DiscoDivas

Disambiguation

This page is for "a Genetic Distance-assisted PRS Combination Pipeline for Diverse Genetic Ancestries, an R script for calculating PRS for diverse populations, especially admixed populations.

Required R packages

• data.table
• dplyr
• stringr
• rio
• optparse

Command line to run DiscoDivas.R

Here is an example command line to run DiscoDivas.R

Rscript DiscoDivas.R -m file/med.g1000G.4pop.tsv \
-p ukbb.pca.txt.gz \
--prs.list afr.prs.tsv.gz,${wd}/eas.prs.tsv.gz,eur.prs.tsv.gz,sas.prs.tsv.gz \
-s IID,PRS \
-A 1,0.9,1,1 \
--regress.PCA T \
-o combine.linear.prs

• -m ukbb.pca.med.1000G.tsv

  med.g1000G.4pop.tsvcan be found in the files. Please notice that the order of the list after -A and --prs.list should has the same the order as the order of cohorts/ populations in this file. User should replace the file by other file that contains the median value of top PC of the validation cohorts actually being used in the analysis.

  The format should be:

  • must have a header; does't contain row name or row index;
  • the first column is the name of cohorts or populations; the seconds column and other columns after that are PCs

POP	PC1	PC2	PC3	PC4	PC5	PC6	PC7	PC8	PC9	PC10
AFR	-10358.4	837.239	193.871	-103.469	195.704	64.7022	-400.128	-49.6729	0.0890048	106.919
EAS	2707.46	6490.42	-166.124	-1408.4	134.848	-0.94705	-193.3735	-362.328	8.36484	598.289
EUR	1280.17	-5075.18	-794.419	-1903.47	172.565	224.566	-208.995	-321.358	-5.28843	113.007
SAS	1276.87	-1124.26	4017.77	1557.19	200.125	25.1516	-195.331	-349.297	-417.462	45.8134

• -p ukbb.pca.txt.gz

  ukbb.pca.txt.gz should be replaced by user's file contains the PC of the testing individual. It can be plain text file or .gz file. Please notice that in all the list, items should be separated by , and no space should be inserted.

  The format is:

  • must have header; does't contain row name or row index;
  • the first column is the individual ID; the seconds column and other columns after that are PCs

• --prs.list afr.prs.tsv.gz,${wd}/eas.prs.tsv.gz,eur.prs.tsv.gz,sas.prs.tsv.gz

  The list of PRS files of the testing cohort that are based on the weight fine-tuned in each of the validation cohorts. They can be plain text file or .gz file. The format is:

  • must have header; does't contain row name or row index;
  • must have a column of individual ID and a column of PRS
  • The individual ID in the PRS files should be the same as the individual ID in the PCA file

• -s IID,PRS The column names of individual ID and PRS in the PRS files. If the column names are the different in different files, please list them in the same order of prs.list : ID1,PRS1,ID2,PPRS2,ID3,PRS3,ID4,PRS4

• -A 1,0.9,1,1

  The list of shrinkage parameter for each PRS based on their quality. The default is a list of 1

• --regress.PCA T

  Choose whether combining PRS after correcting the ancestry information or not. Default and recommanded value is TRUE/T

• -o

  The prefix of output file. The format of output file is .tsv.gz

Supporting files

Please visit files for the suppporting files