PRED-LD

PRED-LD: A tool for GWAS summary statistics Imputation, using precalculated LD statistics

Web tool

Web tool available at : https://compgen.dib.uth.gr/PRED_LD/ or 195.251.108.198:3839/PRED_LD

Demo LD reference panel (Download before running PRED-LD)

Demo LD ref folder

LD resources

Population Data Overview

HapMap Populations

The following table lists the populations included in the HapMap project, along with their symbols:

Population	Symbol
Yoruba in Ibadan, Nigeria	YRI
Han Chinese in Beijing, China	CHB
Japanese in Tokyo, Japan	JPT
CEPH/Utah Collection (NIGMS Human Genetic Cell Repository)	CEU
Maasai in Kinyawa, Kenya	MKK
Luhya in Webuye, Kenya	LWK
Chinese in Metropolitan Denver, CO, USA	CHD
Gujarati Indians in Houston, TX, USA	GIH
Toscani in Italia	TSI
Mexican Ancestry in LA, CA, USA	MXL
African Ancestry in SW USA	ASW

Pheno Scanner and TOP-LD Populations

The following table outlines the population symbols as recognized by Pheno Scanner and TOP-LD:

Population	Symbol
Americans (Only in Pheno Scanner)	AMR
South Asians	SAS
East Asians	EAS
Europeans	EUR
Africans	AFR

Installation guide

PRED-LD is written in Python (ver. 3.8.2)

Clone or download PRED-LD from: https://github.com/pbagos/PRED-LD

git clone  https://github.com/pbagos/PRED-LD

After downloading the .zip folder of PRED-LD from GitHub, extract it to a working directory.
Το install the requirements, pip needs to be installed. Download the script for pip, from: https://bootstrap.pypa.io/get-pip.py.
Open a terminal/command prompt, cd to the folder containing the get-pip.py file and run:
```
python get-pip.py
```
To install the mentioned requirements with pip, open a terminal/command prompt and run:
```
pip install -r  requirements.txt
```

Arguments

PRED-LD accepts the following command-line arguments:

--file-path: The path to the input file containing SNP data. The file should be in tab-separated format (TSV,TXT)
--r2threshold: A float value specifying the R2 threshold for LD filtering
--pop: A string indicating the population code to use for LD calculations (EUR, EAS, SAS, AFR, AMR, YRI etc.), depending on the LD reference resource (--ref argument)
--maf: A float value indicating the minor allele frequency (MAF) threshold
--ref: A string indicating the LD Reference files (Pheno_Scanner, TOP_LD, Hap_Map, all_panels)
--imp_list: A filename (.txt) to define specific rsIDs to impute (each SNP has a new line, no header)

Usage

Your input data should be in a tab-separated text file (TXT format). Ensure the file contains the necessary SNP information and adheres to the specified format.

snp	chr	pos	beta	SE
rs5747010	22	16855618	-0.02	0.12
rs5746647	22	17057138	0.05	0.25
rs5747988	22	17073066	-0.17	0.22
rs5747999	22	17075353	-0.07	0.14
rs2096537	22	17094749	0.27	0.15
rs4428101	22	17118747	-0.22	0.13

To run PRED-LD, navigate to the directory containing the script and execute it with the required arguments. Make sure you have unzipped in the same working directory the ref folder. Demo LD ref folder (Download before running PRED-LD)

Examples

Here is an example command:

python pred_ld.py --file-path /path/to/your/data.txt --r2threshold 0.8 --pop EUR --maf 0.01 --ref TOP_LD

Example 1 (Simple Imputation)

python pred_ld.py --file-path PRED_LD_demo.txt --r2threshold 0.8 --pop EUR --maf 0.01 --ref TOP_LD

Example 2 (Use a list to impute specific rsIDs)

python pred_ld.py --file-path PRED_LD_demo.txt --r2threshold 0.8 --pop EUR --maf 0.01 --ref TOP_LD --imp_list missing_snps.txt

Example 3 (Use all panels to perform Imputation)

python pred_ld.py --file-path PRED_LD_demo.txt --r2threshold 0.8 --pop EUR --maf 0.01 --ref all_panels