The upload data file is similar to BED format. Each field is tab-delimited. The first two columns are chromsome and start position of a short tag and they are required when binning the data.
An example of input is: Chr1 10000 10025 [other columns]
the input data will be mapped to the whole human genome and be divided into equal size bins(minimum size is 25 bp). A cutoff value(minimum cutoff is 1) can be used to filter out those bins which values are below than the user input cutoff. The output are three columns delimited by TAB which are: [chromsome] [bin position] [count value]
The purpose of normalization is to identify systematic experiment errors and eliminate them to reveal the true biological signals. We provide following methods:
Against total read counts:This method is based on the assumption that the total number of DNA bindings would be similar in different cell types under different biological conditions. The number of fragments aligned is normalized using the total number of matched fragments in each sample. Hence, the normalized data will have equal number of tags across samples.