SUMMARY STATISTICS FROM GENOME-WIDE ASSOCATION ANALYSIS OF LDL
SUBFRACTIONS AND THEIR RESPONSE TO STATIN TREATMENT
December 6, 2016

This webpage contains summary data accompanying the following
publication:

Shim H, Chasman D I, Smith J D, Mora S, Ridker P M, Nickerson D A,
Krauss R M and Stephens M (2015). A Multivariate genome-wide
association analysis of 10 LDL subfractions, and their response to
statin treatment, in 1868 Caucasians. PLoS ONE 10(4): e0120758.
http://doi.org/10.1371/journal.pone.0120758

Code implementing the data processing steps and analysis procedures
can be found here:

https://github.com/heejungshim/prepare-summary-data-from-subfraction-analysis

All data contained here are licensed under a Creative Commons
Attribution 4.0 International license (CC-BY-4.0):

https://creativecommons.org/licenses/by/4.0

To use these data for research, please cite the above paper published 
in PLoS ONE.

DOWNLOADS
=========

1. README.txt: this file.
2. phenotypes.tar.gz: normalized averages and differences of LDL 
   subfractions. 
3. genotypes.tar.gz: genotype data for selected SNPs.
4. summary_stats_avg.tar.gz: summary statistics from genome-wide
   association analysis of averaged phenotype values.
5. summary_stats_diff.tar.gz: summary statistics from genome-wide
   association analysis of differential phenotype values.

Each summary statistics file contains effect estimates (regression
coefficients; see *.beta columns) and standard errors (see *.se
columns) for all included phenotypes. Phenotypes are taken as the
average and difference of the following blood lipid measurements:
LDL-C (L), LDL peak particle diameter (Ld), IDL1 (i1), IDL2 (i2), IDL3
(i3), LDL1 (l1), LDL2a (l2a), LDL2b (l2b), LDL3a (l3a), LDL3b (l3b),
LDL4a (l4a), LDL4b (l4b), triglycerides (Tg), HDL-C (H), total
cholesterol (T) and ApoB levels (A). Refer to the PLoS ONE paper for
more information.

The phenotype data are stored in two files: average.txt contains the
normalized average values of the LDL subfractions difference.txt
contains the normalized differences in the LDL subfractions. Each line
(row) of the file corresponds to a sample, and each column gives a
blood lipid measurement: LDL-C (L), LDL peak particle diameter (Ld),
IDL1 (i1), IDL2 (i2), IDL3 (i3), LDL1 (l1), LDL2a (l2a), LDL2b (l2b),
LDL3a (l3a), LDL3b (l3b), LDL4a (l4a), LDL4b (l4b), triglycerides
(Tg), HDL-C (H), total cholesterol (T) and ApoB levels (A). Note that
the normalization is implemented as a simple quantile transform. See
the PLoS ONE paper for more details about the phenotype data.

File genotypes.tar.gz contains genotype data for several SNPs that
were highlighted in the paper. 

*Give more details about the genotype data here.*

CONTACT INFO
============

For questions and feedback, please contact:

Matthew Stephens
Department of Human Genetics
University of Chicago
Cummings Life Science Center room 422
920 E 58th St
Chicago, IL 60637
mstephens@uchicago.edu

CONTRIBUTORS
============

Heejung Shim 
Daniel I. Chasman 
Joshua D. Smith
Samia Mora
Paul M. Ridker 
Deborah A. Nickerson 
Ronald M. Krauss 
Matthew Stephens