Fine-mapping transcriptome-wide association studies

The repository contains the main results for the fine-mapping of transcriptome-wide association studies (TWAS) reported in

Probabilistic fine-mapping of transcriptome-wide association studies Nicholas Mancuso, Malika K. Freund, Ruth Johnson, Huwenbo Shi, Gleb Kichaev, Alexander Gusev, Bogdan Pasaniuc. bioRxiv 2018.

Results

LIPIDS.FOCUS.XX.tsv contains the gene-based fine-mapping analysis over TWAS associations using prior variance of XX = {40, 80}.

Column	Description
Trait	The lipids trait investigated
Expression.Reference	Short description of the gene expression reference panel
Gene	Canonical gene name
Chr	Chromosome
Tx.Start	Transcription start site
Tx.End	Transcription end site
Region	Identifier for a risk region
TWAS.Z	Z-score for TWAS test of association
TWAS.P	P-value for TWAS test
Resid.Z	Residual Z-score after accounting for average signal at region
PIP	Marginal posterior probability for explaining observed association signal at Region
In.Cred.Set	Boolean indicating whether the gene-model is in the 90% credible gene set

LIPIDS.FUSION.tsv contains the lipids (HDL, LDL, TG, total chol) TWAS association statistics.

Column	Description
Trait	The lipids trait investigated
Expression.Reference	Short description of the gene expression reference panel
Gene	Canonical gene name
Chr	Chromosome
Tx.Start	Transcription start site
Tx.End	Transcription end site
Expression.h2g	SNP-hertiability explained by cis-SNPs in reference panel
Best.GWAS.SNP	SNP ID of the best GWAS SNP in the prediction model (i.e. smallest p-value)
Best.GWAS.SNP.Z.score	Z-score of the best GWAS SNP
Top.eQTL.SNP	SNP ID of the best cis-eQTL SNP in the reference panel
Top.eQTL.R2	Variance-explained in measured gene expression by best cis-eQTL SNP
Top.eQTL.Z	Z-score of the best cis-eQTL SNP
Top.eQTL.GWAS.Z	GWAS Z-score of the best cis-eQTL SNP
N.SNPs	Number of SNPs in the model
Non.zero.weights	Number of SNPs with non-zero weights in the model
Best.fit.model	Model with the best prediction accuracy
Model.5CV.R2	Adjusted estimate of variance-explained in measured gene expression using 5-fold CV
Model.5CV.P	P-value of cross-validation R2 estimate
TWAS.Z	Z-score for TWAS test of association
TWAS.P	P-value for TWAS test

Software and support

For performing fine-mapping of TWAS associations please see FOCUS.

For performing TWAS using summary-data, computing expression weights using custom expression panels, and additional post-TWAS analyses please see FUSION.

If you have any questions or comments please contact [email protected]