### Install TwoSampleMR from Source

Source: https://github.com/mrcieu/twosamplemr/blob/master/README.md

Install the TwoSampleMR package directly from its GitHub repository. Ensure 'remotes' package is installed first. This method is useful for installing the latest development version.

```r
install.packages("remotes")
remotes::install_github("MRCIEU/TwoSampleMR")
```

--------------------------------

### Example: Split Exposure Data

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

An example demonstrating how to use the `split_exposure` function to split exposure data.

```r
exposure_list <- split_exposure(exposure_dat)
```

--------------------------------

### Run Steiger Filtering Example

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/sensitivity-pleiotropy-tests.md

Example demonstrating how to apply `steiger_filtering` to harmonized data after adding R-squared values.

```R
harmonized_dat <- add_rsq(harmonized_dat)
dat_filtered <- steiger_filtering(harmonized_dat)
```

--------------------------------

### Example: Split Outcome Data

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

An example showing the usage of the `split_outcome` function for splitting outcome data.

```r
outcome_list <- split_outcome(outcome_dat)
```

--------------------------------

### Run Rucker's Test Example

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/sensitivity-pleiotropy-tests.md

Example of how to run Rucker's test and print the results. The Q_prime_pval column indicates the significance of potential outliers.

```R
rucker_results <- mr_rucker(dat = harmonized_dat)

# Q_prime_pval suggests presence of outliers if significant
print(rucker_results)
```

--------------------------------

### Install TwoSampleMR from MRC IEU R-universe

Source: https://github.com/mrcieu/twosamplemr/blob/master/README.md

Use this command to install the latest version of the TwoSampleMR package from the MRC IEU R-universe repository. This is the recommended method for Windows and macOS users.

```r
install.packages("TwoSampleMR", repos = c("https://mrcieu.r-universe.dev", "https://cloud.r-project.org"))
```

--------------------------------

### Generate MR Scatter Plot Example

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/visualisation-functions.md

Example of how to create and display an MR scatter plot. Requires MR results and harmonized data as input.

```R
mr_res <- mr(dat = harmonized_dat)
scatter <- mr_scatter_plot(mr_res, harmonized_dat)
print(scatter$plot)
```

--------------------------------

### Example Usage of mr_ivw_fe

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/mr-methods.md

Shows a practical example of using the mr_ivw_fe function. Provide your exposure and outcome data, including their respective beta and standard error values.

```r
result <- mr_ivw_fe(
  b_exp = dat$beta.exposure,
  b_out = dat$beta.outcome,
  se_exp = dat$se.exposure,
  se_out = dat$se.outcome
)
```

--------------------------------

### Example Usage of mr_ivw

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/mr-methods.md

Demonstrates how to call the mr_ivw function with sample data. The input data should contain beta and standard error values for exposure and outcome.

```r
result <- mr_ivw(
  b_exp = dat$beta.exposure,
  b_out = dat$beta.outcome,
  se_exp = dat$se.exposure,
  se_out = dat$se.outcome
)
```

--------------------------------

### Exposure Data Structure Example

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/types.md

Displays the first few rows of an exposure data frame to illustrate the expected column structure and data types. This is useful for verifying data input.

```r
# Structure
head(exposure_dat)
#   SNP beta.exposure se.exposure effect_allele.exposure other_allele.exposure
# 1 rs12345678      0.05      0.01                      A                      G
# 2 rs23456789      0.07      0.011                     T                      C
# ...
#   eaf.exposure pval.exposure exposure id.exposure
# 1          0.3        1.2e-10 Exposure    expos_1
# 2          0.4        5.6e-08 Exposure    expos_1

# Attributes: none standard
```

--------------------------------

### Generate MR Forest Plot Example

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/visualisation-functions.md

Example of how to generate and display an MR forest plot using singlesnp and mr results. Requires prior MR analysis outputs.

```R
singlesnp_res <- mr_singlesnp(dat = harmonized_dat)
mr_res <- mr(dat = harmonized_dat)

forest <- mr_forest_plot(singlesnp_res, mr_res)
print(forest$plot)
```

--------------------------------

### Example: One-to-Many Forest Plot

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/visualisation-functions.md

Demonstrates how to generate and print a forest plot for one-to-many MR analysis. Requires harmonized data and MR results.

```r
# MR of one exposure against many outcomes
one_to_many_results <- mr(dat = harmonized_dat)

forest <- forest_plot_1_to_many(one_to_many_results, exponentiate = FALSE)
print(forest$plot)
```

--------------------------------

### Example: Subset MR Results

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

Demonstrates subsetting MR results to include only those from the 'Inverse variance weighted' method.

```r
ivw_results <- subset_on_method(mr_results, "Inverse variance weighted")
```

--------------------------------

### Run Steiger Directionality Test Example

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/sensitivity-pleiotropy-tests.md

Example of how to use `mr_steiger` after harmonizing data and adding R-squared values. It shows how to filter the results to include only forward directionality.

```R
# Must run add_rsq first
harmonized_dat <- add_rsq(harmonized_dat)

steiger_results <- mr_steiger(dat = harmonized_dat)

# Filter to forward direction only
forward_only <- subset(steiger_results, directionality == "Forward")
```

--------------------------------

### Display and Filter MR Methods

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/mr-methods.md

After obtaining the list of all MR methods with `mr_method_list()`, you can display specific columns like 'obj', 'name', and 'use_by_default'. You can also filter this list to get only the methods that are used by default in `mr()` analyses.

```r
all_methods <- mr_method_list()
print(all_methods[, c("obj", "name", "use_by_default")])

# Get only default methods
default_methods <- subset(all_methods, use_by_default)
```

--------------------------------

### Run MR Analyses with Specific Methods

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/mr-methods.md

Execute MR analyses using a custom list of methods by passing a character vector of method names to the `method_list` argument of the `mr` function. This example includes Wald Ratio, IVW, Egger Regression, and Weighted Median methods.

```r
mr_results <- mr(
  dat = harmonized_dat,
  parameters = default_parameters(),
  method_list = c("mr_wald_ratio", "mr_ivw", "mr_egger_regression", "mr_weighted_median")
)

# View main results
head(mr_results$mr)
```

--------------------------------

### Get Information on Available MR Methods

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/mr-methods.md

Retrieve a data frame containing details about all available MR methods using the `mr_method_list` function. This includes function names, display names, PubMed IDs, descriptions, and whether they are used by default or include heterogeneity testing.

```r
mr_method_list()
```

--------------------------------

### Format Data for MR Analysis (Outcome)

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/core-data-functions.md

This example demonstrates how to use `format_data` for outcome data, specifying custom column names for SNPs, effect sizes, and standard errors. Ensure your input data frame (`mydata`) contains these columns.

```r
dat <- format_data(
  mydata,
  type = "outcome",
  snp_col = "rsid",
  beta_col = "effect_size",
  se_col = "std_error"
)
```

--------------------------------

### Get Default Parameters for MR Methods

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Retrieves the default numeric parameters used by various MR methods. These parameters can be customized to control the behavior of MR analyses.

```r
params <- default_parameters()
# Parameters include:
# - num_wald_ratio_snps: Number of SNPs for Wald ratio
# - nboot: Number of bootstrap iterations for bootstrap methods
# - seed: Random seed for reproducibility
# And various other method-specific parameters
```

```r
params <- default_parameters()
params$nboot <- 1000  # Increase bootstrap iterations

mr_results <- mr(
  dat = harmonised_dat,
  parameters = params
)
```

--------------------------------

### Specify Local LD Reference Panel

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

The `bfile` parameter takes the prefix of your PLINK binary files (.bed, .bim, .fam). Ensure these files are in the specified directory.

```r
bfile = "reference_panel"
```

```r
clump_data(
  dat,
  bfile = "/path/to/reference_panel"
)
```

--------------------------------

### Configure Bootstrap Iterations for MR Methods

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Set the number of bootstrap iterations for MR methods by modifying the `nboot` parameter within the `parameters` list. The default is 1000.

```r
params <- default_parameters()
params$nboot <- 5000

mr_results <- mr_egger_regression_bootstrap(
  b_exp, b_out, se_exp, se_out,
  parameters = params
)
```

--------------------------------

### mr_method_list

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/mr-methods.md

Returns information about available MR methods including function names, descriptions, and properties. This is useful for understanding which methods are available and their default usage.

```APIDOC
## mr_method_list

### Description
Returns information about available MR methods including function names, descriptions, and properties.

### Method Signature
```r
mr_method_list()
```

### Return

data.frame with columns:
- `obj`: Function name (e.g., "mr_wald_ratio", "mr_ivw")
- `name`: Display name
- `PubmedID`: PMID for method reference
- `Description`: Method description
- `use_by_default`: Whether included in default `mr()` analysis
- `heterogeneity_test`: Whether method includes heterogeneity testing

### Example

```r
all_methods <- mr_method_list()
print(all_methods[, c("obj", "name", "use_by_default")])

# Get only default methods
default_methods <- subset(all_methods, use_by_default)
```
```

--------------------------------

### Customize and Use MR Parameters

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/mr-methods.md

Demonstrates how to obtain default parameters, modify specific values (e.g., `num_wald_ratio_snps`), and then use these customized parameters in the `mr()` function.

```r
params <- default_parameters()
# Customize parameters
params$num_wald_ratio_snps <- 1
mr_results <- mr(harmonized_dat, parameters = params)
```

--------------------------------

### Configure LD Clumping Window Size

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Set the `clump_kb` parameter to define the kilobase window size for LD clumping. Larger values capture more distant LD, with 10000 (10 Mb) being a common default for European ancestry.

```r
clump_data(dat, clump_kb = 10000)
```

--------------------------------

### Set OPENGWAS_JWT in .Renviron File

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Shows how to set the OPENGWAS_JWT environment variable by adding it to the user's .Renviron file. This ensures the token is available across R sessions.

```bash
# In ~/.Renviron
OPENGWAS_JWT=your_jwt_token_here
```

--------------------------------

### Configure P-value Threshold for Seed SNPs

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

The `clump_p1` parameter filters seed SNPs based on their p-value. Setting it to 5e-8, for example, ensures only genome-wide significant SNPs are used for clumping.

```r
clump_data(dat, clump_p1 = 5e-8)  # Only clump SNPs genome-wide significant
```

--------------------------------

### List Available Enrichment Methods

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

Returns a data frame describing the available enrichment analysis methods. No parameters are required.

```r
enrichment_method_list()
```

--------------------------------

### Robust Adjusted Profile Score (RAPS) MR

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Use `mr_raps` for the Robust Adjusted Profile Score (RAPS) MR method. This approach offers robust estimation, particularly when dealing with potential violations of standard MR assumptions. Ensure the `mr.raps` package is installed.

```r
mr_raps(
  b_exp,
  b_out,
  se_exp,
  se_out,
  parameters = default_parameters()
)
```

--------------------------------

### Specify PLINK Binary Path

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Use the `plink_bin` parameter to provide the explicit path to your PLINK executable. If set to NULL, the package attempts to auto-detect it.

```r
# Auto-detect
clump_data(dat, plink_bin = NULL)
```

```r
# Specify explicit path
clump_data(
  dat,
  plink_bin = "/usr/local/bin/plink"
)
```

--------------------------------

### Perform Bootstrap Rucker's Q-Q' Test

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/sensitivity-pleiotropy-tests.md

Conducts a bootstrap-based version of Rucker's Q-Q' test for heterogeneity and outlier detection. Uses the same parameters as mr_rucker().

```R
mr_rucker_bootstrap(
  dat,
  parameters = default_parameters()
)
```

--------------------------------

### Harmonize Data with Custom Strand Assignment per Outcome

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Provide a vector to `action` to specify different strand assignment strategies for multiple outcomes. For example, `action = c(2, 2, 3)` applies action 2 to the first two outcomes and action 3 to the third.

```r
harmonise_data(
  exposure_dat,
  outcome_dat,
  action = c(2, 2, 3)  # Three outcomes, third uses action=3
)
```

--------------------------------

### make_dat

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/core-data-functions.md

Creates a standardized dataset combining exposure and outcome data with specified columns.

```APIDOC
## make_dat

### Description
Creates a standardized dataset combining exposure and outcome data with specified columns.

### Method
(Not specified, likely a data processing function)

### Endpoint
(Not applicable, local function)

### Parameters
#### Request Body
- **exposure_dat** (data.frame) - Required - Formatted exposure data.
- **outcome_dat** (data.frame) - Required - Formatted outcome data.
- **outcome_snps** (logical) - Optional - Whether to use outcome SNPs for merging (TRUE) or exposure SNPs (FALSE). Defaults to FALSE.

### Return
data.frame with combined exposure and outcome data, merged by SNP.

### Request Example
```r
combined_dat <- make_dat(exposure_dat, outcome_dat)
```
```

--------------------------------

### GRIP MR Method

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Implements the Generalized Regression using Instrumental Polynomial (GRIP) method for MR. Similar parameter requirements to other MR methods.

```r
mr_grip(
  b_exp,
  b_out,
  se_exp,
  se_out,
  parameters = default_parameters()
)
```

--------------------------------

### Wrapper for Applying MR Methods

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

A flexible wrapper function to apply any specified MR method to harmonized data. Useful for easily switching between different MR techniques.

```r
results <- mr_wrapper(
  dat = harmonized_dat,
  method_name = "mr_ivw"
)
```

--------------------------------

### Add MR-GRIP Method to TwoSampleMR

Source: https://github.com/mrcieu/twosamplemr/blob/master/NEWS.md

Demonstrates how to add the new MR-GRIP method to the list of methods used by the `mr()` function. This method implements the MR-GRIP approach for Mendelian Randomization.

```R
mr(dat, method_list = "mr_grip")
```

```R
mr(dat, method_list = c(subset(mr_method_list(), use_by_default)$obj, "mr_grip"))
```

--------------------------------

### Format Data for MR Analysis (Exposure)

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/core-data-functions.md

This snippet shows the default usage of `format_data` for preparing exposure data. It includes all possible column mapping parameters for comprehensive data standardization.

```r
format_data(
  dat,
  type = "exposure",
  snps = NULL,
  header = TRUE,
  phenotype_col = "Phenotype",
  snp_col = "SNP",
  beta_col = "beta",
  se_col = "se",
  eaf_col = "eaf",
  effect_allele_col = "effect_allele",
  other_allele_col = "other_allele",
  pval_col = "pval",
  units_col = "units",
  ncase_col = "ncase",
  ncontrol_col = "ncontrol",
  samplesize_col = "samplesize",
  gene_col = "gene",
  id_col = "id",
  min_pval = 1e-200,
  log_pval = FALSE,
  chr_col = "chr",
  pos_col = "pos"
)
```

--------------------------------

### Simple Median MR

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/mr-methods.md

Implements the simple (unweighted) median MR method. This is a basic median-based approach.

```r
mr_simple_median(
  b_exp,
  b_out,
  se_exp,
  se_out,
  parameters = default_parameters()
)
```

--------------------------------

### Citation for Two Sample MR

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/overview.md

This is the citation for the Two Sample MR project, as published in the European Journal of Epidemiology.

```text
Hemani G, Bowden J, Davey Smith G (2018).
Evaluating the potential role of pleiotropy in Mendelian randomization studies.
European Journal of Epidemiology, 33(9), 807-818.
DOI: 10.1007/s10654-018-0403-y
```

--------------------------------

### Read Exposure Data with Log P-values

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Set `log_pval = TRUE` when your input p-values are already -log10 transformed. This avoids re-transformation.

```r
read_exposure_data(
  filename = "data_with_log_p.txt",
  log_pval = TRUE
)
```

--------------------------------

### enrichment_method_list

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

Returns a list of available enrichment analysis methods. This function does not take any parameters and returns a data frame describing the available methods.

```APIDOC
## enrichment_method_list

### Description
Returns list of available enrichment analysis methods.

### Parameters
#### Path Parameters
- None

#### Query Parameters
- None

#### Request Body
- None

### Method Signature
```r
enrichment_method_list()
```

### Parameters
- None

### Return
data.frame describing available enrichment methods.
```

--------------------------------

### Simple Fixed Effects Meta-Analysis MR Method

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Implements a simple fixed effects meta-analysis MR method. Requires exposure and outcome effect sizes and their standard errors.

```r
mr_meta_fixed_simple(
  b_exp,
  b_out,
  se_exp,
  se_out,
  parameters = default_parameters()
)
```

--------------------------------

### Format Data for One-to-Many Visualization

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/visualisation-functions.md

Formats MR results data into a structure suitable for one-to-many analysis visualization. Takes MR results as input.

```r
format_1_to_many(
  exposure_results
)
```

--------------------------------

### get_r_from_bsen

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

Calculates correlation coefficient from beta, standard error, and sample size.

```APIDOC
## get_r_from_bsen

### Description
Calculates correlation coefficient from beta, standard error, and sample size.

### Parameters
#### Path Parameters
- **beta** (numeric vector) - Required - Effect size estimates.
- **se** (numeric vector) - Required - Standard errors.
- **n** (numeric vector) - Required - Sample sizes.

### Return
numeric vector of correlation coefficients.

### Example
```r
r <- get_r_from_bsen(beta = 0.05, se = 0.01, n = 100000)
rsq <- r^2
```
```

--------------------------------

### Local Data Analysis for MR in R

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/overview.md

This pattern analyzes MR using local data files (exposure.txt, outcome.txt). It reads data, harmonizes it, and runs the MR analysis. Ensure the files are correctly formatted and located.

```r
exposure <- read_exposure_data(
  filename = "exposure.txt",
  sep = "\t",
  clump = TRUE
)

outcome <- read_outcome_data(
  filename = "outcome.txt",
  sep = "\t"
)

harmonised <- harmonise_data(exposure, outcome)
results <- mr(harmonised)
```

--------------------------------

### Set Proxy SNP Lookup Batch Size with proxy_splitsize

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Configure the batch size for proxy SNP lookups using the `proxy_splitsize` parameter. The default value for this parameter is 500.

```r
extract_outcome_data(
  snps = instruments$SNP,
  outcomes = "ieu-a-5",
  proxy_splitsize = 500
)
```

--------------------------------

### Align Proxy Alleles to Target Alleles

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Set `align_alleles = 1` to ensure that proxy SNP alleles are aligned to the target alleles. The default is 1.

```r
extract_outcome_data(
  snps = instruments$SNP,
  outcomes = "ieu-a-5",
  align_alleles = 1
)
```

--------------------------------

### Format Proteomic QTL Data

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/harmonisation-functions.md

Use this function to format proteomic (pQTL) data from a text file. Specify column names if they differ from defaults. The function returns a data.frame suitable for exposure or outcome data.

```r
format_proteomic_qtls(
  filename,
  type = "exposure",
  snp_col = "SNP",
  phenotype_col = "protein",
  beta_col = "beta",
  se_col = "se",
  eaf_col = "eaf",
  effect_allele_col = "effect_allele",
  other_allele_col = "other_allele",
  pval_col = "p"
)
```

```r
pqtl_dat <- format_proteomic_qtls("protein_qtl.txt", type = "exposure")
```

--------------------------------

### Set Random Seed for Reproducible MR-Presso Results

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Ensure reproducible results for stochastic methods like MR-Presso by setting the `seed` parameter. The default seed is 123.

```r
set.seed(123)
run_mr_presso(harmonised_dat, seed = 123)
```

--------------------------------

### mr_grip

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Applies the Generalized Regression using Instrumental Polynomial (GRIP) method for MR analysis.

```APIDOC
## mr_grip

### Description
Generalized Regression using Instrumental Polynomial (GRIP) method.

### Parameters
Same as `mr_penalised_weighted_median()`.

### Return
List with GRIP regression results.
```

--------------------------------

### Create Contingency Table for Allele Combinations

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

Generates a contingency table counting allele combinations from exposure and outcome data. Requires character vectors of alleles.

```r
contingency(alleles_exp, alleles_out)
```

```r
cont <- contingency(
  alleles_exp = dat$effect_allele.exposure,
  alleles_out = dat$effect_allele.outcome
)
```

--------------------------------

### Fetch and Display Available GWAS Studies

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/core-data-functions.md

Fetches all available GWAS studies and displays the first few rows, showing key metadata like ID, trait, number of SNPs, and sample size.

```r
all_studies <- available_outcomes()
head(all_studies[, c("id", "trait", "nsnp", "sample_size")])
```

--------------------------------

### Unweighted Regression MR Method

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Implements the unweighted regression MR method. Requires exposure and outcome effect sizes and their standard errors.

```r
mr_uwr(
  b_exp,
  b_out,
  se_exp,
  se_out,
  parameters = default_parameters()
)
```

--------------------------------

### mr_meta_fixed_simple

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Simple fixed effects meta-analysis MR method. A simplified approach to fixed-effects meta-analysis for MR.

```APIDOC
## mr_meta_fixed_simple

### Description
Simple fixed effects meta-analysis MR method.

### Parameters
Same as `mr_penalised_weighted_median()`.

### Return
List with elements: b, se, pval, nsnp.
```

--------------------------------

### Create Density Plot of Causal Estimates

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/visualisation-functions.md

Generates a density plot of causal estimates. Requires MR results and harmonized data. Customizable with line type and size.

```R
mr_density_plot(
  mr_results,
  dat,
  linetype = 1,
  linesize = 1
)
```

--------------------------------

### Maximum Likelihood MR Method

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/mr-methods.md

Utilize mr_two_sample_ml for maximum likelihood MR analysis. Parameters are the same as mr_ivw.

```r
mr_two_sample_ml(
  b_exp,
  b_out,
  se_exp,
  se_out,
  parameters = default_parameters()
)
```

--------------------------------

### Median of Estimates (MOE) MR Method

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Implements the Median of the Estimates (MOE) MR method. Requires exposure and outcome effect sizes and their standard errors.

```r
mr_moe(
  b_exp,
  b_out,
  se_exp,
  se_out,
  parameters = default_parameters()
)
```

```r
result <- mr_moe(
  b_exp = dat$beta.exposure,
  b_out = dat$beta.outcome,
  se_exp = dat$se.exposure,
  se_out = dat$se.outcome
)
```

--------------------------------

### Perform Heterogeneity Test (Q statistic)

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/sensitivity-pleiotropy-tests.md

Use `mr_heterogeneity` to assess the consistency of SNP estimates by performing a heterogeneity test. It returns the Q statistic, degrees of freedom, and p-value. This is useful for identifying potential violations of the MR assumptions.

```r
mr_heterogeneity(
  dat,
  parameters = default_parameters(),
  method_list = mr_ivw()
)
```

```r
het_results <- mr_heterogeneity(
  dat = harmonized_dat,
  method_list = c("mr_ivw", "mr_egger_regression")
)

# High p-value suggests homogeneity
print(het_results)
```

--------------------------------

### Configure LD Clumping r-squared Threshold

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Use the `clump_r2` parameter to set the LD r-squared threshold. A value of 0.001 is very strict, while 0.01 is moderate and commonly used.

```r
clump_data(dat, clump_r2 = 0.001)
```

--------------------------------

### mr_wrapper

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

A versatile wrapper function to apply any specified MR method to harmonized data.

```APIDOC
## mr_wrapper

### Description
Wrapper function to apply any MR method to data.

### Parameters
#### Parameters
- **dat** (data.frame) - Required - Harmonized data.
- **method_name** (character) - Required - Name of MR method to apply.
- **parameters** (list) - Optional - Method parameters. Defaults to `default_parameters()`.

### Return
data.frame with results from specified method.

### Example
```r
results <- mr_wrapper(
  dat = harmonized_dat,
  method_name = "mr_ivw"
)
```
```

--------------------------------

### Perform Rucker's Q-Q' Test

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/sensitivity-pleiotropy-tests.md

Applies Rucker's Q-Q' test to detect outlier SNPs and evaluate heterogeneity. Use the Q_prime_pval to identify potential outliers.

```R
mr_rucker(
  dat,
  parameters = default_parameters()
)
```

--------------------------------

### Format Metabolomic QTL Data

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/harmonisation-functions.md

Use this function to format metabolomic QTL data from a text file. Specify column names if they differ from defaults. The function returns a data.frame suitable for exposure or outcome data.

```r
format_metab_qtls(
  filename,
  type = "exposure",
  snp_col = "SNP",
  phenotype_col = "metabolite",
  beta_col = "beta",
  se_col = "se",
  eaf_col = "eaf",
  effect_allele_col = "effect_allele",
  other_allele_col = "other_allele",
  pval_col = "p"
)
```

```r
metab_dat <- format_metab_qtls("metabolite_qtl.txt", type = "exposure")
```

--------------------------------

### Retrieve Population Allele Frequencies

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

Fetches allele frequencies for specified SNPs from a reference panel for a given population. Requires a character vector of SNP rs IDs and a population code.

```r
get_population_allele_frequency(
  snps = instruments$SNP,
  pop = "EUR"
)
```

--------------------------------

### Set API Query Batch Size with splitsize

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Use the `splitsize` parameter to control the batch size for API queries when the number of SNPs or outcomes exceeds this value. The default is 10000.

```r
extract_outcome_data(
  snps = instruments$SNP,
  outcomes = "ieu-a-5",
  splitsize = 10000
)
```

--------------------------------

### format_1_to_many

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/visualisation-functions.md

Formats data specifically for one-to-many MR analysis visualization.

```APIDOC
## format_1_to_many

### Description
Formats data for one-to-many analysis visualization.

### Parameters
#### Path Parameters
- **exposure_results** (data.frame) - Required - MR results.

### Return
Formatted data.frame ready for visualization.
```

--------------------------------

### weighted_median_bootstrap

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Bootstrap version of weighted median. It calculates MR estimates using bootstrapping for the weighted median method.

```APIDOC
## weighted_median_bootstrap

### Description
Bootstrap version of weighted median. It calculates MR estimates using bootstrapping for the weighted median method.

### Parameters
Same as `weighted_median()`, plus parameters list.

### Return
List with bootstrap results.
```

--------------------------------

### Perform Basic Multivariable MR Analysis

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/multivariable-mr.md

Use `mv_basic` for a simplified multivariable MR analysis. This function requires pre-harmonized exposure data.

```r
mv_basic(
  mv_harmonized_data
)
```

```r
basic_results <- mv_basic(mv_harmonized)
```

--------------------------------

### get_population_allele_frequency

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

Retrieves population allele frequencies from reference panels.

```APIDOC
## get_population_allele_frequency

### Description
Retrieves population allele frequencies from reference panels.

### Parameters
#### Path Parameters
- **snps** (character vector) - Required - SNP rs IDs.
- **pop** (character) - Optional - Population code ("EUR", "AFR", "SAS", "EAS", etc.). Defaults to "EUR".

### Return
data.frame with SNP allele frequencies for specified population.

### Example
```r
euron_freqs <- get_population_allele_frequency(
  snps = instruments$SNP,
  pop = "EUR"
)
```
```

--------------------------------

### Penalised Weighted Median MR

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Implements the penalised weighted median MR method. Use this for robust estimation when some instruments may be invalid.

```r
mr_penalised_weighted_median(
  b_exp,
  b_out,
  se_exp,
  se_out,
  parameters = default_parameters()
)
```

```r
result <- mr_penalised_weighted_median(
  b_exp = dat$beta.exposure,
  b_out = dat$beta.outcome,
  se_exp = dat$se.exposure,
  se_out = dat$se.outcome
)
```

--------------------------------

### Subset MR Results by Method

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/statistical-utility-functions.md

Subsets MR results based on a specified method name or pattern. Requires MR results and a method pattern.

```r
subset_on_method(mr_results, method_pattern)
```

--------------------------------

### Create Standardized Dataset

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/core-data-functions.md

Combines exposure and outcome data into a standardized format, merging by SNP. The `outcome_snps` parameter determines whether to prioritize outcome or exposure SNPs for merging.

```r
make_dat(exposure_dat, outcome_dat, outcome_snps = FALSE)
```

--------------------------------

### Extract Outcome Data with Proxy SNP Search

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Set `proxies = TRUE` to enable searching for LD proxies for missing SNPs, increasing the number of available SNPs for analysis. The default is TRUE.

```r
extract_outcome_data(
  snps = instruments$SNP,
  outcomes = "ieu-a-5",
  proxies = TRUE
)
```

--------------------------------

### format_proteomic_qtls

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/harmonisation-functions.md

Formats proteomic (pQTL) data from text files into a data.frame suitable for exposure or outcome data in TwoSampleMR.

```APIDOC
## format_proteomic_qtls

### Description
Formats proteomic (pQTL) data from text files.

### Parameters
#### Path Parameters
- **filename** (character) - Required - Path to pQTL file.
#### Query Parameters
- **type** (character) - Optional - Default: "exposure" - "exposure" or "outcome".
- **snp_col** (character) - Optional - Default: "SNP" - Column name for SNP ID.
- **phenotype_col** (character) - Optional - Default: "protein" - Column name for protein/phenotype.
- **beta_col** (character) - Optional - Default: "beta" - Column name for effect size.
- **se_col** (character) - Optional - Default: "se" - Column name for standard error.
- **eaf_col** (character) - Optional - Default: "eaf" - Column name for effect allele frequency.
- **effect_allele_col** (character) - Optional - Default: "effect_allele" - Column name for effect allele.
- **other_allele_col** (character) - Optional - Default: "other_allele" - Column name for other allele.
- **pval_col** (character) - Optional - Default: "p" - Column name for p-value.

### Return
data.frame formatted as exposure or outcome data.

### Example
```r
pqtl_dat <- format_proteomic_qtls("protein_qtl.txt", type = "exposure")
```
```

--------------------------------

### Weighted Median Bootstrap

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/additional-methods.md

Performs a bootstrap version of the weighted median calculation. Use for robust estimation when dealing with potential outliers or wanting to assess variability.

```r
weighted_median_bootstrap(
  b_exp,
  b_out,
  se_exp,
  se_out,
  parameters = default_parameters()
)
```

--------------------------------

### Run MR-PRESSO for Outlier Detection

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/sensitivity-pleiotropy-tests.md

Executes the MR-PRESSO method to identify and correct potential outlier SNPs. Configurable with number of distributions, significance threshold, and random seed.

```r
run_mr_presso(
  dat = harmonized_dat,
  NbDistribution = 10000,
  SignifThreshold = 0.05
)
```

```r
print(presso_results$outliers)
```

--------------------------------

### Set Minimum P-value Threshold

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Use `min_pval` to set the minimum allowed p-value threshold, preventing numerical underflow in calculations. The default is 1e-200.

```r
read_exposure_data(filename, min_pval = 1e-200)
```

--------------------------------

### Perform Jackknife Rucker's Q-Q' Test

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/api-reference/sensitivity-pleiotropy-tests.md

Executes a jackknife leave-one-out version of Rucker's test to assess SNP contributions to heterogeneity. It shares the same parameters as mr_rucker().

```R
mr_rucker_jackknife(
  dat,
  parameters = default_parameters()
)
```

--------------------------------

### Set OPENGWAS_JWT Environment Variable in R

Source: https://github.com/mrcieu/twosamplemr/blob/master/_autodocs/configuration.md

Demonstrates how to set the OPENGWAS_JWT environment variable within an R session for authenticating with the OpenGWAS API. This is useful for functions that retrieve data from protected endpoints.

```r
Sys.setenv(OPENGWAS_JWT = "your_jwt_token_here")
```