Author: Defne Yanartas
Date: March 2023
Tripping SNPs is a web application tool to visualize SNPs on a map and track them over time. This README file explains how to use the tool for a set of plink files and annotation file.
Sections:
- Installing and set-up
- Data
- Programs (usage)
- Results
Install plink
wget https://zzz.bwh.harvard.edu/plink/dist/plink-1.07-x86_64.zip #download to the bin
unzip plink-1.07-x86_64.zip #after that, add the directory to the pathSet up the project directory and git repository. For the rest of the file I will not wrote each time that I commit changes but the user can decide when to commit and push their changes to the repository and remote.
mkdir Data
mkdir Programs
mkdir Results
gitinit
git remote add origin [email protected]:defneyanartas/Tripping_SNPs.git
git branch -M main
git push --set-upstream origin main
conda create --name tripping-r-env r-base It is important to have the directory and file names exactly as stated here in this document.
Fetch the plink files (fam, bim and bed) from:https://github.com/sarabehnamian/Origins-of-Ancient-Eurasian-Genomes/tree/main/steps/Step%20. Rename the base as Eurasian. Then proceed to use plink to recode the binary files to readable files. Output will be a map and ped file.
plink --bfile Eurasian --recode --out Eurasian --nowebGet the annotation file from : https://drive.google.com/drive/folders/1LQyaf6zNSWbklSlf7stQmwFL8N3RZ9NJ?usp=sharingLinks to an external site. and name it "Eurasian.anno" The files created in PLINK are too large and will not be tracked or pushed to github but the binary files and annotation file is provided in the "1.Data" directory so the files necessary for the program to run can be created by following the PLINK command I have written.
Scripts were written in R (version 4.2.1) in RStudio. The functions_file.R contains the commands for loading the libraries, reading the plink and annotation files, together with functions that clean the dataframe and calculate MAF. The application script contains a line of code that sources the functions file and the shiny application. They are very well annotated within the scripts. Below is a session information.
htmlwidgets_1.6.1 compiler_4.2.1 magrittr_2.0.3 fastmap_1.1.0 R6_2.5.1 cli_3.6.0
leaflet_2.1.1 htmltools_0.5.4 tools_4.2.1 rstudioapi_0.14 crosstalk_1.2.0 digest_0.6.31
rlang_1.0.6
We need to run the application in the conda environment that we have created. We also need to install the required packages.
conda activate tripping-r-env r-base
conda install -c conda-forge r-shiny
conda install -c conda-forge r-tidyverse
conda install -c conda-forge r-leaflet
conda install -c conda-forge r-dplyr
conda install -c conda-forge r-shinywidgets
conda install -c conda-forge r-leaflet.extrasYou can run the application by running the code below from your project directory. Notice that scripts must be placed under "2.Programs" directory and the data under "1.Data" directory.
Rscript Programs/Tripping_SNP_shiny.R && R -e "shiny::runApp('Programs/Tripping_SNP_shiny.R', launch.browser = TRUE)"
It might take some time to read data depending on the size. When a line like "http://127.0.0.1:4164/" appears, either there will be a popup window on your browser or if not, copy the address to your browser and this should bring up the GUI, then you can select your options and start visualizing.
The default SNP is depicted together with no populations picked when the application is started. Minor allele and the minor allele frequency (MAF) are also displayed underneath the map for the whole dataset of the chosen SNP.
The user can select a different SNP, certain populations or a time period.
When the user ticks the time path check box, a time tracking line is drawn showing the samples from the oldest to the newest. Also when the cursor is on a marker, information regarding the sample is depicted.
