A high discrimination motif scanner (variable sensitivity)

This is a high discrimination scanner designed to recognise DNA motifs within a long stretch of DNA. Most importantly, this implementation shows how to implement a variable sensitivity for detection, by varying the pseudocount values. It uses two models for discrimination, one model representing the target and the second model representing the background. The model representing the target uses a set of DNA motifs, whereas the model that represents the background uses motifs colected from within the long stretch of DNA being analyzed. The scanner uses the motifs of the two models to compute two position weight matrices (PWMs). It uses the PWMs to search for regions that resemble the motifs from the set that are above the background. But how it does that? Consecutive score values extracted from sliding windows determine the formation of a representative signal above a long stretch of DNA. The score peaks from the signal determine the location of possible motif variations that resemble those in the motif set. Further step-by-step explanations regarding this scanner can be found in the book entitled: Algorithms in Bioinformatics: Theory and Implementation. The application also uses the Sequence Logo project shown here.

Live demo: https://gagniuc.github.io/A-high-discrimination-motif-scanner-and-variable-sensitivity/

References

• Paul A. Gagniuc. Algorithms in Bioinformatics: Theory and Implementation. John Wiley & Sons, Hoboken, NJ, USA, 2021, ISBN: 9781119697961.