If you submit a correctly-formatted file using the form above, your job will be entered into our queue for running on our server in the Roslin Institute, University of Edinburgh. Very large jobs (those with more than 1000 SNPs, or more than 400 SNPs mapping to FANTOM5 TSS) may take a long time and these will be pushed down the queue during busy periods, and may be cancelled if they are taking too long. Please contact us () if you have a very large job, or download our code below and run it on your own server.
Upload a list of SNPs (or genomic locations) in the format described below. The SNPs should share some common feature, such as putative association with a given phenotype at a permissive p-value threshold (eg. 5e-6), such that it might reasonably be expected that some of the SNPs in the entry set will share an expression profile across the FANTOM5 expression atlas.
A standard tab- or space-delimited bed formatted file is sufficient for upload. We use a slightly modified bed format by adding a SNP ID to the 4th column, but this is optional. An example is shown here:
chr start end [optional_snp_id]
chr5 40622948 40622949 rs4957326
chr6 31263050 31263051 rs2853926
chr6 31274379 31274380 rs9264942
chr7 153487943 153487944 rs2098112
chr9 117538333 117538334 rs4574921
chr9 117547771 117547772 rs10114470
Coordinates must be hg19 - use LiftOver if neeeded.
The background file is a bed formatted file containing the genomic positions of all loci that your study could possibly have identified.
By default the locations in your uploaded chosen locations file will be mapped onto the FANTOM5 network and compared against randomly permuted genomic locations. This will use our circular permutations method (see manuscript) to randomly select locations on the genome in the same spactial distribution as your input SNPs.
In many studies the true background of variants is not the entire genome. For example, in a SNP chip GWAS study, the true background is the total SNP content of the genotyping chip used, because only these variants could possibly have been detected by your study. In this case it is more reliable to specify the true background by uploading a file containing all of the genomic positions of every SNP on the SNP chip used.
This specifies the number of randomly-generated circular permutations against which your data are compared. Default is 100, 1000 takes longer.
This allows you to keep track of your analyses more easily.
After you submit the form you will be redirected to a link on this site. Your results will appear there - usually within a few hours but for larger analyses (either bigger input files, or more connected groups of genomic variants) you might have to wait for a few days. Please contact us ( ) if you are waiting for longer than that and we'll do what we can to help.
Click here to see an example of the results format.
For a full description of the method, see our manuscript: Baillie JK et al. Shared Activity Patterns Arising at Genetic Susceptibility Loci Reveal Underlying Genomic and Cellular Architecture of Human Disease. PLOS Computational Biology 14, no. 3 (March 1, 2018): e1005934. PMC5849332.
We are very grateful to recieve funding from the following sources: Wellcome Trust, BBSRC, Intensive Care Society, MRC, NIH.