:mount_fuji: Fuji plot :mount_fuji:
a circos representation of multiple GWAS results.
Overview
Fuji plot (a circos representation of multiple GWAS results) was developed to efficiently summarize significant SNPs across dozens of traits (Kanai, M. et al., Nat. Genet. 2018). Technically, it is a wrapper script for Circos, generating input data and configuration files from GWAS results.
Fuji plot is named after :mount_fuji: Mt. Fuji in Japan :mount_fuji:, whose ridge and skirts greatly resemble our figure!
Requirements
- R with dplyr and stringr packages.
- To install these packages, type
install.packages("tidyverse")in R console.
- To install these packages, type
- Circos
- Please follow their instructions to install Circos.
- For the current version (v0.69-6), you need to apply our patch to avoid an infinite loop in text arrangements. Alternatively, you can download and use our patched version. I will report this issue soon.
Usage
Please add circos to your PATH or set CIRCOS_PATH in fujiplot.R appropriately.
Rscript fujiplot.R input_example/input.txt input_example/traitlist.txt output_exampleInputs
Input files used in Kanai, M. et al. (2018) are provided under ./input_example.
Significant trait-associated loci (input.txt)
This file provides a concatenated list of all significant trait-associated loci (for each trait). It also defines a unique locus across multiple traits to highlight pleiotropic loci. The required fields are as follows:
LOCUS_ID: Unique ID for each locusCATEGORY: Trait categoryTRAIT: Trait nameCHR: ChromosomeBP: Position (hg19)MARKER: Marker IDGENE: Gene symbol
Trait list (traitlist.txt)
This file provides a list of traits and their categories. It defines an order of traits and a color of each category in a figure. The required fields are as follows:
CATEGORY: Trait categoryTRAIT: Trait nameCOLOR: Category color
Outputs
All output files will be created under ./output.
Circos outputs (circos.{png,svg})
This is the raw output from Circos. Unfortunately, we could not figure out the appropriate way to make our highlights (radial lines) over scatter plots. To this end, we edited the svg file using Adobe Illustrator and produced the final figure.
Optional: a bar plot (barplot.pdf)
This corresponds to Fig. 1a of Kanai, M. et al. (2018). Again, you need to do some Illustrator edits to make it pretty. You can suppress this output by setting OUTPUT_BARPLOT = FALSE in fujiplot.R.
Citation
When using Fuji plot, please cite the following paper.
- Kanai, M., et al. Genetic analysis of quantitative traits in the Japanese population links cell types to complex human diseases. Nat. Genet. (2018) doi:10.1038/s41588-018-0047-6
Contact
Masahiro Kanai (mkanai@g.harvard.edu)