Fig 1.
This figure demonstrates the training of the imputation model using the reference data. The inputs required for SWAM are a set of reference genotypes with sample matched measured expression, and the multiple imputation models to be included. The list of multiple imputation models must also include a model derived from the reference data, which can be done via prediXcan. SWAM uses these models to impute tissue-specific expression levels from the reference genotypes. These imputed expression sets are then compared with the measured expression of the reference set. The weights are calculated based on the similarity between the measured and imputed expression and the covariance structure of tissues. For full details, see the Materials and Methods section.
Fig 2.
Simulation study comparing SWAM with naïve average, best tissue and single tissue methods.
We performed each simulation 10,000 times, with the following default settings: 10 total tissues (1 target, 4 relevant, 5 irrelevant), 100 SNPs (2 per tissue), 10% genetic heritability, 50% shared heritability between relevant tissues. In addition, the sample size of the target tissue was 100 individuals, and the remaining tissues had 200 individuals. This was done to emphasize the importance of integrating information from other tissues when the quality of the target tissue model is limited. Five methods–Single Tissue, UTMOST, Best Tissue, Naïve Average, and SWAM were compared. Panel (A) shows the effects of changing the shared heritability for the relevant tissues. We note that each tissue has 10 causal SNPS–for the relevant tissues, 5 of these causal SNPS is shared with the target tissue while the other 5 are independent of all simulated tissues. In panel (B), we varied the number of relevant tissues, from 0 to 10. Panel (C) shows the improvement when the total number of tissues is increased, with the number of irrelevant tissues fixed at 50% of the total. Panel (D) shows the performance of the approaches for different levels of genetic heritability. This simulation demonstrates the range of heritability that we would expect to see the most improvement. Panel (E) shows the effects of target tissue sample size. The x-axis pertains to the sample size of the target tissue only, and all other tissues were fixed at 200 individuals. Finally, panel (F) shows the performance of the methods at different p-value thresholds, using the default simulation settings.
Fig 3.
Empirical validation of SWAM using lymphoblastoid-cell line data from GEUVADIS consortium.
We used our LCL-targeted SWAM model to impute expression levels based on the genotypes of 344 European samples. We then calculated the concordance between imputed expression and measured LCL expression. We repeated this for all of the other methods mentioned here. (A) shows the performance of SWAM against the single-tissue models from 49 tissue-specific predictDB models derived from GTEx version 8. In (B), we derived various SWAM models using every combination of the following: 1) all GTEx v6 tissues, 2) all GTEx v7 tissues, 3) all GTEx v8 tissues, and 4) Depression Gene Network (DGN) single tissue whole blood model from predictDB. Here, we also included the UTMOST LCL model, naïve average and best tissue models, all derived from GTEx v6.
Fig 4.
TWAS on LDL trait targeting liver using SWAM, UTMOST and PrediXcan models.
TWAS was performed using metaXcan on the LDL trait from the Global Lipids Genetics Consortium (GLGC) GWA analysis. For a consistent comparison, the SWAM and UTMOST models were derived from GTEx version 6 tissues, and the prediXcan model used was GTEx v6 liver. The number of associations were: 66, 56 and 15 for SWAM, UTMOST and prediXcan respectively. P-values were capped at 10−20 in these plots.