jPREdictor

jPREdictor is a software tool that enables the user to define motifs, search for them on sequences, weight motifs using training sets and score sequences using the weighted motifs. One particular type of short sequence fragments, namely transcription factor binding sites, are commonly found enriched and clustered in such elements. Therefore, these binding sites are commonly represented as motifs - a necessary requirement for prediction to take place.

With the use of motifs, it is possible to weight them by applying a positive or negative training set to search for them in sequences and eventually score a given sequence. The process of gene regulation is complex and involves many factors, one of which is cis-regulatory elements. Understanding these elements and their role in gene regulation is key to the development of organisms.

The prediction process in jPREdictor is based on clusters of individual motifs, and also allows for arbitrary combinations of these motifs into multi-motifs. The user can select maximal and minimal distances between the motifs in the multi-motif clusters. Individual motifs in jPREdictor can be of heterogeneous classes, such as simple sequence motifs or position-specific scoring matrices.

Finally, cluster scores in this software tool are obtained from the weighted occurrences of multi-motifs. This means that weights for each cluster are derived from positive and negative training sets, ensuring that predictions are reliable and accurate.