STEP 1

Select the alignment mode

help
Which mode should I choose?
warning
Please select an alignment mode.
Sample data is available in both modes.
One-vs-one mode demonstration
The default one-vs-one mode is suitable if you have medium-sized query regions, usually marked by epigenomic modifications, (e.g. ChIP-Seq peaks, hundreds to thousands of base pairs long), and much longer target regions. It will match the query region(s) to a sub-region of the target region(s) (one-against-one alignment, multiple paired inputs also supported). This mode allows defining target regions using homologous information.
Many-vs-many mode demonstration
The many-vs-many mode attempts to align every input query region to all target regions to find the best hit (the regions do not need to be paired). Therefore, it would be useful if you have query and target regions of similar lengths, and would like to find one best match for the query region among a set of target regions. This mode could also be handy if you'd like to use promoter regions as input data, which you can define with gene names.

STEP 2

Upload data

Query regions

Genome assembly:
Peak regions:
or upload your own:
warning Please either select a provided paired public dataset, or upload your own peak file (BED6 format).
Input regions:
BED6 fields:
chrom chromStart chromEnd name score strand
Example:
chr12 8207583 8209349 Example 0 +
or upload a file:
warning Please either provide the input regions (BED6 or gene names) in the text box, or upload a text file.
Please specify promoter region ranges if you use gene names for input: bps upstream from TSSs, bps downstream from TSSs.
warning Please use a non-negative integer.

Target regions

Genome assembly:
warning Please do not use the same reference when using homologous regions to specify your target in one-vs-one mode.
Peak regions:
or upload your own:
warning Please either select a provided paired public dataset, or upload your own peak file (BED6 format).
Input:
Please select the alignment mode to see specific search types.
Result for "{{postedClusterText}}"
  • {{ cluster.id }} open_in_browser
({{ clusterMessage }})
  • {{ species }} genes
    • {{ gene.description }}
warning Please provide a cluster name. You may use the text box to search for a cluster including the gene of interest. Please use the open_in_browser button to select the cluster you want.
bps upstream, bps downstream.
warning Please use non-negative integers.
BED6 fields:
chrom chromStart chromEnd name score strand
Example:
chr12 8207583 8209349 Example 0 +
or upload a file:
warning Please either provide the search regions (BED6{{ enhancerSelected ? '' : ' or gene names' }}) in the text box, or upload a text file (BED6{{ enhancerSelected ? '' : ' or gene names' }}).

STEP 3

Set the alignment parameters

help
What are these parameters?
The weight parameter w represents the relative weight of the epigenome to the sequence. EpiAlignment considers only sequence when w = 0, while it increasingly relies on epigenomic information as w rises. A w value greater than 0.15 is not recommended.
Alignment parameters
s: sequence substitution rate.
μ: sequence deletion rate.
κ: rate of switching between the two epigenomic states 0 and 1.
πA, πC, πG and πT: equilibrium probabilities of the four nucleotide bases.
π1: equilibrium probability of the epigenomic state 1.
Epigenome weight w (relative to sequence):
warning Please use non-negative numbers for the epigenome weight (preferably less than 0.15).
Alignment parameters:
s:
μ:
κ:
πA:
πC:
πG:
πT:
π1:
warning Please use positive numbers for all parameters.

STEP 4

Submit data

warning Please provide a valid email address or leave this field blank if you don't want to receive any notifications.
Due to the many processing steps and larger search regions involved, one-vs-one mode with larger regions runs longer than many-vs-many mode and may take minutes.
It is highly recommended to provide an email here to get notified when results become available. You may also bookmark the result page after submitting your data.

Select paired peak datasets from ENCODE Project and other public datasets

Here we provide some pre-compiled paired peak datasets from the ENCODE and mouse ENCODE project and other public datasets. More data will be added here after they are matched.
Please click on one of the blue cells to use the paired data between human and mouse in EpiAlignment.
Loading ENCODE data ...
ENCODE Datasets {{ filter.label }}
{{ sampleEntry.label }}
Public Datasets {{ filter.label }}
{{ sampleEntry.label }}