ScoreBigwig - loosolab/TOBIAS GitHub Wiki

Background

The main task in footprinting is to identify regions of protein binding across the genome. Using single basepair cutsite tracks (as produced by ATACorrect), TOBIAS ScoreBigwig is used to calculate a continuous footprinting score across regions. The calculation of the TOBIAS footprint score is as follows:

This score takes into account the depletion of signal (negative counts), while also including the general accessibility of the nearby region. This helps to improve binding prediction for transcription factors with low footprintability (i.e. showing weak footprints). It is therefore important to note is that if there is no depletion, the score will converge to the accessibility of the flanking regions. A clear shape of a footprint is therefore not a necessity for a high footprint score to occur.

---

Example command

$ TOBIAS FootprintScores --signal test_data/Bcell_corrected.bw --regions test_data/merged_peaks.bed --output Bcell_footprints.bw --cores 8

~1 minute

---

Input parameters

• --signal
  Signal bigwig containing cutsites per basepair - e.g. {prefix}_corrected.bw from ATACorrect

• --regions
  Limits the computation of footprints to regions found in --regions .bed-file (in most cases the peaks of accessible chromatin).

• --score The score to apply to the input --signal. The default is "footprint", which calculates the footprint score depicted above. Other options include "sum" and "mean", which can be used to calculate other types of scores on the input --signal.

Full input parameters can be found by running TOBIAS ScoreBigwig --help.

Output

• Bigwig containing calculated scores. Positions with no score assigned can be interpreted as 0.