Enlight FAQ

The eQTL data sets come from two sources. The first one is the eQTL browser of Pritchard Lab at University of Chicago, the other one is the GTEx project. Both of them underwent format conversion and genome build liftover. The GTEx eQTL data is split into different subsets based on analysis IDs. Analysis IDs usually represent different tissues. However, sometimes they merely represent different measurements for expression. The name of a data set can tell you the tissue and analysis assay (if more than one are assays are used). Tissue-specific information is unavailable for data from University of Chicago as no such meta-data is provided

All ENCODE data sets come from UCSC genome browser with necessary genome build liftover. Wherever there are replicates, only the first one is selected. All data sets are in BED format. The score column is extracted for plotting and annotation.

The interaction matrix data comes from the data published in the original HiC method paper: Liberman-Aiden et al, Science, 2009. The resolution is 100 Kbp for intrachromosomal interaction and 1 Mbp for interchromosomal interaction. The ratio of Observed/Expected is reported for each region in the heatmap plot. The original genome build is hg18, UCSC liftover tool was applied to convert it to hg19.

If you want to plot over multiple regions, just open the dialogue for multiple region specification on the submission page. If you want to try different combinations of, say, ENCODE data tracks, you can open another tab, change your settings and submit new requests.

The query file uploaded by user will first be annotated by ANNOVAR. If any record is not accepted by ANNOVAR, it will go into a file ending with '.invalid_input'. To understand what is the standard ANNOVAR format, please go here.

The score is potentially different for each study. Mostly it is a -log10 P-value, but sometimes it can be a posterior probability or log10 Bayes Factor. It really depends on what study is reported for that eQTL. Enlight does not differentiate among different studies used by UChicago eQTL browser. Please check out the text annotation and look at here to figure out which study is reported and what the score means.

Under such circumstances, the highest score (smallest p value) will be plotted.

Most likely it is because there is no SNP in the region you specified. If you believe there should be more figures generated, please email me (yunfeigu@usc.edu) detailed information about your submission (input file, region specification, other settings).