Detection of significant genomic alterations via simultaneous minimal sojourns at a state by independent continuous-time Markov chains (Q2516395)

The authors study the connection between a disease and recurrent alterations in genomic profiles of humans affected by the disease, each patient profile being modeled as a 2-state jump Markov process. They derive a closed-form upper bound of the probability for a given alteration to occur in a given number of patients. This upper bound leads to a cumulative probability of a waiting time associated with a specific birth and death process. An explicit expression is given for the Laplace transform of the waiting time related to the upper bound. The theoretical results are applied to a variety of situations encountered in cancer genomics.