The following alignment files were extracted from Rye et al. and Micsinai et al. in order to generate the peak calling benchmarks. They are used as examples on how to operate the SigSeeker webserver.
In general it is possible to process any ChIP-Seq, Histone-Seq, Methyl-Seq, DNase-Seq or ATAC-Seq aligned file in the .bam format with SigSeeker. Examples for epigenetic alignment tools are BWA, Bowtie and Bowtie.
The ensemble is setup in accordance with the experiment parameters here. While it is possible to characterize the ensemble in detail based on the algorithmic parameters it is recommended for beginners to leave the default parameters and just assign the total number of samples and control files to be processed. The example presented here will analyze one of the benchmarking datasets provided by Rye et al., in particular we will analyze one replicate compared to one control for the SRF dataset. Consequently we need to specify that we want to use 1 sample and 1 control in the form and the organism will have to be switched from mouse to human (hg18 in this case).
For the purpose of this example we will name the project "SRFBenchmark", the sample is "SRF" and the since we are using the control file for SRF we need to link it to the sample data by specify it as "CTR". The alignment data to be submitted is the Gm12878 human data set for SRF, in particular sample 4 and control 10.
After finishing the upload, which may take a long time (especially for large complex experiments), the user can initiate the analysis and will be presented with a link to the result screen. This result screen will not be populated before the analysis has completed. We are currently implementing an email notification system so that the user will not have to refresh the results screen.
The results for this particular analysis can be found here. The results for the ensemble based analysis can be found on the "Comparison" tab, which will provide a breakdown of various tool combinations as well as the union specific peak calling results described in the Bioinformatics paper.