Answers are grounded in this paper and cite their sources.
We examine scan statistics for one-dimensional marked Poisson processes. Such statistics tabulate the maximum weighted count of event occurrences within a window of predetermined width over all windows within an observed interval. We derive analytical formulas and also give an importance sampling method for approximating the tail probabilities of scan statistics. Because high-throughput genomic sequencing has led to the availability of massive amounts of biomolecular sequence data, it is often of interest to search long DNA or protein sequences for local regions that are enriched for a certain characteristic. Thus scan statistics have become a useful tool in modern computational biology. We illustrate the application of our p value approximations with such examples.
No takes yet. Share an insight, caveat, or question.
Chan et al. (2007) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: