Introduction🎯

BioCycleMR is an R package crafted with the objective of enhancing Mendelian Randomization (MR) analysis in the field of biomedical research. Designed to integrate diverse exposure data types, the tool is an amalgamation of traditional medicinal insights and the dynamism of modern genomics.🌱

Installation🎉

Installing the Package

install.packages("devtools")
devtools::install_github("DaXuanGarden/BioCycleMR")

Load and Commence

library(BioCycleMR)
results <- runBioCycleMR(your_exposure_data, your_outcome_data)
print(results)

Detailed Function Descriptions📕

get_finn

Usage:

To process FINN data from the "finn" directory and save the results to "finn_r" using 64cores:

library(BioCycleMR)
get_finn(finn_dir = "finn", save_dir = "finn_r", cores = 64)

Parameters:

1. finn_dir (default: "finn"):
   • Description: Specifies the directory where the raw FINN data files are located.
   • Type: String
2. save_dir (default: "finn_r"):
   • Description: Directory where the processed results should be saved.
   • Type: String
3. cores (default: 64):
   • Description: Number of cores to be used for parallel processing.
   • Type: Integer

Immune cells

1. Retrieve Data for 731 Immune Cells

   Obtain data for 731 immune cells. Choose to use preprocessed data or customized parameters (we have used the built-in default parameter p1 = 1e-05). The network problem can be solved well using this function, where we use the parameter count_try_max to adjust the maximum number of network attempts

   immc_data_preprocessed <- get_immc(use_preprocessed = TRUE)
   immc_data_custom <- get_immc(use_preprocessed = FALSE, 
                                p1 = 1e-05, 
                                p2 = 5e-08, 
                                r2 = 0.001, 
                                kb = 10000, 
                                mc_cores = 10)

   In order to avoid the problem of poor network of some users, we have been prepared to mess with the exposure data of the immunophenotypes with p=1e-5 and p=5e-6

   data("immune_cell_raw1e5")
   data("immune_cell_raw5e6")

2. Preprocess FinnGen R9 Data

   We recommend that you preprocess FinnGen R9 Data into a format with the suffix .rda (You will need to place the files downloaded from the FinnGen R9 Data in the specified directory).

   get_finn(finn_dir = "finn", save_dir = "finn_r", cores = 30)

3. Retrieve GWAS Datasets Using Keyword or ID

   Identify potential genetic instruments for MR analysis using keywords or IDs.

   get_gwas_id("Myocardial infarction")
   get_gwas_id("finn-b-N14_ENDOMETRIOSIS")

4. Convert Local VCF File for Two-Sample MR Analysis

   Convert VCF files for MR studies.

   get_local("ieu-a-2.vcf.gz", "exposure")

5. Calculate F-values and MAF

   Estimate F-statistics and minor allele frequency for MR studies.

   data("dx_immu_cell_raw_df")
   result_list <- get_f_maf(dat_object = immu_cell_raw,F_value = 10,maf_threshold = 0.01,)
   immu_cell_f = result_list[[1]],
   immu_cell_f_select = result_list[[2]],
   SNP_stats_f = result_list[[3]]

6. Complex Operations with get_tsmr

   The code below will create folders and output the results of MR, heterogeneity, horizontal pleiotropy, and PRESSO level pleiotropy tests

   get_tsmr(immu_cell_f_select, finn_r_dir="finn_r", cores = 64)

7. Compute Effect Size Estimates from .rda Files

   Derive effect size metrics from .rda files.

   get_effect("~/path_to_directory", immune_ref_data)

Future Developments🐾

BioCycleMR is designed with adaptability and scalability in mind. While it stands as a reflection of the current knowledge and skills of its creators, they envisage it to evolve, incorporating advancements in biomedical research and feedback from the scientific community.

Community Engagement and Feedback👏🏻

Open communication channels and collaborations are the lifeblood of BioCycleMR. The creators earnestly invite the community to pitch in, share insights, suggest enhancements, or even critique -- every interaction is a step towards refinement.

Creators😊

• Xuanyu Wang: A 2021 undergraduate student from the College of Traditional Chinese Medicine, Tianjin University of Traditional Chinese Medicine. Research Interests in Traditional Chinese Medicine with a special focus on Cardiology and Cardiovascular diseases.

• Yangyang Zhang: A 2021 undergraduate student from Shanghai Medical College, Fudan University. Research Interests include Obstetrics-Gynecology and reproductive medicine.

Contact✍🏻

For discussions, feedback, or potential collaborations:

📧 Xuanyu Wang(DaXuanGarden) 📧 Yangyang Zhang (pigudogzyy)