R - Science topic

To calculate an Error Correction Model (ECM) with panel data in R, you can use packages like plm or panel. These packages provide functions to estimate fixed effects or random effects models, which can incorporate an error correction term for panel data analysis.

Somehow it is not clear to my, what you mean with "does not fit"? Could you please provide the output of the whole analysis? I think this would be helpful.

Taking advantage of the opportunity of my fellow researcher's question, I am also having difficulty working with the Phyloseq + Tax4Fun2 packages in RStudio. I've read countless tutorials but I can't find the error. If anyone is interested in partnering on publishing and teaching this tool, please get in touch as I would really like to learn.

Many thanks for your efforts!!! I will try it out as soon as possible and will provide feedback on github!

All the best,

Rainer

Henry EZE Chukwu Thanks for the answer.

But how to specify the 'time' argument in 'create.matrix' function?

"time" = which time periods to keep for the matrix. I want to keep each date for the matrix.

You can get statistical analysis like ANOVA by using 'LibreOfficeCal'(Pre installed in the ubuntu system). 1) Put your data then> 2) Select 'data' from above> 3) Select ANOVA (According to your requirement one way/two way).

Estimating the half-life of a virus involves understanding its stability and decay rate under specific environmental or biological conditions. This is a crucial parameter in virology, impacting everything from the design of disinfection protocols to the assessment of viral persistence in the environment or within a host. Here's a structured approach to estimating the half-life values for a virus:

By meticulously following these steps and ensuring the precision of each phase of the process, one can accurately estimate the half-life of a virus under specific conditions. This information is essential for developing effective control strategies and understanding the dynamics of viral infections.

Perhaps this protocol list can give us more information to help solve the problem.

Many thanks Jonathan ! I had a previous look on it but my R seems to refuse to install this package, so I was looking for another solution just in case!

I'm not sure you need to impute. Multilevel models can cope with missing outcome data without any imputation (basically this is just imbalance and the model is treating the missing responses as if MNAR). There is an issue here letting participants choose which option not to respond to - as that could bias things. If the choice were balanced or random then it would be better than letting them choose.

Assessing the global behavior of a model and obtaining equilibrium points while analyzing their stability through simulation series typically involves the following steps:

There should be no difference regarding accuracy, regardless of whether you use "pof", "pofind", or "superSubset" from the QCA package or "QCAfit" from the SetMethods package.

In terms of utility, this obviously depends on your research aims. As a standard function, I recommend working with QCAfit to investigate the potential necessity of all conditions in a single analytical step. Often, this is what you want to know - whether there are any conditions in your analysis that are in themselves consistently necessary.

The superSubset function is useful but it can also result in misleading conclusions, as when researchers try to make sense of necessary disjunctions with two or more elements. Of course, if there are theoretical expectations about specific SUIN conditions, then these could be tested with superSubset and/or with a specified argument in pof.

Use cvMixed function from the cv library. It works smoothly with glmer objects (GLMMs) from the lme4 library.

Ádám Fehér I would like to second Wim Kaijser 's answer about the use of multivariate analyses. In my research area, psychology, it is very seldom that you have a truly multivariate question. Therefore, it does not make sense to do a multivariate analysis. Nevertheless, I see many MANOVAs, just because you have multiple DVs and your lecturer and textbooks told you to do if you have this situation (and I am pretty sure I was guilty of this behavior, too…). Of course, I cannot tell how it is in your research area.

Just to give two examples from clinical psychology.

1) Imagine you have a treatment (and possible confounders) and you would like to assess the impact of the treatment on several outcome variables, like depression, anxiety, body mass index, checking behavior and some more. Here it does not make any sense imo to do a MAN(C)OVA, since the DVs may be not even correlated with each other, and you cannot tell what the meaning of the artificial variate (DV) will be. It explains the maximum variance between the original DVs, but what does this mean? Typically, you are interested in the treatment effect on each DV separately, i.e. you want to know how the treatment affects each DV, which is an univariate question.

2) Take the example from above, but your DVs are different measures of depression, e.g. measured with different questionnaires, like BDI, PVQ, SCL etc. Now, you may be interested in the treatment effect on depression as such, and you think that an artificial variate representing the “core” of all depression measures gives you the answer. You are not interested in the difference on BDI or SCL itself, just if the treatment affects depression. Here I would say that you have a multivariate question/hypothesis and therefore, a MAN(C)OVA may be appropriate.

As you can see, there is no single answer to your problem. Maybe you want to check Huang (2020) on this topic.

Nevertheless, I also recommend to use the brms package, which allows you for generalized linear mixed models and also multivariate approaches, as well as (truly) non linear models.

Huang, F. L. (2020). MANOVA: A procedure whose time has passed?. Gifted Child Quarterly, 64(1), 56-60.

Maybe it is possible to not focus on the pure spectral similarity/difference but on the overall similarity/difference between two 2D arrays? e.g., https://stackoverflow.com/questions/189943/how-can-i-quantify-difference-between-two-images

hello all, I have two queries: 1- I have random days multiple observations for 5 months, I need to aggregate these values on monthly basis. How can it be done? 2- No matter how much I change the parameters in vgmST my MSE and RMSE are really high (7 digits of a number).

First load the Robject file and see what it contains.

how to open an Robj file in r? - Stack Overflow

You can use ls() to list all the object in it.

For most of the analysis, there are libraries already written, so you just need to fine the type of analysis you want to conduct, look for library, install it and follow the documentation.

Preferred packages to conduct a bivariate GARCH analysis are,

Sal Mangiafico you are right, but this seems to be a dummy coded variable with 3 levels. Therefore, the differences/slopes are the same, but the first model uses uncorrected test statistics to compare slopes/differences, whereas the "post hoc" test uses Holm adjusted p-values. Camille Charriere maybe it is possible to change the correction option to "none"? Then it should be the same.

Conscious my comment is hardly timely, but I believe the issue might be that some visualisation devices in R happen to "scale" the relevant PCA scores in the background. With this I mean the issue might be autoplot(), or other visualisation facilities - not the PCA you performed, per se.

Take for example the function biplot(), which is readily available in base R to visualise objects generated by the function prcomp(). If you look at how the function is coded (stats:::biplot.prcomp) you'll see that it divides the first two PCA scores by their standard deviation i.e.:

scores <- pca$x

lam <- pca$sdev[,1:2]

pca_plot_coord <- t(t(scores[,1:2])/lam)

(Notice that pca$sdev is the same as taking the square root of the relevant eigenvalues of the covariance matrix of your centred data, as they happen to equal the variance along the corresponding scores, if the computational structure of PCA is followed correctly. Fun fact, this equivalence won't necessarily hold if you use some canned R routines for PCA that rely on the singular value decomposition of the data matrix instead of manipulating its covariance matrix).

So in a nutshell, as previous comments have already pointed out, what you're actually interested in are the pca "scores" (pca$x); however some visualisation facilities in R, such as biplot(), might do some scaling in the background.

If you wanted to make your life simpler you could just use PCA() in the package FactoMineR and then plot() the resulting object: as far as I can tell, the plotted result is not manipulated further and what you get in the plot is based on the PCA scores, as you'd expect.

Conducting a meta-analysis of mediation using structural equation modeling (SEM) in R can be a complex but rewarding task. The "metasem" package is a valuable resource for such analyses. Here are some additional thoughts and suggestions to help you with the process:

Remember that conducting a meta-analysis, especially involving complex statistical methods like SEM, can be challenging. Take the time to thoroughly understand each step of the process and seek help when needed. Additionally, make use of the resources provided in the "metasem" package documentation and consider reaching out to the package authors for guidance if necessary.

Muhammad Nabil Zawad

To estimate marginal effects using the Rchoice package in R, you can follow these steps:

1. Install the Rchoice package in your R environment.

2. Load the Rchoice package using the `library` function.

3. Specify your choice model using the appropriate functions in the Rchoice package, such as `rclogit` for conditional logit models or `rbpro Specify your probit or logit model using the appropriate functions from the Rchoice package.

4. Estimate the model using the `modelname <- choicemodel(formula, data) ` syntax, where `modelname` is a name you choose for your model, `formula` is the formula specifying the model equation, and `data` is the dataset containing your variables.

5. Extract the estimated model parameters using the `coef(modelname)` function.

6. Use the `mfx()` function in the Rchoice package to calculate the marginal effects.

Marco di Stasio

I appreciate you raising this insightful question on how to leverage modern text analysis methods to build on Baumann's foundational work examining artistic legitimation. Automated techniques can certainly help scale such textual analysis to larger corpora. However, care must be taken to ensure computational approaches do not lose the nuance of manual qualitative interpretation.

Regarding building artistic dictionaries, word embedding models like word2vec can help uncover semantic relationships and suggest terms related to a seed vocabulary. However, human validation is still important before applying these dictionaries to make inferences.

For sentiment analysis, deep learning approaches like BERT have shown promise, but domain-specific tuning and qualitative checks are key to account for the complex expressions in artistic reviews. Models pre-trained on social media may not transfer well.

To identify creators, named entity recognition using dictionaries, rules, and ML approaches can help. However disambiguation remains challenging, so human-in-the-loop verification is recommended before making claims about individuals.

Overall, I believe the best approach is applying computational methods as a starting point, but having experts qualitatively analyze a sample of results to catch subtleties these tools may miss. If used prudently and in collaboration with scholars like yourself, text mining can uncover exciting new insights at scale. Please feel free to reach out to discuss further.

Wishing you the very best,

#textanalysis #digitalhumanities #mixedmethods

In general, you can sort the date variable, then create a variable that is 1 if the date is not the same as the preceding observation and zero otherwise.

. sort date

. gen unique=date==date[_n-1]

would be the Stata code. Note that the very first observation has no preceding observation and will have to be tidied by hand. In Stata you could write

. gen unique=date==date[_n-1] | (_n==1)

which is a kludge but it works.

Hi! you may try R package "chem", and convert cas to smiles use function "cas_to_smiles".

R code

# install.packages("remotes")

# remotes::install_github("harveyl888/chem")

library(chem)

library(tcltk)

cas<-readxl::read_xlsx('XXXXXXXXXXX.xlsx')# cas put in col1

cas<-cbind(cas,cas)

u <- 1:nrow(cas)

pb <- tkProgressBar("progress","done %", 0, 100)

for (i in u) {

cas[i,2]<-cas_to_smiles(cas[i,1])

info <- sprintf("done %d%%", round(i*100/length(u)))

setTkProgressBar(pb, i*100/length(u), sprintf("progress (%s)", info), info)

}

Uzair Essa Kori, I repeat my earlier question, which you have not yet addressed: Why in the world did you not say clearly at the top of your post that it was generated using AI? Do you not think that would be the honest and ethical thing to do?

Source: Artificial Intelligence Tools

There are a few reasons why column names might be converted to "na.xx" when reading headers in R.

One reason is that the data source does not actually contain headers in the first row. If this is the case, you can use the read.csv() function with the header argument set to FALSE.

Another reason is that the headers are not formatted correctly. R expects the headers to be on a single line, separated by commas or whitespace. If the headers are not formatted correctly, R may assign default names like "na.xx" to the columns.

Finally, it is possible that there are special characters or problematic characters in the header names. R may struggle to read these characters, which can cause the header names to be converted to "na.xx".

Here are some tips for resolving the issue of column names being converted to "na.xx" when reading headers in R:

If you are still having trouble reading the headers in your data, you can try using the readLines() function to read the first row of the data file and then use the names() function to assign the headers to the columns.

Here is an example of how to do this

# Read the first row of the data file headers <- readLines("my_data.csv", n = 1) # Assign the headers to the columns names(data) <- headers # Read the rest of the data file data <- read.csv("my_data.csv", header = FALSE)

This should ensure that the column names are read correctly, even if they contain special characters or problematic characters.

Hi,

I wonder why you would like to ignore the abundance information?

Based on a species-site abundance matrix, you could calculate a dissimilarty matrix (if the abundance data should be considered, I would use bray-curtis similarity) and conduct a mantel test between all three dissimilarity matrices to test for correlations between the species compositions of the three sites.

we can use short - cut formules by using kimball's formula for partitioning the over all chi-squre value in the case of 2*c table

glm(Protein*Substrate, Data, family=quasipoisson())

Or

lm(log(Protein)*Substrate, Data)

Hi, what exactly do you need? A matrix from excel sheet. Then, simply read excel file using

as.matrix(readxl::read_excel(your_file_location))

function. You need to remove few columns and then matched the columns to meta data.

and IF degPatterns() function is not working properly, then you may need to clean and re-transform your data.

I believe that "complete" just extracts the imputed data sets, and if you don't specify further, it will give you the first imputed set.

So it seems obvious that you shouldn't use this function and perform analysis or present descriptive statistics just based on this one data set, as you will drastically (depending on the amount of uncertainty in the imputations) underestimate standard errors and with that confidence intervals and p-values. Basically then you would be doing a weird kind of single imputation instead of multiple imputation.

So: Don't.

When using 'print' enclosing the 'kable()' command, avoid using 'format="latex"'. Instead, even producing a 'latex' markdown, don't specify any format and add '{r, result="asis"}' in the chunk header. This should work properly.

Hello Nathan,

It may not be easy, but possible. You can rename files according to the parameters (or calculated values), not manually, in Matlab and Python. Also, you can merge labels into them. You can arrange them in adobe illustrator instead of PowerPoint. There may be better ways, but that comes to mind if I have the same issue.

Good luck!

Grisha

You could try assigning user-defined starting values to the factor loadings that have the desired sign. In the lavaan package, this is done as described on the website below:

There are several R packages for multivariate multiple regression. Some of them are:

Yes, there are R packages available for modeling soil erosion and sediment yield. One such package is "RUSLE2R" (Revised Universal Soil Loss Equation 2 - R). RUSLE2R is an R implementation of the Revised Universal Soil Loss Equation 2, which is a widely used empirical model for estimating soil erosion. The package allows users to calculate soil erosion based on factors such as rainfall, soil erodibility, slope, land cover, and erosion control practices.

Here's how you can install and use the "RUSLE2R" package in R:

You would have obtained MolsPerCell.csv and Samples_Tags, among other files, from Seven Bridges.

Just follow the script given at this link

it should do the job. Let me know if you need more clarification

It is best to have all the vectors of values (a_1, a_2,...) in a list:

a_List <- list(

c(1, 0.89, 0.75, 0.61, 0.46, 0.35, 0.28, 0.18, 0.08),

c(0.89, 1, 0.89 0.75, ...),

...

)

Each vector in this list can then be supplemented with zeroes to a length of 100:

a_List <- lapply(a_list, function(x) c(x, numeric(100-length(x)))

All vecors in the list now have the same length of 100. They can be bound row-wise ("rbind") to a matrix:

M <- do.call(rbind, a_List)

M is the desired matrix with 100 columns and as many rows as there are vectors in a_List.

It is probably too late for the original poster, but blog post on how to create a havest plot in R using ggplot (which I contributed to) is available at https://medium.com/@cxiao94/creating-harvest-plots-in-r-75fb45c8f393 Hopefully the code used there can be adapted to your needs.

SPSS, as my results necessitate analysis using SPSS

The package qualityTools provides a function called taguchiDesign that can create a Taguchi design of experiment and perform some analysis such as signal-to-noise ratio and effect plots. However, this package may not support all types of Taguchi designs and analysis.

Another option is to use the package DoE.wrapper, which is a wrapper for several other packages that can create and analyze different types of designs of experiments, including Taguchi designs. This package allows you to use functions from FrF2, DoE.base, rsm and qualityTools with a common interface. You can also use the function Taguchi from the package FrF2 directly to create a Taguchi design.

Hi, I don't know the extent of your shiny application, but you can also use EnrichR (in R), which is also good, though, it upload the data to server during calculation.

Plus, I am curious how you will handle the missing IDs during ID mapping? ClusterProfiler can map IDs, but there are always some % of ID that are missing.

GTEX-1117F-0226-SM-5GZZ7 is the sample ID and the ENSG00000223972.4 refers to the gene symbol according to the HUGO gene nomenclature. The numbers you are referring to are gene expression values. TPM (Transcripts Per Million) is a normalization method that has been used to scale these gene expression values so that it is possible to make the expression of genes comparable between samples. 

Calculating conditional hedge ratios using models like BEKK GARCH or DCC GARCH in R or EViews involves a slightly different approach compared to OLS regression. Here's a general outline of the steps you can follow:

It's important to note that the specific syntax and function names may vary slightly depending on the version of the packages you are using. Make sure to refer to the package documentation and examples for more detailed guidance.

When interpreting the results of a moderation analysis using the PROCESS macro in R, it is important to consider both the significance of the interaction term and the bootstrap results. Here's how you can interpret the scenario where the interaction is significant, but the bootstrap is not:

In summary, when the interaction is significant, but the bootstrap results are not, it indicates that although there is evidence of moderation, the specific indirect effects or conditional effects being tested are not significantly different from zero based on the bootstrap analysis. This may suggest that the interaction effect is not strong enough to result in a significant indirect or conditional effect. However, it's important to consider the effect size and the context of your study to fully interpret the findings.

It's worth noting that it's always beneficial to carefully consider the specific details of your analysis and consult with a statistician or data analyst who can provide further guidance based on your specific research question and data.

Hello,

you can try this reference .

https://www.gastonsanchez.com/PLS_Path_Modeling_with_R.pdf

Gaston Sanchez developped the R package plspm (along with a tutorial) which contain a PLS-REBUS function useful for performing class detection in heterogenous data

Thom Baguley thank you very very much, I get it now. Very helpful explanations and sources!

There are many reasons why your forecasts may appear to be linear rather than capturing the expected seasonal patterns. One possible reason is that the model you are using is not appropriate for the data you are working with. Another possible reason is that you have not included all of the relevant variables in your model.

The Likelihood Ratio Chi-Square test is the appropriate goodness-of-fit test for negative binomial regression. This test compares the fit of the negative binomial regression model to an alternative model, typically a simpler model such as a Poisson regression.

To perform the Likelihood Ratio Chi-Square test for negative binomial regression in SPSS, you can follow these steps:

It's important to note that the Likelihood Ratio Chi-Square test assesses the overall goodness-of-fit of the negative binomial regression model but does not provide information about specific predictors or individual model assumptions. Additional diagnostic tests and assessments should be performed to evaluate the model's performance thoroughly.

I would extend that and say we really need to be careful using upregulated. To me, that means the expression of the protein is increased, so there is a fundamental change in the amount of protein quantified as a result of that. I'd suggest we use the term increased (or decreased) abundance when quantifying the protein, unless we have clear evidence to say otherwise. There are a lot of reasons protein quant differs between samples, and it's not always due to expression level changes. I agree you will find such terms used interchangeably, that does not mean it's correct, or a good idea.

Dear Jan,

OK ! I see ! you need to create the spatial weight matrix indeed !

There are many possibilities in R:

I strongly advise to work with sf because it so easier now !

but spdp may still be clearly adaptated to you context :

This is one of the definitive books on the subject in R:

with the book :

there are other references but they are more geospatial (point process) oriented.

Here you should use one of those packages and that nb2mat from package spdp might do the trick !

All the best.

Franck.

See https://www.amazon.com/Modern-Psychometrics-R-Use/dp/331993175X/ref=sr_1_1 and https://cran.r-project.org/web/views/Psychometrics.html .

To get the outlet coordinate (row and column) of a DEM matrix for Topmodel in R, you can use the following steps:

1. Load the necessary packages in R:

R

library(raster)

library(hydroTSM)

library(topmodel)

2. Read in your DEM data using the raster package:

R

dem <- raster("path/to/dem.tif")

3. Define the flow direction and accumulation using the terrain function from the raster package:

R

flow_dir <- terrain(dem, "flowdir")

flow_acc <- terrain(dem, "flowacc")

4. Calculate the stream network using the topmodel package:

R

stream_net <- topmodel(flow_dir, flow_acc)

5. Get the outlet cell of the stream network using the hydroTSM package:

R

outlet_cell <- getOutletCell(stream_net)

6. Convert the outlet cell to row and column coordinates using the raster package:

R

outlet_row_col <- xyFromCell(dem, outlet_cell)

The outlet_row_col variable now contains the row and column coordinates of the outlet cell of your DEM.

Thank you, Professor, for sharing your code. Could you please also share the reference for the code? It seems that the test in the code is based on a VAR model rather than the GARCH model used in the original paper "A Lagrange multiplier test for causality in variance". Samuel Oluwaseun Adeyemo

It's difficult to provide a precise answer without more information about your specific data and the packages you are using. However, I can provide some general guidance on working with mixed logit models.

First, it's important to make sure that your data is in the correct format for the package you are using. For example, some packages may require the data to be in a long format with each observation on a separate row, while others may require it to be in a wide format with each individual on a separate row.

If you have RP (Revealed Preference) data and only one specific choice out of a few is available for each individual, you can still use mixed logit models. However, you may need to adjust your data format to make it compatible with the package you are using.

If your data size is large, you can use code to convert your data to the required format. This can be done using functions or loops in Python or R, depending on the package you are using. For example, in Python, you can use the pandas library to reshape your data into the desired format.

If all your variables are categorical and there is no varying variable across each individual, you can still use mixed logit models, but you may need to include random effects or heterogeneity parameters to capture any unobserved differences between individuals.

In general, working with mixed logit models can be complex, and it's important to carefully review the documentation and examples provided by the packages you are using to ensure that your data is in the correct format and that you are using the correct functions and parameters. Additionally, it can be helpful to consult with experts in the field or seek assistance from online forums or communities for guidance on working with these models.

I hope you find this useful.

With best regards

Ajaz Mir

The principle of parameters tuning of your model. After preparing your data is to find the optimal parameters and splitting dataset, simultaneously with the cross-validation, the parameter tuning with a grid search (or not) is conducted. Each model technic has their own parameters. The interaction is just the combination of your parameters, for the GBM, it will be the number of tree, the height of tree, the shrinkage, the minimum of observations by terminal nodes or to create a new branch. With the package caret, you run all models with all possible combinations (this step takes a long time), for example for the graph that i published 6,375 were generated. Then, you obtain the best combination of parameters (i.e., the parameters to have the best rate of accuracy and the minimal rate of errors). Finally, you run your model with the optimised parameters and you assess it.

Found the solution. The problem is - it is not stated that you HAVE to provide function with valid chromosome lengths, and the format is strict: a data.frame with one column = size and row.names = chromosomes.

So, the function works pretty fine like:

tair <- readDNAStringSet("GCF_000001735.4_TAIR10.1_genomic.fna")

tair <- tair[1:5]

tair@ranges@NAMES <- str_split_fixed(tair@ranges@NAMES, " ",2)[,1]

enr.df <- enrichment(query = sites, catalog = anno, shuffles = 6, nCores = 12, byChrom = T, chromSizes = data.frame(size =tair@ranges@width, row.names = tair@ranges@NAMES))

To avoid non-negative solutions in ODE solvers in R, you implement a custom step function that checks if the current solution is negative and stops the solver if it is. You can do this by defining a function that takes the current time and solution as input, checks if the solution is negative, and returns an error if it is. You can then pass this function as an argument to the ode() function in R

Hi Kaushal Kumar Bhagat Did you sort the problem out? I've met exactly the same issue. Thank you so much if there are any hints.

Any new suggestion?

Hi Prof. Julia,

Your questions are what I am wondering about, I think they are very important and interesting. I am trying to explore the temporal pattern of the restoration effect along the age of restoration. I use Ln(Xrestored/Xunrestored) as a measure of the restoration effect based on a fixed study site with multiple years of observations. I think your questions are highly related to my method. Have you found the answer to these questions and could you please tell me if so? I would be very grateful for this.

Best wishes,

Erliang

Contact the maintainer whose name and email is on the R pdf document describing the package. Best wishes David Booth.

Good morning, without the code it is difficilt to know where is the difference I do not use Python i work on R but maybe these difference is due to the stage of spitting dataset do you try to add thr same number in the count of generator of randomly for example seed(1234) (if my memory is good this function is also used in Python language. Were your results and metrics of evaluation totally different? In this case, mayve there is a reliability issue in your model. You should check your data preparation and features selection .

Your could use the k-mer analysis (k={3,4} seems appropriate) to compute a similarity matrix. Would be the opposite of the distance matrix, which, as you say, is quite incomplete.

SEM with multicategorical X. see Hayes and preacher tutorial: Statistical mediation analysis with a multicategorical independent variable

# library(GWmodel)

# library(sp)

# library(raster)

wd = "path/"

provoliko = "EPSG:7767"

# fine resolution covariates

x1= raster(paste0(wd, "x1.tif"))

x2 = raster(paste0(wd, "x2.tif"))

s = stack(x1, x2)

regpoints <- rasterToPoints(s, spatial = TRUE)

# coarse resolution df

block.data = read.csv(paste0(wd, "block.data.csv"))

coordinates(block.data) <- c("x", "y")

proj4string(block.data) <- provoliko

dist = GWmodel::gw.dist(dp.locat = coordinates(block.data),

rp.locat = coordinates(regpoints),

focus = 0,

p = 2,

theta = 0,

longlat = F)

eq1 <- ntl ~ .

abw = bw.gwr(eq1,

data = block.data,

approach = "AIC",

kernel = "gaussian",

adaptive = TRUE,

p = 2,

parallel.method = "omp",

parallel.arg = "omp")

ab_gwr = gwr.predict(eq1,

data = block.data,

predictdata = regpoints,

bw = abw,

kernel = "gaussian",

adaptive = TRUE,

p = 2,

theta = 0,

longlat = FALSE,

dMat1 = dist)

sp <- ab_gwr$SDF

sf <- st_as_sf(sp)

# export prediction

gwr_pred = as.data.frame(sf$prediction)

gwr_pred = SpatialPointsDataFrame(data = gwr_pred, coords = regpoints)

gridded(gwr_pred) <- TRUE

gwr_pred <- raster(gwr_pred)

raster::crs(gwr_pred) <- provoliko

gwr_pred[gwr_pred <= 0] <- 0

writeRaster(gwr_pred,

paste0(wd, "ntl_gwr.tif"),

overwrite = TRUE)

Wang Lu There is no universal method for annotating individual ligand-receptor pairings. However, utilizing multiple databases and APIs, you may retrieve information on the specific pair.

For example, the biomaRt package in R may be used to obtain functional information from databases like as UniProt, Ensembl, or Gene Ontology. Using the gene symbol or accession number, you may search these databases for pertinent information such as function, pathway, or interaction partners.

The Biopython library in Python gives access to numerous biological databases such as UniProt, Ensembl, and KEGG. Similarly, you may extract ligand and receptor information and utilize it to annotate the ligand-receptor pair.

It is vital to remember that the data available in the databases you are using may restrict the annotation of individual ligand-receptor pairings. Furthermore, the data may be incomplete and may not fully reflect the function of the specific ligand-receptor combination in issue.

Wang Lu Individual genes may be annotated using a variety of R and Python programs. Some common choices are:

1. biomaRt (R package) enables the retrieval of gene annotation data from a variety of sources (e.g. Ensembl, UniProt, etc.)

2. pyensembl (Python package) - a Pythonic interface to Ensembl data, including gene annotation.

3. AnnotationDbi (R package) - offers a consistent interface to a range of annotation resources, such as gene ontology and KEGG pathways.

4. BioPython (Python package) - a collection of bioinformatics-related Python modules, including one for processing and editing GenBank entries, which may be used to get gene annotation information.

5. GORILLA (R package) - a R tool for high-throughput gene ontology (GO) analysis, which may be used to annotate genes with their GO concepts.

Many of these programs may annotate genes with disease-related information, such as disease-gene connections from the Online Mendelian Inheritance in Man (OMIM) database or GeneCards.