jpHMM (jumping profile Hidden Markov Model) is a probabilistic approach to compare a sequence to a multiple alignment of a sequence family. The jpHMM web server at GOBICS is a tool for the detection of recombinations in HIV-1 and hepatitis B virus (HBV) genomes. For a query sequence phylogenetic recombination breakpoints are predicted and each region of the sequence is assigned to one HIV-1 subtype/HBV genotype. This prediction is based on a pre-calculated multiple alignment of the major HIV-1 subtypes/HBV genotypes. A detailed description of the algorithm and some information about the evaluation can be found here. For information about the output format please see the online submission page