Visualizing local estimates¶
This page describes script to visualize local SNP-heritability estimates and local genetic covariance estimates.
Local SNP-heritability estimates¶
We provide script (misc/local_hsqg_manhattan.py
) that makes Manhattan-style
plot for visualizing local SNP-heritability estimates. The script can be
executed as follows.
python misc/local_hsqg_manhattan.py \ --local-hsqg-est <local SNP-heritability output> \ --out <output file name e.g. trait1_local_hsqg.pdf> \ --trait-name TRAIT1
There following is an example figure generated by the visualization tool. Here, colored bars represent loci that have significant local SNP-heritability.
Local genetic covariance estimates¶
We provide script (misc/local_rhog_manhattan.py
) that makes Manhattan-style
plot for visualizing local genetic covariance estimates. The script can be
executed as follows.
python misc/local_rhog_manhattan.py \ --local-rhog-est <local genetic covariance estimates output> \ --local-hsqg-est <local SNP-heritability output trait 1> <local SNP-heritability output trait 2> \ --out <output file name e.g. trait1_trait2_local_rhog.pdf> \ --trait-names TRAIT1 TRAIT2
Similarly, colored bars represent loci that have significant local genetic covariance.