it difficult to truly compare the different outputs); however, I think there is an important point to address in your question regarding the two sampling approaches (i.e. restrained vs. global).

If you perform a restrained search such that you are sampling fewer completely new protein-protein docking orientations, then your iterations are used to optimize the local interface. In other words, you are sampling deeply in a confined region of space. If you perform a completely global search, then you will identify many more completely different protein-protein docking orientations, but none of them will likely have optimal local interactions. In other words, you are sampling broadly in a large region of space.

If you have data that you can use for restraints (e.g. known protein-protein contacts, co-evolved residues for a conserved interaction, experimental mutagenesis, NMR chemical shifts, etc.), then use them as it will almost certainly improve the prediction. If you do not have any knowledge or data to guide you, and you are truly performing "blind" docking, then I recommend you take the different solutions from the blind/global docking experiments and choose a subset with which to perform restrained docking (e.g. by identifying different clusters, ranking by energy, fitting to known data, etc.).

If you are looking to have more control over your docking experiments than a webserver provides, I strongly encourage using the Rosetta macromolecular modeling software package. There are many protocols available and extensive online documentation. Rosetta is very useful for a number of tasks including protein-protein docking.

I hope this helps. Best,

Ben

https://www.embodi3d.com/

Congratulations on completing your PhD program, I will be happy to collaborate in your study. I am a research assistant at the Rush Alzheimer's Disease Center and will be able to get some large pool of data, and analysis if that will help.

hi George Buzzell, I am interested in this. Please check your DM.

I reckon that cDFT should be a reasonable alternative to separate induction from charge transfer:

DOI: 10.1021/acs.jctc.6b00155

J. Chem. Theory Comput. 2016, 12, 2569−2582

I do recommend to take into consideration AI methods. More specifically machine learning methods like k-NN --- stands for k nearest neighbors --- might bring deeper insight into your problematic.

What I recommend is to study similar problems already resolved by this method and similar ML methods and try to understand the way of implementation into your specific problem.

The best way to tackle the problem is to use Python and relevant libraries that already contain all necessary ML methods. When you are novel in the field, try to find someone around you who us familiar with Python programming. He/she can speed up your learning curve substantially.

Some review papers on this topic would be an excellent start.

To Mr.Jiří Kroc

Yes, especially the application of general theory of CAS that require further development in relation to the specific conditions of management of the organizations.

Kind regards,

Dragoljub

Dear Dr Tomic,

I like your question. I am sharing with you the following video link:

Notice also the very close anatomical intimacy and the hugging between limbic system and ventricular system (CSF mysterious waves)

Regards,

Omer

Hello Libin,

as the equation is nonlinear, you cannot use a simple least squares estimator anymore, because the parameters you want to find are not linearly separable.

So if you want to fit measured data to this model, you have a nonlinear least squares problem. There are several ways to solve this problem:

1. This is the simplest one: use the gradient descend method (also known as backpropagation in ANN). This method only finds a local optimal solution (depending on the function and initialization point), so there might be a set of parameters, which fits the problem better. ( https://en.wikipedia.org/wiki/Gradient_descent )

2. You can also use the Levenberg-Marquardt algorithm, which converges faster than the gradient descend method, but also only finds a local optimum ( https://en.wikipedia.org/wiki/Levenberg–Marquardt_algorithm )

3. use a Maximum likehood estimator: https://en.wikipedia.org/wiki/Maximum_likelihood_estimation

4. Use a global optimization algorithm like the particle swarm optimization or an evolutionary algorithm, which can find the global optimal set of parameters with a very high probability. ( https://en.wikipedia.org/wiki/Particle_swarm_optimization )

Your optimization goal for1,2 and 4 is to minimize the residual sum of squares ( https://en.wikipedia.org/wiki/Residual_sum_of_squares ), which is defined by the deviation of the measured data compared to the model output with the actual parameter set.

Dear Fateme Saberipirooz ,

It can be hard to find a specific dataset to use for make computational models or to even experiment on it.

The following list does contain great datasets for experimentation and it contains a description as well, usage examples and in some cases the used codes. It is focused on machine learning problems:

1. Kaggle datasets. This is by far my favourite dataset location. Basically, each dataset is a small community where you can find discussions about the data, find some public code or create your own projects in Kernels. Sometimes you can find notebooks with algorithms that solve the prediction problem in this specific dataset.

Link: https://www.kaggle.com/datasets

2. Google's datasets search engine. This is other of my favourite dataset locatios. The last year, Google launched this great service that consists in a toolbox that can search for datasets by name. Their main objective is to unify thousands or maybe millions of different repositories for datasets around the world and make that data discoverable.

Link: https://toolbox.google.com/datasetsearch

3. Microsoft datasets. Also the last year, Microsoft launches its “Microsoft Research Open Data”. It contains a data repository in the cloud dedicated to facilitating collaboration across the global research community. It offers a considerable amount of datasets that were used in published research studies.

Link: https://msropendata.com/

4. Github collection. This is a other great source of datasets organized by topics, such as Economics, Finance, Education, Transportation, among others. Most of the datasets listed there are free, but you should always check the licensing requirements before using any dataset.

Link: https://github.com/awesomedata/awesome-public-datasets

5. Some government datasets. Finally, it’s easy to find government-related datasets as well. Many countries have shared datasets to the public as an exercise of transparency. For instance:

There are more datasets in some universities webpages that you can search. In conclusion, there is a global direction towards making more data available to the research community. Those new datasets’ communities will continue to grow and make the data easily accessible so that the crowdsource and the computer science community can continue to innovate fast and bring more creative solutions to every single problem.

Thanks Martin

I have tried this option, and hope the calculations will be fine.

Dr. Magnus

Thanks for your reply.

Yes. I follow the same steps for geometry optimisation using Spartan Student version 6; however, I am performing conformer distribution computations which I can't find how to pause and continue later.

By coading u can do it

algorithm which was the base of my proper algorithm for python abaqus features extraction (I attach part of it). You should take into acccount that stresses are determined at integration points, not at the nodes. For viewing them at nodes, abaqus uses an extrapolation algorithm.

Therefore, my advice is to extract it at integration points or at the element centroid (this is the way used on my algorithm).

I hope you'll find it useful, it took some hard work. You'll only need to add the file writing on it.

# Get ABAQUS interface

from abaqus import *

from abaqusConstants import *

# Open the odb

myOdb = session.openOdb(name=odbPath)

session.viewports['Viewport: 1'].setValues(displayedObject=myOdb)

# Get the frame repository for the step, find number of frames (starts at frame 0)

frames = myOdb.steps[stepName].frames

numFrames = len(frames)

# Isolate the instance, get the number of nodes and elements

myInstance = myOdb.rootAssembly.instances[instanceName]

numNodes = len(myInstance.nodes)

numElements = len(myInstance.elements)

for el in range(0,numElements):

   # Isolate current and previous element's stress field

Stress=myOdb.steps[stepName].frames[fr].fieldOutputs['S'].getSubset(region=myInstance.elements[el],position=CENTROID, elementType='C3D8H').values

   sz = len(Stress)

   for ip in range(0,sz):

      Sxx = Stress[ip].data[0]

      Syy = Stress[ip].data[1]

      Szz = Stress[ip].data[2]

      Sxy = Stress[ip].data[3]

      Sxz = Stress[ip].data[4]

      Syz = Stress[ip].data[5]

myOdb.close()

I have received it and will study. I will inform you in the nearest time.

Borland and colleagues did wonderful work on this topic so if you are interested have a look at the following papers:

1. Borland R, Partos TR, Yong H-H, Cummings KM, Hyland A. How much unsuccessful quitting activity is going on among adult smokers? Data from the International Tobacco Control Four Country cohort survey. Addiction. 2012;107(3):673-82.

2. Herd N, Borland R. The natural history of quitting smoking: findings from the International Tobacco Control (ITC) Four Country Survey. Addiction. 2009;104(12):2075-87.

3. Herd N, Borland R, Hyland A. Predictors of smoking relapse by duration of abstinence: findings from the International Tobacco Control (ITC) Four Country Survey. Addiction. 2009;104(12):2088-99.

4. Partos TR, Borland R, Yong HH, Hyland A, Cummings KM. The quitting rollercoaster: how recent quitting history affects future cessation outcomes (data from the International Tobacco Control 4-country cohort study). Nicotine Tob Res. 2013;15(9):1578-87.

If your question concerns binding or rather visual links, in brief and very general (broad) sense the response is yes. However, the proposition goes much deeper and attempts to resolve visual generalization dilemma using relative positional potentials of visible objects. In our approach, time and energy challenge more classical models following visual feature aspects of objects, which indeed requires various unrealistic prerequisites (e.g., initial conditions, sampling,....). Any assumption based on pre-processes of sensory data fall short in resolving hard problems of visual general transformations. This includes relation binding.....

refer the book by Logan..first course in finite element method..

Dear Rahman,

I see that you look for steric effects that the chemists that we collaborate use for obtaining the nanoparticles complementing the sol-gel technique. Perhaps if you look for the papers of Arturo Lopez-Quintela you sure would find applications.

Factors

· Formalization

· Structure

· A first attempt

· A simple transducer

· The ultimate computing model

· transition function

· auxiliary Memory

I found these are factors

The R script below illustrates the nested versus non-nested (crossed) random effects functionality in the R packages lme4 and nlme. Note that crossed random effects are difficult to specify in the nlme framework. Thus, I've included a back-of-the-envelope (literally a scanned image of my scribble) interpretation of the 'trick' to specifying crossed random effects for nlme functions (i.e., nlme and lme).

## This script illustrates the nested versus non-nested

## random effects functionality in the R packages lme4 (lmer)

## and nlme (lme).

library("lme4")

library("nlme")

data("Oxide")

Oxide <- as.data.frame(Oxide)

## In the Oxide data, Site is nested in Wafer, which

## is nested in Lot. But, the latter appear crossed:

xtabs(~ Lot + Wafer, Oxide)

## So, create a variable that identifies unique Wafers

Oxide$LotWafer <- with(Oxide, interaction(Lot,Wafer))

## For continuous response 'Thickness',

## fit nested model E[y_{ijk}] = a + b_i + g_{ij}

## for Lot i = 1:8 and Wafer j = 1:3 and Site k = 1:3

## where b_i ~ N(0, \sigma_1)

## g_{ij} ~ N(0, \sigma_2)

## and b_i is independent of g_{ij}

## The following four models are identical:

## lme4

lmer(Thickness ~ (1 | Lot/Wafer), data=Oxide)

lmer(Thickness ~ (1 | Lot) + (1 | LotWafer), data=Oxide)

## Note: the equivalence of the above formulations makes

## clear that the intercept indexed by Wafer within Lot

## has the same variance across Lots.

## nlme

lme(Thickness ~ 1, random= ~1 | Lot/Wafer, data=Oxide)

lme(Thickness ~ 1, random=list(~1|Lot, ~1|Wafer), data=Oxide)

## Note: the second formulation illustrates that lme assumes

## nesting in the order that grouping factors are listed. I

## think that this was a poor implementation decision, and

## that the latter should indicate non-nested grouping.

## Fit non-nested model E[y_{ijk}] = a + b_i + g_j

## for Lot i = 1:8 and Wafer j = 1:3 and Site k = 1:3

## where b_i ~ N(0, \sigma_1)

## g_j ~ N(0, \sigma_2)

## and b_i is independent of g_j

## lme4

lmer(Thickness ~ (1 | Lot) + (1 | Wafer), data=Oxide)

lmer(Thickness ~ (1 | Wafer) + (1 | Lot), data=Oxide)

## nlme: There is no 'easy' way to do this with the nlme package,

## and I couldn't get this to work at all with the nlme function.

## This is a trick that gives a random slope for each level of the

## grouping variables, which are indexed by the levels of

## a dummy grouping variable with only one group. We also

## specify, for each grouping factor, that covariance

## matrix is proportional to the identity matrix.

Oxide$Dummy <- factor(1)

Oxide <- groupedData(Thickness ~ 1 | Dummy, Oxide)

lme(Thickness ~ 1, data=Oxide,

random=pdBlocked(list(pdIdent(~ 0 + Lot),

pdIdent(~ 0 + Wafer))))

The image below is my interpretation of the nlme (lme) trick for non-nested (crossed) random effects. The idea is to assign a random slope (no intercept) to each level of the grouping factors, which are each indexed by the levels of a dummy variable with that has exactly one level. The pdIdent function ensures that these random effects are uncorrelated and common variance. The pdBlocked function specifies that the random effects are also independent across the two grouping factors.

Simulation-based power analysis using proportional odds logistic regression programming, R, simulation, statistics

Consider planning a clinicial trial where patients are randomized in permuted blocks of size four to either a 'control' or 'treatment' group. The outcome is measured on an 11-point ordinal scale (e.g., the numerical rating scale for pain). It may be reasonable to evaluate the results of this trial using a proportional odds cumulative logit model (POCL), that is, if the proportional odds assumption is valid. The POCL model uses a series of 'intercept' parameters, denoted , where is the number of ordered categories, and 'slope' parameters , where is the number of covariates. The intercept parameters encode the 'baseline', or control group frequencies of each category, and the slope parameters represent the effects of covariates (e.g., the treatment effect).

A Monte-Carlo simulation can be implemented to study the effects of the control group frequencies, the odds ratio associated with treatment allocation (i.e., the 'treatment effect'), and sample size on the power or precision associated with a null hypothesis test or confidence interval for the treatment effect.

In order to simulate this process, it's necessary to specify each of the following:

control group frequencies

treatment effect

sample size

testing or confidence interval procedure

Ideally, the control group frequencies would be informed by preliminary data, but expert opinion can also be useful. Once specified, the control group frequencies can be converted to intercepts in the POCL model framework. There is an analytical solution for this; see the link above. But, a quick and dirty method is to simulate a large sample from the control group population, and then fit an intercept-only POCL model to those data. The code below demonstrates this, using the polr function from the MASS package.

## load MASS for polr()

library(MASS)

## specify frequencies of 11 ordered categories

prbs <- c(1,5,10,15,20,40,60,80,80,60,40)

prbs <- prbs/sum(prbs)

## sample 1000 observations with probabilities prbs

resp <- factor(replicate(1000, sample(0:10, 1, prob=prbs)),

ordered=TRUE, levels=0:10)

## fit POCL model; extract intercepts (zeta here)

alph <- polr(resp~1)$zeta

As in most other types of power analysis, the treatment effect can represent the minimum effect that the study should be designed to detect with a specified degree of power; or in a precision analysis, the maximum confidence interval width in a specified fraction of samples. In this case, the treatment effect is encoded as a log odds ratio, i.e., a slope parameter in the POCL model.

Given the intercept and slope parameters, observations from the POCL model can be simulated with permuted block randomization in blocks of size four to one of two treatment groups as follows:

## convenience functions

logit <- function(p) log(1/(1/p-1))

expit <- function(x) 1/(1/exp(x) + 1)

## block randomization

## n - number of randomizations

## m - block size

## levs - levels of treatment

block_rand <- function(n, m, levs=LETTERS[1:m]) {

if(m %% length(levs) != 0)

stop("length(levs) must be a factor of 'm'")

k <- if(n%%m > 0) n%/%m + 1 else n%/%m

l <- m %/% length(levs)

factor(c(replicate(k, sample(rep(levs,l),

length(levs)*l, replace=FALSE))),levels=levs)

}

## simulate from POCL model

## n - sample size

## a - alpha

## b - beta

## levs - levels of outcome

pocl_simulate <- function(n, a, b, levs=0:length(a)) {

dat <- data.frame(Treatment=block_rand(n,4,LETTERS[1:2]))

des <- model.matrix(~ 0 + Treatment, data=dat)

nlev <- length(a) + 1

yalp <- c(-Inf, a, Inf)

xbet <- matrix(c(rep(0, nrow(des)),

rep(des %*% b , nlev-1),

rep(0, nrow(des))), nrow(des), nlev+1)

prbs <- sapply(1:nlev, function(lev) {

yunc <- rep(lev, nrow(des))

expit(yalp[yunc+1] - xbet[cbind(1:nrow(des),yunc+1)]) -

expit(yalp[yunc] - xbet[cbind(1:nrow(des),yunc)])

})

colnames(prbs) <- levs

dat$y <- apply(prbs, 1, function(p) sample(levs, 1, prob=p))

dat$y <- unname(factor(dat$y, levels=levs, ordered=TRUE))

return(dat)

}

The testing procedure we consider here is a likelihood ratio test with 5% type-I error rate:

## Likelihood ratio test with 0.05 p-value threshold

## block randomization in blocks of size four to one

## of two treatment groups

## dat - data from pocl_simulate

pocl_test <- function(dat) {

fit <- polr(y~Treatment, data=dat)

anova(fit, update(fit, ~.-Treatment))$"Pr(Chi)"[2] < 0.05

}

The code below demontrates the calculation of statistical power associated with sample of size 100 and odds ratio 0.25, where the control group frequencies of each category are as specified above. When executed, which takes some time, this gives about 80% power.

## power: n=50, OR=0.25

mean(replicate(10000, pocl_test(pocl_simulate(50, a=alph, b=c(0, log(0.25))))))

The figure below illustrates the power associated with a sequence of odds ratios. The dashed line represents the nominal type-I error rate 0.05.

Sharing link;

Hi,

There are many applications in Civil Engineering that depends on complex variables and theories. These include Seepage under dams and flow nets. There is an excellent book by Prof. Harr from Purdue University on this subject. Also, the dynamic response of structures due to external loadings such winds and earthquakes also involves complex variables.

Amir Al-Khafaji

Dear Tat'yana,

Dear Mustafa,

Thank you for your response.

20

Thank you Dr Abdallah

De Nardi and Yang have been working with such a model for quite a bit. Here's one of their latest publications on this topic:

Perhaps the survey by Golosov and Tsyvinski also covers some relevant sources:

Regardless the type of the composite material the biggest problem with DMA technique, concerning creep and relaxation experiments, is the size of the specimens. The specimens are too small, resulting in a distorted image of the viscoelastic behavior, due to edge effects, etc. 

The experimental results also depend on what viscoelastic behavior you are examining. Is it linear or nonlinear viscoelastic behavior? Depending on these parameters the limitations vary.

In my opinion, if you wish to study in-depth the viscoelastic behavior composite materials, you need self-made apparatuses. They are easy to construct and can provide more accurate results. However, if this isn't possible, DMA is a good solution, as long as you provide all the information regarding the experimental procedure (loading speed, experimental time etc.) and appropriate literature to support this procedure. Finally, verification of the experimental results with analytical modeling will help you better understand how the materials behave. 

Thank you so much for your reply Bohuslav Uher sir.......

with my team we have modeled electrolyte and fluoride glass with the periodic boundary conditions, but these conditions are used just to reproduce the macroscopic system from an elementary simulation box, as in The MMC an the MD simulations for example. The periodic boundary conditions are not necessarily related to the structure. If you have any technical question on the use of the periodic boundary conditions do not hesitate.

Can you describe the problem a little more? For example in what domain are you averaging? on all the samples, on a subset of samples or on some features?

Jaime,

To my knowledge, within the basic EEGLAB package, you should use an event-related spectral perturbation calculation for this. It has different options, and the default is a wavelet-based decomposition rather than the FFT which if I remember correctly, Pfurtscheller and colleagues utilized in their initial papers describing the outcome. You have the option to use FFT for this data, which may help with lower frequencies. 

I'm afraid I cannot help you with a code for this, however, if you take advantage of the epoch and EEG data structure in EEGLAB, you should be able to write your own code. 

I hope this helps a little bit!

Kind regards,

Chris

Dear.

Please see these links :

Funny that you should ask this precise question ;)

https://www.researchgate.net/post/How_can_I_obtain_spike_timings_in_seconds_OR_mseconds_from_Brian_Hears?

Jansen, Mende's article is simply wonderful! In biology there's a field that study exactly the interconnection between biological and ecological phenomena called allometric scaling and a lot of works were done to explain why the exponents (from powers-law) varies and tend to assume some characteristics values. Mende needs to be know, because he found some universal behavior in growth phenomena with strong mathematical justifying (exponential towers, that I didn't know), suggesting another explanation to exponents and connecting micro-scale behavior between interconnected and independent phenonema to macro-scale outpus. Would be better if mathematicians tried to derive Mende-Peschel differential equation to biological appreciation. Once more, thanks for your readiness, Jansen.

In organization theory, organizations actually usually constitute the agent. See the long tradition of NK papers that started with Levinthal's 1997 paper: http://pubsonline.informs.org/doi/abs/10.1287/mnsc.43.7.934.

Whether a region can be a decision-making unit - well, depends on if it is regional decision making you want to simulate.

The mathematical model is formulated on the basis of theory, review of literature and logical basis. The regression model is generated after formulating the mathematical model as a method for estimating the conditional Y value given the X(s) Value(s). It is important for the regression model to include the error as a result of the incomplete relationship to represent the effects of the other variables that were omitted from the equation.

Thanks alot for your answer

I think that if the comparison between two computational models of emotions (whether the PIA expressed or not faithfully emotions of humans), as we are dealing with in this dialectic interaction may be based, or emanate from the "Conceptual Science Systems" under the agreement that can provide a theoretical framework for understanding, or provide knowledge on the development of a theory or a mental model using rigorous methodologies and a common programming language artificial intelligence "emotional" (PIAe)

TNO Diana.

For steel frame, I used von mises yield criteria with no hardening. But it do not shows perfectly plastic behavior. I have attached the pictures for the reference (the blue  curve is the result from the software and the black curve (I assume) to be the expected result shape only)

Dear Sudipta,

As Hein Van Gils suggested, I believe spatial metrics are a good way to understand not only quantitatively but qualitatively how the distributions are different/similar.

Measures of overlap between kernels are a nice option, since you can measure how much one distribution looks like the other.There are different ways of calculating kernel overlap: calculating the area/length (for 2D/1D kernels) of a given threshold (e.g., 95% kernel) that overlaps betwenn two distributions; calculating the volume/area that overlaps, etc.

A good review for that, in the context of animal movement (but it may be thought for other kinds of data), is the paper by Fieberg and Kochanny 2005:

Also, if you are working with kernels in R, there is a good package that has already operationalized the calculations of kernel overlap: package adehabitatHR, function kerneloverlap(). Take a look here: