“Predicting cell proportion of 34 circulating immune cell types in human blood from bulk whole blood RNA-seq data”

How does it work?

DeconCell is an R package containing models for predicting the proportions of circulating immune cell subpopulations using bulk gene expression data from whole blood. Models were built using an elastic net and training in 95 healthy dutch volunteers from the 500FG cohort with FACS quantification of 73 circulating cell subpopulations as described in our previous publication. For aditional details on methods and results please go our manuscript (link will be updated soon).

This ready to use web application implements the DeconCell package. Upload a .txt or .csv file (maximum 10mb) and choose the corresponding parameters.


GitHub Repository Get access to the GitHub repository.

R package Do you want to analyze (larger) datasets on your own machine? The R package is also available for download.

Example data Example data is available to show how your data should look like.


Please have a look at our manuscript (link will be updated soon) for more information about the methods used.



Expression quantitative trait loci (eQTL) studies have been commonly used to interpret the function of disease-associated genetic risk factors. To date most eQTL analyses have been conducted in bulk tissues such as whole blood and tissue biopsies, which is likely to mask the cell type context of these regulatory effects. Although this context can be investigated by generating transcriptional profiles from purified cell subpopulations these methods are labor intensive and expensive, limiting the scale of this approach. Here we introduce a statistical framework, which we termed Decon2, for estimating cell proportions using expression profiles from bulk tissue samples (Decon-cell) and consecutive deconvolution of cell type eQTLs (Decon-eQTL). The estimated cell proportions as determined by Decon-cell are in agreement with experimental measurements across cohorts (R>=0.77). Using Decon-cell we were able to predict the proprotions of 34 circulating cell types for 3,194 samples from a population based cohort. Next, 15,616 whole blood eQTLs were identified and subsequently re-distributed to CT specific effects with Decon-eQTL using the predicted cell proportions from Decon-cell. Deconvoluted eQTLs show excelent allelic directional concordance with that of previously published eQTL studies which used either purified cell subpopulations or single cell RNA-Seq (>= 96%). Our method provides an opportunity to re-distribute eQTLs from bulk tissue into the different cell subpopulations, which is helpful to pinpoint the most relevant cell type for complex diseases. Decon2 is available as an R package and Java application and a web tool.